This feed does not validate.


  1. <?xml version="1.0" encoding="utf-8"?>
  2. <?xml-stylesheet type="text/css" href="" ?>
  4. <feed xmlns="" xml:lang="en">
  6. <title type="text">It's Not Magic</title>
  7. <subtitle type="text">
  8. Writings of a techie wizard
  9. </subtitle>
  10. <id></id>
  11. <link rel="alternate" type="text/html" href="" />
  12. <link rel="self" type="application/atom+xml" href="" />
  14. <author>
  15. <name>Peter A. Donis</name>
  16. <uri></uri>
  17. <email></email>
  18. </author>
  20. <rights>Copyright 2011-2015 by Peter A. Donis</rights>
  22. <generator uri="" version="0.9.8">
  23. simpleblog3 0.9.8
  24. </generator>
  26. <updated>2015-01-03T04:07:00Z</updated>
  28. <entry>
  29. <title type="html">Offline</title>
  30. <category term="/general" />
  31. <id></id>
  32. <updated>2015-01-03T04:07:00Z</updated>
  33. <published>2015-01-03T04:07:00Z</published>
  34. <link rel="alternate" type="text/html" href="" />
  35. <content type="xhtml">
  36. <div xmlns="">
  37. <p>As you might have noticed, I haven't posted here in a while, and what with
  38. various things going on, I don't expect to be posting again for a while.
  39. Everything that's here now will stay, but I won't be adding any new posts
  40. for an indefinite period. I hope you've enjoyed what I've posted here, and
  41. thanks for reading!</p>
  42. </div>
  43. </content>
  44. </entry>
  46. <entry>
  47. <title type="html">Science, Heal Thyself (Again)</title>
  48. <category term="/opinions" />
  49. <id></id>
  50. <updated>2014-09-27T01:57:00Z</updated>
  51. <published>2014-09-27T01:57:00Z</published>
  52. <link rel="alternate" type="text/html" href="" />
  53. <content type="xhtml">
  54. <div xmlns="">
  55. <p>Courtesy of
  56. <a href="">Watts Up With That</a>,
  57. I came across a
  58. <a href="">blog post</a>
  59. by Kate Marvel, a climate scientist who says she is "so bored with the
  60. hiatus". The WUWT article makes some good criticisms, though in fairness
  61. to Marvel, it appears to take her post's title a bit too literally--she
  62. isn't bored with the fact of the hiatus, but with all the media attention
  63. it gets, which is not quite the same thing. But here I want to focus on
  64. another aspect of Marvel's post: it's another good illustration of something
  65. I've
  66. <a href="">blogged about before</a>,
  67. namely, why the public finds it hard to trust what scientists say.</p>
  69. <p>First, I want to quote a particular paragraph from Marvel's post in full,
  70. so that when I start to parody its style, as I am about to do, it will be
  71. clear that I am not exaggerating:</p>
  73. <blockquote>
  74.  <p>Look, sometimes the ocean takes up more heat, and sometimes the atmosphere
  75.  does.  This is because the climate system is complex--so complex that people
  76.  literally do nothing all day but study how the air and water on Earth slosh
  77.  around and interact with each other. These pitiable people are called
  78.  scientists, and despite their questionable life choices they are really
  79.  pretty sharp.  While they no doubt appreciate being reminded of the hiatus
  80.  by you, WSJ writer/internet commenter/angry uncle, you may rest assured that
  81.  they are aware of it, perhaps even more so than you!  The question they are
  82.  interested in is not, "how come surface temperatures are rising so slowly?"
  83.  but rather, "why is the ocean doing so much of the work right now, and how
  84.  long will this last"?</p>
  85. </blockquote>
  87. <p>Wow, that sounds really cool! So what you're saying is, there are these
  88. really awesome people who are studying the climate, and they came up with
  89. this graph that shows how the ocean is absorbing a whole lot of heat. And,
  90. from what I can tell, the question they are asking is why the ocean is
  91. absorbing <em>more</em> heat now that it used to, right? That's what "doing so
  92. much of the work right now" means, yes?</p>
  94. <p>The problem is, you see, that if I look at this graph you gave, that's not
  95. what I get from it. First of all, it's hard to tell what it's saying, because
  96. you didn't actually give any <em>data</em> to back it up. So I can't check any real
  97. numbers; all I can do is eyeball this graph to try and pick out trends. For
  98. example, I can try to figure out how much heat the ocean absorbed from, say,
  99. 2000 to 2008 (which appears to be the end of the graph), and compare that
  100. with how much it absorbed in previous time periods of the same length.</p>
  102. <p>And when I do that, I come up with something like this: about 70 units of
  103. heat absorbed from 2000-2008 (the units are 10<sup>21</sup> J, according to
  104. the graph); about 20 units absorbed from 1992-2000; about 60 units from
  105. 1984-1992; about 50 units from 1976-1984; and about 30 units from 1968-1976.
  106. Now, if I were asking questions about this, the question I would be
  107. interested in is not "why is the ocean doing so much of the work right now?",
  108. because it doesn't look like it's doing significantly more work now than it
  109. did in the 1970s or 1980s. The question I would be interested in is "why did
  110. the ocean do so <em>little</em> work in the 1990s?", because that is the time period
  111. that seems to be so different from the ones before and after it. But from
  112. what you say, these "sharp" climate scientists, of which you yourself are
  113. one, are not even asking that question at all.</p>
  115. <p>(If I were really sharp, I might ask about the 1960s and early 1970s too,
  116. and I might even hypothesize that there was some sort of natural cycle in
  117. the way the oceans absorb heat, and that this might affect the climate. I
  118. might also ask, though, how reliable this ocean heat data is back that far
  119. in the first place, since we didn't even start trying to cover the ocean
  120. systematically with temperature measurements until the ARGO buoy project
  121. started in 2003, and that didn't give us reasonably complete coverage until
  122. about 2010. But maybe that's too much for a single post.)</p>
  124. <p>In other words, your graph does not tell me that the ocean is masking warming
  125. now by absorbing more heat than usual. It tells me that, during the 1990s,
  126. the ocean <em>caused</em> warming by absorbing <em>less</em> heat than usual. So the story
  127. you are trying to tell me with this graph is <em>not</em> the story that the graph
  128. itself tells me. That does not inspire my confidence, and it probably does
  129. not inspire the confidence of other members of the public either (to say
  130. nothing of media outlets like the <em>Wall Street Journal</em>).</p>
  132. <p>We do appear to agree on one thing, though:</p>
  134. <blockquote>
  135.  <p>my biggest problem with the hiatus is that it's really so tedious.</p>
  136. </blockquote>
  138. <p>The only difference is that what I find tedious is not the "hiatus" or the
  139. general reporting about it, but the fact that scientists like you trot out
  140. data and graphs and so forth and claim they say one thing, when they really
  141. say something else.</p>
  142. </div>
  143. </content>
  144. </entry>
  146. <entry>
  147. <title type="html">Don't Worry, It's Only Monopoly Money</title>
  148. <category term="/opinions" />
  149. <id></id>
  150. <updated>2014-09-15T03:38:00Z</updated>
  151. <published>2014-09-15T03:38:00Z</published>
  152. <link rel="alternate" type="text/html" href="" />
  153. <content type="xhtml">
  154. <div xmlns="">
  155. <p>Peter Thiel, in a recent
  156. <a href="">article</a>,
  157. says that (as the article's subhead puts it)</p>
  159. <blockquote>
  160.  <p>If you want to create and capture lasting value, look to build a monopoly</p>
  161. </blockquote>
  163. <p>Of course this works out well for the monopolist; but how about the rest
  164. of us?</p>
  166. <p>I'll go ahead and get the obvious criticism out of the way first: Thiel's
  167. actual examples don't justify the "create" part, only the "capture" part.
  168. Actually, "examples" is an overstatement, because he only gives one (at
  169. least, only one that directly addresses the point made in the subhead): he
  170. compares U.S. airline companies to Google. His numbers do show that Google,
  171. the monopoly, captures much more value than the highly competitive airlines
  172. do. But they also show that the airlines <em>create</em> much more value than
  173. Google does.</p>
  175. <p>I'll also go ahead and clear up the obvious objection that the word
  176. "monopoly" raises: Thiel does make it clear that he isn't talking about
  177. monopolies that only got that way because of special favors from the
  178. government or a willingness to bend the rules in a way that other
  179. companies are not. He is talking about</p>
  181. <blockquote>
  182.  <p>the kind of company that is so good at what it does that no other firm
  183.  can offer a close substitute.</p>
  184. </blockquote>
  186. <p>Of course, that definition makes a monopoly sound like a good thing, even
  187. though, as we saw above, they don't actually create more value. But that
  188. isn't really why Thiel thinks monopolies, in his sense, are good (which is
  189. why I got that criticism out of the way quickly, so we could focus on his
  190. real points--never mind that they aren't the same as the point the subhead
  191. makes). His real argument is twofold. First, he says, monopolies are
  192. better for workers:</p>
  194. <blockquote>
  195.  <p>Imagine you're running one of those restaurants in Mountain View. You're
  196.  not that different from dozens of your competitors, so you've got to fight
  197.  hard to survive. If you offer affordable food with low margins, you can
  198.  probably pay employees only minimum wage. And you'll need to squeeze out
  199.  every efficiency: That is why small restaurants put Grandma to work at the
  200.  register and make the kids wash dishes in the back.</p>
  202.  <p>A monopoly like Google is different. Since it doesn't have to worry about
  203.  competing with anyone, it has wider latitude to care about its workers, its
  204.  products and its impact on the wider world.</p>
  205. </blockquote>
  207. <p>This is wrong in at least two ways. First, the problems Thiel attributes
  208. to the restaurants in Mountain View are really problems of small businesses,
  209. not low-margin businesses. They can only afford to pay minimum wage
  210. (assuming they actually do; Thiel only says they "probably" do) because
  211. they have a small customer base, so wages, like everything else, have to
  212. come out of a small pool of resources. There are plenty of competitive
  213. businesses that can afford to pay high wages for people with the skills
  214. to justify them. Intel, for example, is certainly facing stiff competition,
  215. but they don't skimp when it comes to paying chip designers; they can
  216. afford to pay them well, because they have a much larger customer base
  217. and therefore a much larger pool of resources to draw on.</p>
  219. <p>Second, even if it's true in theory that Google, as a monopoly, has more
  220. leeway to care about things, it's not at all obvious that this actually
  221. translates to better outcomes in practice. Plenty of ex-Googlers have
  222. described various downsides of working there. And the "impact on the
  223. wider world" that Google is having is by no means an unmixed blessing.
  224. Along with the undeniable value of the search engine and of some apps
  225. like Google Maps, there is the ad-supported business model, the
  226. increasing amount of personal data on Google's servers, the repeated
  227. pattern of apps being launched, getting wide usage, and then being shut
  228. down because they aren't gaining Google enough revenue, and so on.</p>
  230. <p>Looking at the actual impact of Google, instead of the theoretical
  231. impact, based on the "don't be evil" motto, that appears to be all
  232. Thiel has bothered to consider, also brings up another problem with
  233. his analysis. Monopolies are supposed to be better because they capture
  234. more of the value they create: but Google actually captures <em>no</em> value
  235. directly from any of its applications. Anyone can use any Google app
  236. for free. All of the value Google captures is indirect, from measuring
  237. how many eyeballs land in the various places that it tracks. The major
  238. direct effect of this is on Google's users, not on its workers (and
  239. we'll talk about that further below); but the fact that users have no
  240. way of directly communicating to Google how much value they are getting
  241. from its services is a huge elephant in the room that Thiel never even
  242. mentions, even though capturing value is central to his analysis.</p>
  244. <p>So it's by no means clear that monopolies are better, either for workers
  245. or for users, if we go by the example of Google. But Thiel gives another
  246. argument for why monopolies are better for customers, despite the claims
  247. of economists to the contrary:</p>
  249. <blockquote>
  250.  <p>Profits come out of customers' wallets, and monopolies deserve their bad
  251.  reputation—but only in a world where nothing changes.</p>
  253.  <p>In a static world, a monopolist is just a rent collector. If you corner
  254.  the market for something, you can jack up the price; others will have no
  255.  choice but to buy from you. Think of the famous board game: Deeds are
  256.  shuffled around from player to player, but the board never changes. There
  257.  is no way to win by inventing a better kind of real-estate development.
  258.  The relative values of the properties are fixed for all time, so all you
  259.  can do is try to buy them up.</p>
  261.  <p>But the world we live in is dynamic: We can invent new and better things.
  262.  Creative monopolists give customers more choices by adding entirely new
  263.  categories of abundance to the world. Creative monopolies aren't just good
  264.  for the rest of society; they're powerful engines for making it better.</p>
  265. </blockquote>
  267. <p>On the face of it, this seems backwards. The usual view in economics is
  268. that competition is what forces companies to innovate. But Thiel claims
  269. that competition actually <em>prevents</em> companies from innovating; in fact,
  270. it prevents companies from doing anything beyond day-to-day operations:</p>
  272. <blockquote>
  273.  <p>Monopolists can afford to think about things other than making money;
  274.  non-monopolists can't. In perfect competition, a business is so focused on
  275.  today's margins that it can't possibly plan for a long-term future. Only
  276.  one thing can allow a business to transcend the daily brute struggle for
  277.  survival: monopoly profits.</p>
  278. </blockquote>
  280. <p>In part, the error here is the same as the one we saw above with the
  281. restaurants in Mountain View: having no resources to spare for anything
  282. beyond day-to-day operations is a problem of small businesses, not low
  283. margin businesses. It also doesn't help that Thiel fails to take into
  284. account that the profit margins companies report in their financials
  285. are what's left <em>after</em> all expenses are subtracted, and those expenses
  286. include things like spending to plan for a long-term future. (They also
  287. include employee wages and salaries, which is why a low-margin business
  288. is not forced to pay low wages, as we saw above, provided it has enough
  289. of a customer base to justify hiring highly skilled people.) Airlines,
  290. for example, spend plenty on things like new aircraft, and that spending
  291. is subtracted before they quote their profits.</p>
  293. <p>But the real problem here is that, while monopoly profits do provide an
  294. extra pool of resources for the company to spend as it wishes, they do
  295. not guarantee that this extra spending will actually translate into extra
  296. value. We saw an aspect of that above, when we looked at the effect of
  297. Google's business model on its users: the fact that Google captures no
  298. value directly from its users means that Google has no way of knowing the
  299. value of its various services to users. Of course, figuring out a way
  300. for users to be able to, for example, directly pay for Google search
  301. is a very hard problem. But solving it certainly seems like it would be
  302. of great value, and surely, if any company in the world is in a position
  303. to solve that kind of a problem, it's Google. So why, if monopolies
  304. really work the way Thiel claims they do, hasn't Google solved it?</p>
  306. <p>The standard economist's answer to this question is that Google has
  307. no incentive to solve the problem. The only way to create such an
  308. incentive would be for some competitor to put Google in a situation
  309. where the problem had to be solved in order to keep its user base.
  310. In the absence of such competition, Google is free to do, not what
  311. its users actually want (since it has no way of knowing that), but
  312. what it thinks its users might want, at least enough to increase its
  313. ad revenues. In other words, Google is using its monopoly profits,
  314. not to create wonderful new benefits for users, but to run expensive
  315. experiments on what users will click on; any new benefit for users is
  316. a lucky side effect (and is likely to go away once Google realizes
  317. that it isn't going to increase their revenues). Google is doing this
  318. simply because it can--because nothing is forcing it to do what seems
  319. obvious from the user's viewpoint and simply let users tell it
  320. directly how valuable its services are, by paying for them.</p>
  322. <p>Of course there's an obvious objection to this: if Google started
  323. charging for search, say, people would simply stop using it. But if
  324. that is really true, then that just sharpens the argument I made
  325. earlier: monopolies like Google may <em>capture</em> more value, but they
  326. <em>create</em> less. If Google search isn't worth paying for, let alone its
  327. other services, then Google is creating even less value than we thought.
  328. And that just makes Thiel's claims even harder to swallow.</p>
  330. <p>The other examples Thiel gives don't make his case any better. For
  331. example, the change from AT&amp;T's monopoly of phone service to the
  332. current state was a classic example of increasing competition benefiting
  333. users by reducing prices and increasing the availability of services; it
  334. was certainly <em>not</em> a case of a new and improved monopoly displacing the
  335. old one, which is Thiel's stated reason for mentioning it. One can
  336. perhaps justify viewing Microsoft and Apple as monopolies in Thiel's
  337. sense, depending on how narrowly you want to draw the line around what
  338. would count as a "close substitute". But, as with the case of Google,
  339. the benefits of Windows and iPhones have brought with them significant
  340. downsides for users: for example, the litany of security flaws in Windows,
  341. and a greatly increased cost of switching systems and applications. What
  342. has kept these problems from being worse than they are is competition:
  343. from Macs and Linux for Windows, and from Android for iOS.</p>
  345. <p>Thiel may be confused about competition because he is confused about
  346. how economists model it. He says:</p>
  348. <blockquote>
  349.  <p>Economists copied their mathematics from the work of 19th-century
  350.  physicists: They see individuals and businesses as interchangeable atoms,
  351.  not as unique creators. Their theories describe an equilibrium state of
  352.  perfect competition because that is what's easy to model, not because it
  353.  represents the best of business... But every new creation takes place far
  354.  from equilibrium.</p>
  355. </blockquote>
  357. <p>It's true that, in order for new creation to take place, the economy has
  358. to be out of equilibrium. But new creation is not what disturbs the
  359. equilibrium; the equilibrium is always being disturbed anyway, simply
  360. because people's needs and wants are always changing. Economists know
  361. perfectly well that the "equilibrium state of perfect competition" exists
  362. only in theory, not in reality, just as physicists know that no real
  363. system is ever in perfect thermodynamic equilibrium, because the system
  364. is always interacting with its environment, and the environment is always
  365. changing. But a physical system's <em>drive</em> towards equilibrium is what
  366. enables useful work to be done; and similarly, an economy's drive towards
  367. equilibrium--competition--is what enables useful creation to be done.
  368. Without that driving incentive, monopoly profits will simply get frittered
  369. away on things that may look good to the company, but don't actually
  370. benefit the rest of us.</p>
  371. </div>
  372. </content>
  373. </entry>
  375. <entry>
  376. <title type="html">How Not To Support Your Customers</title>
  377. <category term="/opinions" />
  378. <id></id>
  379. <updated>2014-07-18T03:12:00Z</updated>
  380. <published>2014-07-18T03:12:00Z</published>
  381. <link rel="alternate" type="text/html" href="" />
  382. <content type="xhtml">
  383. <div xmlns="">
  384. <p>The latest round of the Netflix-Verizon tiff that I
  385. <a href="">recently blogged about</a>
  386. has now appeared in a
  387. <a href="">post by Verizon</a>
  388. and a
  389. <a href="">response from Level 3</a>.
  390. First, Verizon purports to describe the problem and its solution:</p>
  392. <blockquote>
  393.  <p>Even though there is no congestion on our network, we're not satisfied
  394.  if our customers are not. We fully understand that many of our customers
  395.  want a great streaming experience with Netflix, and we want that too.
  396.  Therefore, we are working aggressively with Netflix to establish new,
  397.  direct connections from Netflix to Verizon's network.</p>
  398. </blockquote>
  400. <p>Which sounds good, but now look at Level 3's response explaining what
  401. would actually be needed to fix the problem:</p>
  403. <blockquote>
  404.  <p>[W]e could fix this congestion in about five minutes simply by
  405.  connecting up more 10Gbps ports on those routers. Simple. Something
  406.  we've been asking Verizon to do for many, many months, and something
  407.  other providers regularly do in similar circumstances. But Verizon
  408.  has refused. So Verizon, not Level 3 or Netflix, causes the congestion.
  409.  Why is that? Maybe they can’t afford a new port card because they've
  410.  run out - even though these cards are very cheap, just a few thousand
  411.  dollars for each 10 Gbps card which could support 5,000 streams or more.
  412.  If that's the case, we’ll buy one for them. Maybe they can't afford the
  413.  small piece of cable between our two ports. If that's the case, we'll
  414.  provide it. Heck, we'll even install it.</p>
  415. </blockquote>
  417. <p>In other words, Verizon wants Netflix to make a huge investment in a
  418. "direct connection" between the two networks, when all that's really
  419. needed is a few port cards and cables, the cost of which wouldn't even
  420. amount to rounding error in Verizon's accounting (and as you can see,
  421. they wouldn't even have to spend that since Level 3 has offered to cover
  422. all the costs).</p>
  424. <p>But that seems daft: Verizon customers are having a serious problem
  425. that has a simple fix, yet Verizon refuses to allow that fix. What
  426. could Verizon possibly be thinking? Here's Level 3's answer to that:</p>
  428. <blockquote>
  429.  <p>This congestion only takes place between Verizon and network providers
  430.  chosen by Netflix. The providers that Netflix does not use do not
  431.  experience the same problem. Why is that? Could it be that Verizon
  432.  does not want its customers to actually use the higher-speed services
  433.  it sells to them? Could it be that Verizon wants to extract a pound of
  434.  flesh from its competitors, using the monopoly it has over the only
  435.  connection to its end-users to raise its competitors' costs?</p>
  436. </blockquote>
  438. <p>If you're wondering how Netflix and Verizon are competitors, see
  439. <a href="">here</a>.</p>
  441. <p>It's worth noting that Verizon's talk about "direct connection" leaves
  442. me wondering exactly what the Netflix-Verizon deal I referred to in my
  443. <a href="">previous post</a>
  444. was supposed to accomplish, since the whole point of that deal was
  445. supposed to be giving Netflix a direct connection to Verizon's network,
  446. similar to the deal it made with Comcast. But if that were really the
  447. case, Level 3, which is a transit provider, would not even come into the
  448. picture. It's possible that, as
  449. <a href="">Ars Technica notes</a>,
  450. Verizon is simply taking time to implement the direct connections that
  451. their deal with Netflix makes possible, and until that implementation
  452. is complete, at least a part of Netflix traffic to Verizon customers
  453. goes via Level 3. But Verizon's post, quoted above, certainly seems to
  454. imply that "direct connection" is an <em>alternative</em> to what Netflix is
  455. doing now, not something Netflix has already paid Verizon for but Verizon
  456. has not finished implementing yet. Either way, this confusion certainly
  457. doesn't help Verizon's credibility.</p>
  459. <p>I'll leave you with this statement in Verizon's post, which is
  460. particularly ironic in view of all the above:</p>
  462. <blockquote>
  463.  <p>Verizon is focused on providing its customers with the best Internet
  464.  experience possible.</p>
  465. </blockquote>
  467. <p>As long as you don't try to experience Verizon's competitors, apparently.</p>
  468. </div>
  469. </content>
  470. </entry>
  472. <entry>
  473. <title type="html">Netflix Neutrality (Again)</title>
  474. <category term="/opinions" />
  475. <id></id>
  476. <updated>2014-06-06T03:26:00Z</updated>
  477. <published>2014-06-06T03:26:00Z</published>
  478. <link rel="alternate" type="text/html" href="" />
  479. <content type="xhtml">
  480. <div xmlns="">
  481. <p>In an entirely predictable development, at least if you've been keeping up
  482. with my
  483. <a href="">previous</a>
  484. <a href="">posts</a>
  485. on net neutrality, Netflix is now
  486. <a href="">having a tiff</a>
  487. with Verizon over slow delivery of Netflix content to customers.
  488. It seems that Netflix has been displaying messages to customers when videos
  489. take a long time to buffer, telling them that the reason is congestion on
  490. their ISP's network. Verizon, of course, didn't like that very much, so
  491. they sent Netflix a cease-and-desist letter telling them to stop blaming
  492. Verizon for slow video delivery.</p>
  494. <p>What's interesting about this is that Netflix has a similar deal with
  495. Verizon to the one it made with Comcast, which I referred to briefly in
  496. <a href="">my first net neutrality post</a>.
  497. The deal means that Netflix traffic does not have to go through a third
  498. party to get to Verizon's customers; Netflix has direct connections to
  499. Verizon's network (and to Comcast's), so the problem has to be either on
  500. Netflix's end or Verizon's end (update: further developments have shown
  501. that it's more complicated than that--see below). And in all the heavy
  502. weather Verizon is making about this, the one thing they are conspicuously
  503. <em>not</em> saying is that the problem is on Netflix's end.</p>
  505. <p>In other words, Verizon's customers are asking for Netflix data; the data
  506. is slow getting to the customer because Verizon's network is indeed slow
  507. (since if Netflix's end were slow, you can be sure Verizon would be saying
  508. so, loudly--update: it looks like they are saying loudly that the problem
  509. is not their network being slow, but that doesn't mean it's not their fault;
  510. see below); but Verizon <em>does not want its customers to know that</em>. As
  511. Netflix's spokesman says, quoted in the CNBC article,</p>
  513. <blockquote>
  514.  <p>This is about consumers not getting what they paid for from their
  515.  broadband provider. We are trying to provide more transparency, just like
  516.  we do with the ISP Speed Index, and Verizon is trying to shut down that
  517.  discussion.</p>
  518. </blockquote>
  520. <p>Of course, trying to shut down the discussion now is closing the door after
  521. the horse has left the barn. The only thing the cease and desist letter did
  522. was ensure that even people who are <em>not</em> Verizon customers, like me, now
  523. know that Verizon's network is slow (update: or at least that there is a
  524. significant problem that Verizon is not fixing as they should). The only way
  525. Verizon can really fix this problem is to, well, <em>fix</em> it, by upgrading its
  526. network (update: or fixing its connections with transit providers). But I'm
  527. not holding my breath.</p>
  529. <h1>Update (17 July 2014)</h1>
  531. <p>As I note in a
  532. <a href="">follow-up post</a>
  533. to this one, the fact that Verizon and Netflix have made the deal referred
  534. to above does not immediately take Internet transit providers (like Level 3,
  535. who handles Netflix traffic) out of the game. In fact, it's still not
  536. entirely clear, at least not from Verizon's public statements, exactly
  537. what the technical implications of the Netflix-Verizon deal are. See the
  538. follow-up post for more on that. However that may be, though, the bottom
  539. line is still the same: Verizon doesn't want its customers to know the
  540. real reason why their Netflix streaming is having problems, or what
  541. options for fixing it (some of which are quite simple, as I discuss in
  542. the follow-up post) have been refused by Verizon, for reasons which have
  543. nothing to do with serving the needs of their customers.</p>
  544. </div>
  545. </content>
  546. </entry>
  548. <entry>
  549. <title type="html">Net Neutrality Redux: Peer Pressure?</title>
  550. <category term="/opinions" />
  551. <id></id>
  552. <updated>2014-05-06T01:43:00Z</updated>
  553. <published>2014-05-06T01:43:00Z</published>
  554. <link rel="alternate" type="text/html" href="" />
  555. <content type="xhtml">
  556. <div xmlns="">
  557. <p>If you've read my
  558. <a href="">previous post</a>
  559. and are still wondering, even after the Postscript, whether I was really
  560. being fair, you may be interested in
  561. <a href="">this</a>
  562. from Level 3, another major Internet transit provider like Cogent, which
  563. I mentioned in my last post. It should come as no surprise that they are
  564. also having problems with major broadband providers.</p>
  566. <p>As the article notes, Level 3 has a total of 51 peers, i.e., other major
  567. Internet providers with whom they connect so that they can route traffic
  568. from their own customers to the portions of the Internet that they don't
  569. serve directly. Level 3 has no issues with most of those peers; there are
  570. only 12 with whom they observe significant congestion issues:</p>
  572. <blockquote>
  573.  <p>A port that is on average utilised at 90 percent will be saturated,
  574.  dropping packets, for several hours a day. We have congested ports saturated
  575.  to those levels with 12 of our 51 peers. Six of those 12 have a single
  576.  congested port, and we are both (Level 3 and our peer) in the process of
  577.  making upgrades – this is business as usual and happens occasionally as
  578.  traffic swings around the Internet as customers change providers.</p>
  580.  <p>That leaves the remaining six peers with congestion on almost all of the
  581.  interconnect ports between us. Congestion that is permanent, has been in
  582.  place for well over a year and where our peer refuses to augment capacity.
  583.  <strong>They are deliberately harming the service they deliver to their paying
  584.  customers. They are not allowing us to fulfil the requests their customers
  585.  make for content.</strong></p>
  587.  <p>Five of those congested peers are in the United States and one is in
  588.  Europe. There are none in any other part of the world. All six are large
  589.  Broadband consumer networks with a dominant or exclusive market share in
  590.  their local market. In countries or markets where consumers have multiple
  591.  Broadband choices (like the UK) there are no congested peers.</p>
  592. </blockquote>
  594. <p>Emphasis mine.</p>
  595. </div>
  596. </content>
  597. </entry>
  599. <entry>
  600. <title type="html">Why We Need Net Neutrality</title>
  601. <category term="/opinions" />
  602. <id></id>
  603. <updated>2014-04-30T03:14:00Z</updated>
  604. <published>2014-04-30T03:14:00Z</published>
  605. <link rel="alternate" type="text/html" href="" />
  606. <content type="xhtml">
  607. <div xmlns="">
  608. <p>In the wake of the
  609. <a href="">Federal Court ruling</a>
  610. in January that struck down key portions of the FCC's
  611. <a href="">Net Neutrality</a>
  612. regulations, it looks like the agency is now considering allowing ISPs
  613. to have a
  614. <a href="">"fast lane"</a>
  615. for preferred traffic, which means traffic that content providers are
  616. willing to pay the ISP extra for carrying. Needless to say, the content
  617. providers, such as Netflix, are
  618. <a href="">not in favor of this</a>.
  619. And also needless to say, ISPs like Comcast are hastening to assure us that
  620. <a href="">these aren't the droids we're looking for</a>.
  621. (Notice that the Netflix article is full of technical details, while the
  622. Comcast post is just corporate doublespeak--not to mention that the
  623. boilerplate disclaimers are more than twice the length of the actual post.)</p>
  625. <p>I'm not going to rehash all the arguments and counter-arguments here.
  626. Instead, I want to tell a little fable to illustrate why we, the ordinary
  627. users of the Internet, should be very concerned about any such "fast
  628. lane" regulation being put in place.</p>
  630. <hr>
  632. <p>You are at the intersection where you normally turn left to get to your
  633. favorite store. However, something seems to have happened to the road there.
  634. Instead of the usual smooth paved surface, it's all pockmarked with potholes
  635. and the surface in between the potholes is rough and gravelly. You're not
  636. even sure your car will tolerate being driven over that surface; certainly
  637. you'll have to take it a lot slower than you normally do. The store itself
  638. looks no different, nor does its parking lot (which is the private property
  639. of the store owners), and there's no obvious reason for the change in the
  640. road leading there.</p>
  642. <p>The road to the right, which leads to the SuperMegaStore that you never shop
  643. at, looks even nicer than it normally does. New lines have been painted, new
  644. street lights have been installed, and there's even a big sign at the turn
  645. now telling you about all the great bargains available at the SuperMegaStore
  646. if you just Turn Right Now. The SuperMegaStore itself, along with its huge
  647. parking lot, is the same as it's always been; once again, there's no obvious
  648. reason for the change in the road. But it seems like a <em>lot</em> more traffic is
  649. turning right now rather than left, which isn't surprising considering the
  650. conditions of the respective roads.</p>
  652. <p>A man happens to be standing to one side of the intersection, observing the
  653. traffic and making notes on a clipboard. You pull over to the shoulder of
  654. the road and walk up to him.</p>
  656. <p>"Hi," you say. "I don't want to interrupt, but I'm a bit curious about what's
  657. going on."</p>
  659. <p>The man makes a last note, then looks up.</p>
  661. <p>"You mean what's going on with the roads, no doubt?"</p>
  663. <p>"Yes," you say.</p>
  665. <p>"Well, it's not really that surprising. The Fast Lane Regulation went into
  666. effect at midnight last night."</p>
  668. <p>"The Fast Lane Regulation?" you ask. "I remember hearing about it on the news,
  669. but I never really understood what it was all about."</p>
  671. <p>"Well, it's really very simple," the man says. "As of midnight last night,
  672. the owners of roads can charge extra fees to the businesses the roads lead to,
  673. in order to maintain the roads' throughput. Otherwise, the owners wouldn't be
  674. able to recover the costs of building enough road capacity to serve the needs
  675. of the businesses. SuperMegaStore paid its fee, plus the extra surcharge for a
  676. road upgrade, and an advertising fee to have a larger sign put in, so that's
  677. what it got."</p>
  679. <p>"And the other store?" you ask, though you can already see what the answer is
  680. going to be.</p>
  682. <p>"They couldn't afford the fee," the man says, "so their road got downgraded."</p>
  684. <p>"But it was a perfectly good road before," you say. "I suppose I can see it
  685. not getting new lines painted or fancy street lights, but why make it worse
  686. than it was before?"</p>
  688. <p>"Well, obviously, if a store, or any other business, can get the same road
  689. quality without paying the fee, businesses won't want to pay the fee," the man
  690. says.</p>
  692. <p>"But I'm not sure I understand," you say. "The owners of these roads already
  693. get paid for building them, by the taxpayers, and even by tolls. I had to
  694. pay a toll on the main highway coming in here. And now they're charging the
  695. businesses the roads lead to as well? Isn't that getting paid twice for the
  696. same thing?"</p>
  698. <p>"I can't comment on that," the man says. "I'm sure the Road Commission took
  699. such things into consideration before it made the regulation."</p>
  701. <p>"But now I can't get to my store," you say. "Even if my car made it over that
  702. road this one time, it certainly won't be able to do it regularly. How am I
  703. going to get my shopping done?"</p>
  705. <p>"I can't comment on that either," the man says.</p>
  707. <p>As you drive around, you see the effects of the Fast Lane Regulation
  708. everywhere. Your favorite restaurant is now reachable only by a gravel track,
  709. while the chain restaurants whose food you can't stand have wide access roads
  710. and large billboards pointing them out. The MegaTheater has what almost seems
  711. to be a superhighway leading to it, while the smaller art theater that shows
  712. the old movies you like is now on the other side of a dilapidated one-lane
  713. bridge. It seems like you can't get anywhere you would like to go.</p>
  715. <p>Then you begin to wonder: what about <em>new</em> businesses? It seems like they
  716. would have to be able to pay the fees just to come into existence; otherwise
  717. starting them would be pointless, since it would be so difficult to reach
  718. them. It used to be fairly common in your town to see a new business starting
  719. up; but that was when everyone could count on customers having access to any
  720. place they chose to go on equal terms. What will it be like now?</p>
  722. <p>You keep thinking that this just doesn't make sense. Roads are supposed to
  723. be common infrastructure for everybody; they're not supposed to privilege
  724. some businesses over others. How could something like this Fast Lane
  725. Regulation even happen? Businesses already compete based on the quality
  726. and price of what they provide; competing on the basis of extra fees to
  727. allow customers to reach them doesn't seem fair. There must be some way to
  728. fix it.</p>
  730. <hr>
  732. <p>So what is the moral of this little tale? Well, each Federal Court that has
  733. ruled on this subject has given the FCC a very broad hint: if it wants to
  734. regulate ISPs as
  735. <a href="">common carriers</a>,
  736. it can regulate them however it wants. This is exactly what was done with
  737. the telephone companies in the 1930's, and it is why the sort of scenario
  738. described in the above fable never materialized for telephones. I've been
  739. racking my brain trying to come up with a reason, other than the obvious
  740. one--corporate influence--why the FCC would not be taking this obvious
  741. course, but I've been unable to do it. But regardless of the reason, the
  742. course the FCC is considering now would be very, very bad for the Internet
  743. and for us, its users.</p>
  745. <h1>Postscript</h1>
  747. <p>By the way, if you're wondering whether the above fable was really being
  748. fair by downgrading the roads that didn't pay the fee, consider
  749. <a href="">this graph</a>
  750. showing the performance of Netflix by ISP.</p>
  752. <p>Also, just to add some more fuel to the fire, consider
  753. <a href="">this article</a>
  754. talking about the wider implications of the Comcast-Netflix deal. Basically,
  755. Comcast, Verizon, and the other major internet providers are no longer just
  756. ISPs; they also now own a considerable portion of the Internet's backbone,
  757. which is where competition between providers used to do the most good. In
  758. other words, the major ISPs are working hard to create a situation where
  759. they control <em>every</em> pathway from the rest of the Internet to you. As the
  760. article notes, this will make it harder for <em>any</em> regulation by the FCC to
  761. be implemented fairly, though I don't agree with the article's title that
  762. it makes such regulation obsolete. We still need common carrier regulation
  763. in this environment; we just need to also push back against the way the
  764. Internet's structure is evolving away from a decentralized peer-to-peer
  765. network and towards a system of monolithic walled gardens. But one battle
  766. at a time.</p>
  768. <h1>Post-Postscript</h1>
  770. <p>This just in, a
  771. <a href="">Netflix-Verizon deal</a>
  772. similar to the Netflix-Comcast deal has been made. If you're wondering why
  773. Comcast and Verizon just happened to be the first two major ISPs to strike
  774. this deal with Netflix, it may help to know that, prior to these deals, the
  775. same transit provider,
  776. <a href="">Cogent</a>,
  777. was serving Netflix content to both of them, and that, prior to these deals,
  778. the ISPs were making
  779. <a href="">quite a bit of noise</a>
  780. about the fact that, because of the volume of Netflix traffic, Cogent was
  781. sending them a lot more data than they were sending Cogent, which apparently
  782. was not kosher (at least not to the ISPs) under the existing peering
  783. agreements between the ISPs and the transit providers, which assumed that
  784. the traffic between them would be roughly balanced, and allowed the peering
  785. to be free (i.e., no money changing hands either way) on that basis.</p>
  787. <p>Of course, an obvious course for Verizon or Comcast to take if peering was
  788. becoming unbalanced with Cogent would be to stop giving Cogent peering for
  789. free, and start charging them for the excess traffic generated by Netflix.
  790. So one way of looking at the situation is that, as
  791. <a href="">this article</a>
  792. suggests, it may actually be cheaper for Netflix to pay Comcast and Verizon
  793. directly and cut out the middleman, if the alternative is for the middleman
  794. to no longer get peering with the ISPs for free. As the article notes, even
  795. if the issues with Cogent were resolved now, Netflix might want to change the
  796. transit provider it uses at some point, which it apparently has done fairly
  797. often in the past, and the new provider would then have the same issues.</p>
  799. <p>What's missing from all of this, though, is any acknowledgment of the people
  800. who are actually the ultimate source of all this traffic: the <em>customers</em> of
  801. Comcast and Verizon (and other ISPs) who want to watch streaming movies over
  802. broadband. The ISPs are talking as though it's all Netflix' fault for
  803. generating so much traffic, without even mentioning that it's <em>their own
  804. customers</em> who are actually creating the traffic. Sure, they happen to be
  805. doing so by watching Netflix right now, but it could just as well be Hulu
  806. next week, or some new service next year that doesn't even exist now. If
  807. <em>all</em> of these services just happen to have problems connecting with a few
  808. particular ISPs, is that the service's fault, or the ISP's fault? Isn't there
  809. a saying that "the one common factor in all of your failed relationships is
  810. you"?</p>
  811. </div>
  812. </content>
  813. </entry>
  815. <entry>
  816. <title type="html">Does Bernie Sanders Read This Blog?</title>
  817. <category term="/opinions" />
  818. <id></id>
  819. <updated>2014-04-28T21:18:00Z</updated>
  820. <published>2014-04-28T21:18:00Z</published>
  821. <link rel="alternate" type="text/html" href="" />
  822. <content type="xhtml">
  823. <div xmlns="">
  824. <p>Some time back I made a
  825. <a href="">proposal for campaign finance reform</a>.
  826. Now I find that Senator Bernie Sanders has proposed a
  827. <a href="">constitutional amendment</a>
  828. that is identical to my proposal. I don't know if Sanders reads this blog,
  829. but however he got the idea, I'm for it.</p>
  830. </div>
  831. </content>
  832. </entry>
  834. <entry>
  835. <title type="html">News Flash: IPCC Says Burning Food Is A Bad Idea</title>
  836. <category term="/rants" />
  837. <id></id>
  838. <updated>2014-03-27T02:32:00Z</updated>
  839. <published>2014-03-27T02:32:00Z</published>
  840. <link rel="alternate" type="text/html" href="" />
  841. <content type="xhtml">
  842. <div xmlns="">
  843. <p>The Daily Telegraph
  844. <a href="">reports</a>
  845. that, based on the latest draft of the IPCC AR5,</p>
  847. <blockquote>
  848.  <p>The United Nations will officially warn that growing crops to make "green" biofuel harms the environment and drives up food prices</p>
  849. </blockquote>
  851. <p>(hat tip:
  852. <a href="">Watts Up With That</a>
  853. ). At first glance, this looks promising, an actual outbreak of sanity for
  854. the IPCC, something like
  855. <a href="">admitting that climate model forecasts are inaccurate</a>.
  856. But just as with that previous item, you shouldn't get your hopes up too
  857. much; as you can see even from the brief quote above, the obvious reason
  858. for not using food crops to make biofuels (the one that's in the title of
  859. this post) is <em>not</em> the primary reason the IPCC gives for their about-face
  860. on this issue.</p>
  862. <p>The primary reason the IPCC gives is that</p>
  864. <blockquote>
  865.  <p>growing biofuel crops on a large scale requires either the conversion of
  866.  agricultural land used for food crops or the destruction of forests to free
  867.  up land, possibly offsetting any reduction in carbon emissions from the use
  868.  of biofuels.</p>
  869. </blockquote>
  871. <p>In other words, the IPCC isn't really concerned about rising food prices;
  872. after all, if they had been, the AR4, back in 2007, would not have made such
  873. an aggressive recommendation to <em>increase</em> the use of biofuels. It's not as
  874. though burning food only just started to drive up food prices. No, it's all
  875. about CO2 alarmism.</p>
  877. <p>Of course, even if we restrict the discussion to the climate aspect, the
  878. IPCC is admitting that they screwed up. Did they just now discover that
  879. growing biofuel crops requires the use of land? Couldn't exactly the same
  880. analysis have been done back in 2007? Why wasn't it? Of course, nobody is
  881. asking those questions. And if the IPCC can screw up something this basic,
  882. what does that say about their ability to get it right on more complex
  883. issues, like, oh, say, predicting what Earth's climate will be like in
  884. fifty or a hundred years? Of course, nobody is asking those questions
  885. either.</p>
  887. <p>But to me, all that is secondary to the real issue, which is that the IPCC,
  888. and all the governments that make policy based on what the IPCC says, were
  889. willing to make it more difficult for a significant fraction of the world's
  890. population to get enough to eat, right now, based on the belief that it would
  891. lead to some vague benefit to the climate fifty or a hundred years hence.
  892. Which they're now saying isn't going to be a benefit anyway (not that I
  893. believed them whey they said it would, but the point is that now even they
  894. admit it's not). And the people who wouldn't get enough to eat had no say
  895. in the matter. These are the people whom we are supposed to trust with the
  896. future of our planet. Personally, I don't trust them to add two and two
  897. correctly. But maybe that's just me.</p>
  898. </div>
  899. </content>
  900. </entry>
  902. <entry>
  903. <title type="html">Constitution Worship?</title>
  904. <category term="/rants" />
  905. <id></id>
  906. <updated>2014-03-21T03:58:00Z</updated>
  907. <published>2014-03-21T03:58:00Z</published>
  908. <link rel="alternate" type="text/html" href="" />
  909. <content type="xhtml">
  910. <div xmlns="">
  911. <p>Some time back I
  912. <a href="">noted</a>
  913. that what was then a common sentiment (I found it in an op-ed in the New
  914. York Times, which is proof of it being a common sentiment if anything is)
  915. about the Constitution seemed backwards to me. The claim was that we were
  916. getting into trouble about the "fiscal cliff" because we were too obsessed
  917. with following the Constitution; but as I showed in that post, the real
  918. problem was that we weren't following it <em>enough</em>.</p>
  920. <p>Now I've come across a
  921. <a href="
  922. worship.html">lecture</a>
  923. given by Michael Karman at Johns Hopkins University on Constitution Day,
  924. 2010, entitled "A Skeptical View of Constitution Worship", which goes even
  925. further than the NYT op-ed did. My basic response is the same: the problem
  926. is not that we "worship" the Constitution, it's that we ignore it.
  927. But the lecture presents such a tempting target that I can't help going
  928. beyond that; so here goes.</p>
  930. <p>I'll start with a key point that the lecturer doesn't appear to be aware of
  931. (or if he is, he's done a swell job of concealing it): we can <em>amend</em> the
  932. Constitution. For example, the lecturer bemoans the fact that the original
  933. Constitution allowed slavery, and even gave it legal protections:</p>
  935. <blockquote>
  936.  <p>[I]t's hard to celebrate a Constitution that explicitly guaranteed the
  937.  return of fugitive slaves to their masters, protected the international
  938.  slave trade for 20 years, and enhanced the South’s national political
  939.  representation to reflect its slaveholding.</p>
  940. </blockquote>
  942. <p>Well, we fixed that problem with the Thirteenth, Fourteenth, and Fifteenth
  943. Amendments. The lecturer also complains about the original Constitution
  944. being anti-democracy:</p>
  946. <blockquote>
  947.  <p>The Framers were trying to create a powerful national government that
  948.  was as distant from popular control as possible: very long terms in office,
  949.  large constituencies, indirect elections. They thought of democracy as rule
  950.  by the mob. They didn't think poor people could be trusted with the suffrage.
  951.  They didn't think women should vote.</p>
  952. </blockquote>
  954. <p>We amended the Constitution to fix those things too: popular election of
  955. Senators, giving women the right to vote, prohibiting poll tax to remove a
  956. major roadblock to voting. The lecturer might feel, of course, that these
  957. reforms didn't go far enough: perhaps we need even more protection for
  958. voting rights; perhaps we should pass the Equal Rights Amendment to ensure
  959. that women's rights are respected. But why aren't the reforms we <em>have</em> made,
  960. by amending the Constitution, mentioned at all? Do they somehow not count?</p>
  962. <p>Similarly, the lecturer complains about features of the Constitution that
  963. "still bind us", without appearing to be aware that we can amend those things
  964. too. (It's true that one thing he complains about, having two Senators for
  965. every State regardless of population, would be much harder to change, since
  966. there is an explicit provision about that.) But more than that, the lecturer
  967. appears to assume without question that whatever he thinks is a good idea,
  968. must in fact <em>be</em> a good idea, so if the Constitution makes it harder for us
  969. to do it, the Constitution must be bad.</p>
  971. <p>For example, he complains that no foreign-born person can be President,
  972. and that the Electoral College is a bad idea. Well, guess what? If you think
  973. those things should be changed, <em>propose an Amendment to change them</em>. And
  974. then we can actually have a substantive debate about whether these changes
  975. would, in fact, be good for the country. Don't blame the Constitution for
  976. the fact that it makes you go through that laborious process instead of just
  977. dictating to everybody that whatever you think is a good idea is what we're
  978. going to do. There's a reason the Constitution is set up to make these kinds
  979. of changes hard: to protect us from ourselves. And judging by our performance
  980. when we disregard these protections, the Framers were quite right to try to
  981. put roadblocks in our way to prevent us from charging ahead to make changes
  982. whenever we feel like it.</p>
  984. <p>In other words, the lecturer doesn't understand what the Constitution is
  985. really about. It's not about finding the "right" set of provisions and
  986. enforcing them on everyone. It's about finding a structure that lets people
  987. with very different ideas about what is "right" coexist peacefully in the
  988. same country, going about their business and not interfering with each other,
  989. by agreeing on a common set of basic guidelines that everyone can live with,
  990. and <em>stopping there</em>.</p>
  992. <p>Which brings us to the lecturer's third point: we ignore the Constitution
  993. anyway. This is often true, as I argued myself in my previous post on this
  994. topic. However, unlike the lecturer, I view that as a bug, not a feature.
  995. And it doesn't make much sense to argue that following various Constitutional
  996. provisions is bad for the country, as the lecturer has just done in points 1
  997. and 2, when you're also arguing that we don't follow the Constitution.</p>
  999. <p>Which also brings us to the fourth point: the Supreme Court ignores the
  1000. Constitution. I certainly agree with that; I've
  1001. <a href="">said</a>
  1002. <a href="">the</a>
  1003. <a href="">same</a>
  1004. <a href="">thing</a>
  1005. myself. But once again, is this supposed to be a feature, or a bug? Is the
  1006. right response to just admit that we ignore the Constitution, and discard it?
  1007. Or is the right response to start actually taking it seriously? Which means
  1008. that if we really have a problem with a Constitutional provision, because
  1009. our values have changed from those of the Framers, we <em>amend</em> it, like the
  1010. Framers explicitly <em>told</em> us to do?</p>
  1012. <p>Would it be worth it to do all that work? Well, that brings us back to point 0:
  1013. how many of our freedoms, which the writer justifiably admires (I do too), do we
  1014. owe to the Constitution? Let's see:</p>
  1016. <blockquote>
  1017.  <p>It's a wonderful thing that one can criticize the president--even call him a
  1018.  socialist or a coddler of terrorists, if you like--and not worry about being
  1019.  arrested for it.</p>
  1020. </blockquote>
  1022. <p>In other words, freedom of speech. First Amendment: check.</p>
  1024. <blockquote>
  1025.  <p>It's a great thing that one can pursue one's own religious beliefs with a
  1026.  great deal of tolerance...</p>
  1027. </blockquote>
  1029. <p>First Amendment again: check.</p>
  1031. <blockquote>
  1032.  <p>...and that a black man can be president of a country that held blacks in
  1033.  slavery just 150 years ago and that still had an entrenched system of white
  1034.  supremacy until roughly 50 years ago.</p>
  1035. </blockquote>
  1037. <p>We already covered that one above: check. (And a case can be made that the
  1038. reason it took so long to overcome Jim Crow after the Thirteenth, Fourteenth,
  1039. and Fifteenth Amendments were passed was that we ignore the Constitution when
  1040. we feel like it. If we had really taken those Amendments seriously, we might
  1041. have had a black President sooner.)</p>
  1043. <blockquote>
  1044.  <p>It's a great thing that in America a woman came very close to being elected
  1045.  president of the United States two years ago and that one probably will win
  1046.  such an election sometime fairly soon.</p>
  1047. </blockquote>
  1049. <p>We covered giving women the right to vote above too: check. (And I'm all in
  1050. favor of a woman President as long as it's <em>not</em> Hillary Clinton.)</p>
  1052. <blockquote>
  1053.  <p>I personally think it's a wonderful thing that in many states gay and
  1054.  lesbian couples can get married just like straight couples, and that, I would
  1055.  predict, they will be able to do so in most of the country within another
  1056.  decade or so.</p>
  1057. </blockquote>
  1059. <p>Ok, you've got me on this one: as Justice Scalia is fond of pointing out, the
  1060. Constitution says nothing about marriage.</p>
  1062. <p>Oh, wait: what's that in the Fourteenth Amendment? "Equal protection of the
  1063. laws"? I take it back: the Constitution has this one covered too. As I've
  1064. blogged
  1065. <a href="">before</a>.</p>
  1067. <p>The funny part is that the lecturer gets things <em>almost</em> right at the end:</p>
  1069. <blockquote>
  1070.  <p>In the end, we, the American people, determine what sort of country we live
  1071.  in--</p>
  1072. </blockquote>
  1074. <p>Right on! But then he muffs it:</p>
  1076. <blockquote>
  1077.  <p>--the Constitution and the courts play a relatively marginal role in that
  1078.  process.</p>
  1079. </blockquote>
  1081. <p>And as we've seen, when that does happen, the process does not work. But then
  1082. it gets even better: we have a statement that is a truth <em>and</em> a deeply flawed
  1083. misunderstanding at the same time:</p>
  1085. <blockquote>
  1086.  <p>To paraphrase the great jurist with the greatest of names--the Honorable
  1087.  Learned Hand--no constitution and no court are going to rescue us from white
  1088.  supremacy or sexism or homophobia or Japanese American internment or FBI
  1089.  profiling of Arabs and Muslims.</p>
  1090. </blockquote>
  1092. <p>This is true as a matter of history: as we've seen, the Constitution and the
  1093. Supreme Court did <em>not</em> rescue us from many bad things. But that's <em>our</em>
  1094. fault. We chose to ignore the Constitution, and chose not to hold our elected
  1095. representatives and the Supreme Court accountable when <em>they</em> ignored the
  1096. Constitution, and indeed, we got all these bad things. What would have
  1097. happened if we had <em>upheld</em> the Constitution, and voted people out of office
  1098. when they ignored it, and protested when the Court failed to uphold it?
  1099. Unfortunately we can't go back and find out what would have happened in the
  1100. past; but we could at least give it a try for the future.</p>
  1101. </div>
  1102. </content>
  1103. </entry>
  1105. <entry>
  1106. <title type="html">What If They Gave A Crisis And Nobody Came?</title>
  1107. <category term="/rants" />
  1108. <id></id>
  1109. <updated>2014-01-17T04:55:00Z</updated>
  1110. <published>2014-01-17T04:55:00Z</published>
  1111. <link rel="alternate" type="text/html" href="" />
  1112. <content type="xhtml">
  1113. <div xmlns="">
  1114. <p>It's been obvious for quite some time, at least to anyone not marinated
  1115. in the ideology of climate change alarmism, that the models being used to
  1116. produce the IPCC's forecasts of doom
  1117. <a href="">do not match reality</a>.
  1118. But now it's become so glaring that even the IPCC itself has admitted it
  1119. in the
  1120. <a href="">Summary for Policymakers</a>
  1121. (SPM) from Working Group I for its Fifth Assessment Report (AR5)
  1122. (hat tip:
  1123. <a href="">Watts Up With That</a>
  1124. ).</p>
  1126. <p>Actually, of course, "admitted it" is optimistic phrasing: a more apt
  1127. description would be "attempted to pretend nothing is actually wrong".
  1128. You have to look carefully to see the admissions; for example, as noted
  1129. in the Watts Up With That post, one is in a footnote on p. 14 of the SPM
  1130. (and in a font small enough that I had to zoom in to read it on my
  1131. computer). Also, there is no discussion that I can find of a key point that
  1132. is often overlooked when comparing the climate models to reality: the
  1133. different model projections are based on different assumptions about how
  1134. much CO2 will be emitted in the future, and actual CO2 emissions thus far
  1135. have been similar to the most pessimistic set of models (i.e., the ones
  1136. that assumed the most CO2 emissions), which have overpredicted actual
  1137. temperatures significantly <em>more</em> than the average of all the models, which
  1138. is what is usually quoted when comparing the models to actual observations.
  1139. So the IPCC's admissions still don't fess up to the full extent of the
  1140. problem: the model predictions are even worse than they admit.</p>
  1142. <p>(It's also worth noting that, despite admitting, however obliquely, that
  1143. its models cannot predict future climate, the IPCC continues to fill its
  1144. report with predictions of future climate. Perhaps it's force of habit.)</p>
  1146. <p>The IPCC report is not the only venue in which climate change alarmism is on
  1147. the defensive. The
  1148. <a href="">Warsaw Climate Change Conference</a>
  1149. ended in November, and despite the positive spin on the conference website,
  1150. the general sense was that it
  1151. <a href="">fell short</a>
  1152. even of somewhat limited expectations. Part of the reason for that may have
  1153. been that alarmists' efforts to link any sort of adverse event to climate
  1154. change have been facing increasing skepticism. The fact that the conference's
  1155. educational materials
  1156. <a href="">lied about sea level rise</a>,
  1157. claiming that the sea level began rising in the late 1800's "for the first
  1158. time since the last ice age", didn't help either. (The materials also claimed
  1159. that Northern Hemisphere snow cover is decreasing, which it isn't.) And back
  1160. in October, the U.S. Supreme Court
  1161. <a href="">agreed to hear a case</a>
  1162. challenging the EPA's regulation of CO2 emissions under the Clean Air Act.</p>
  1164. <p>You'd think that a bunch of people who claim to be honest scientists would
  1165. feel some contrition over all this. Instead, they continue to dial up the spin.
  1166. One meme which has become popular is the "Hiroshima bomb" comparison, which
  1167. I referred to in
  1168. <a href="">a recent post</a>.
  1169. The meme has now even appeared as an
  1170. <a href="">app</a>
  1171. that counts the "Hiroshima bombs" of heat being added to the climate. Of
  1172. course, as I showed in that recent post, these numbers don't actually amount
  1173. to much at all when put in perspective (and the link above gives more numbers
  1174. showing the same thing). Another common meme is "denialists are harassing
  1175. us"; a good recent example is
  1176. <a href="">this piece in The Guardian</a>
  1177. about the case currently before the Virginia Supreme Court regarding a
  1178. Freedom of Information Act request for emails from Michael Mann and other
  1179. climate scientists. The Guardian's position on this is interesting:</p>
  1181. <blockquote>
  1182.  <p>Freedom of Information (Foia) laws...were enacted at the federal level
  1183.  and also in many states to help insure transparency and accountability in
  1184.  government. They have proved invaluable tools for journalists and public
  1185.  interest organizations seeking to uncover information that some in government
  1186.  would prefer to hide. But applying these so called "sunshine laws" to
  1187.  academics at state-run academic institutions is something new.</p>
  1188. </blockquote>
  1190. <p>In other words, governments shouldn't be allowed to hide information, but
  1191. academics doing science under government grants (which are Federal grants,
  1192. by the way, despite that bit about "state-run" institutions), science which
  1193. is claimed to justify public policies with huge costs? Sure, hide all the
  1194. information you want, no sweat. The author does helpfully explain why:
  1195. forcing academics to openly share information would</p>
  1197. <blockquote>
  1198.  <p>have a chilling effect on the free and open sharing amongst colleagues
  1199.  which is essential in the scientific process.</p>
  1200. </blockquote>
  1202. <p>So "free and open sharing" apparently means "not giving information to people
  1203. who disagree with you". Well, it's nice to have that clarified.</p>
  1205. <p>But then the piece goes on to make a point that does make some sense. It
  1206. quotes Michael Halpern of the Union of Concerned Scientists:</p>
  1208. <blockquote>
  1209.  <p>"Freedom of information laws rightly exempt internal communications and
  1210.  deliberations in order to facilitate the free exchange of ideas," Halpern says.</p>
  1211. </blockquote>
  1213. <p>Now as a pure matter of principle, I actually agree with this--<em>if</em> it is
  1214. limited to "internal communications and deliberations" (a distinction that
  1215. is, of course, notably lacking in the article up to this point--not to
  1216. mention the fact that it's notably lacking when the media is pestering the
  1217. government for information, but that's another post). I'm really not
  1218. interested in reading Michael Mann's private emails. I don't care what
  1219. discussions he has behind closed doors or what sort of groupthink goes on
  1220. in his research group--as long as it <em>stays</em> there.</p>
  1222. <p>But what I <em>do</em> care about, as I've
  1223. <a href="">said before</a>,
  1224. is scientists like Michael Mann doing bad science, then declaring a planetary
  1225. emergency based on it, and then obstructing at every turn any attempt to get
  1226. the details on which the science is based, to demonstrate that it's bad
  1227. science. Mann hasn't just withheld private emails; he's done his best to
  1228. withhold raw data, statistical methods, and anything else that could be
  1229. used to check his work. And when that information finally came out, in
  1230. spite of his best efforts to the contrary, it showed that his science was,
  1231. in fact, bad science. Yet instead of owning up, or at least <em>shutting</em> up,
  1232. he continues to peddle climate change alarmism.</p>
  1234. <p>As the recent efforts at spinning the IPCC AR5 show, this behavior is
  1235. typical of climate change alarmists: the more their conclusions are
  1236. discredited, the louder they shout that hey, there really is a planetary
  1237. emergency--really! Cross my heart and hope to die! Why is this? Of course
  1238. I telegraphed my answer in the title of this post. These are people of
  1239. <a href="">Heinlein's class one</a>,
  1240. who are afraid of losing their cushy position as the ones who get to tell
  1241. others what to do and pronounce moral judgment on everyday activities like
  1242. driving your car. If they give a crisis and nobody comes, they might have
  1243. to find honest work.</p>
  1244. </div>
  1245. </content>
  1246. </entry>
  1248. <entry>
  1249. <title type="html">The Non-Beatings Will Continue Until Morale Improves</title>
  1250. <category term="/rants" />
  1251. <id></id>
  1252. <updated>2013-11-28T04:31:00Z</updated>
  1253. <published>2013-11-28T04:31:00Z</published>
  1254. <link rel="alternate" type="text/html" href="" />
  1255. <content type="xhtml">
  1256. <div xmlns="">
  1257. <p>A few weeks ago the Federal Reserve announced that it would
  1258. <a href="">continue "quantitative easing"</a>
  1259. at its current level. The reason, as explained in the press release just
  1260. linked to (though in rather oblique language, as is the usual practice with
  1261. such things), was basically that, while the economy appears to be recovering,
  1262. the Fed isn't sure that it's recovering strongly enough. Which leads to the
  1263. obvious next question: how much longer will this have to go on?</p>
  1265. <p>To see just how acute this question really is, you have to bear in mind that
  1266. the reason the Fed is doing "quantitative easing" (QE) in the first place is
  1267. that it has already maxed out its other tools. The Fed's target interest rate
  1268. is already as low as it can go, i.e., zero, and the press release makes it
  1269. clear that there are no plans to change that any time soon (in fact, the
  1270. release notes that the rate will probably remain at effectively zero for some
  1271. time <em>after</em> "quantitative easing" ends, whenever that is). Bank
  1272. <a href="">reserve requirements</a>
  1273. are about as low as they can possibly go without admitting openly that they
  1274. are basically zero (and note that the upper limit of the "low-reserve tranche"
  1275. in which only 3 percent of liabilities must be held in reserve, has been rising
  1276. steadily, as shown on the table further down the page, and is scheduled to rise
  1277. again in January 2014). Yet the economy continues to be sluggish.</p>
  1279. <p>Of course, there is no shortage of theories as to why the economy has not
  1280. responded more emphatically to all this nice treatment. The standard Keynesian
  1281. answer, which you can find even at
  1282. <a href="">The Economist</a>
  1283. these days, is that without QE the recession would have been much worse. On
  1284. this theory, the Fed should certainly not be even <em>thinking</em> about scaling
  1285. back QE (as they almost did in September); if anything, they should be
  1286. thinking about <em>expanding</em> it. The main problem with this theory is that, if
  1287. the Keynesians actually believed it, they <em>would</em> be advocating expanding QE,
  1288. yet none of them are. True, they aren't advocating reducing the pace either;
  1289. <a href="">this column</a>
  1290. by Paul Krugman is a good example of the current Keynesian wisdom:</p>
  1292. <blockquote>
  1293.  <p>On the whole, I'm sympathetic to skepticism about the effectiveness of QE,
  1294.  predictably. After all, I’ve been arguing for forward guidance instead for
  1295.  15 years. On the other hand, right now investors are not making a clear
  1296.  distinction between QE and forward guidance; taper talk has been accompanied
  1297.  by a clear shift in expectations toward the notion that the Fed will raise
  1298.  short-term rates sooner rather than later. So I wouldn’t be tapering now--it
  1299.  sends a bad signal at a time when recovery remains very weak and fragile.</p>
  1300. </blockquote>
  1302. <p>That bit about "forward guidance" just means that, instead of QE, Krugman
  1303. would prefer that the Fed just cross-its-heart-and-hope-to-die-double-promise
  1304. that it really, really, really won't raise interest rates, and hope that by
  1305. itself is enough to get the economy to recover. You will note, of course, as
  1306. I did above, that the Fed basically did exactly that in the press release; in
  1307. fact they have been doing it pretty consistently for the past few years. So
  1308. if it were going to work, you would expect that it already would have worked,
  1309. with or without the added push of QE. Another beautiful theory spoiled by an
  1310. ugly fact.</p>
  1312. <p>Let's try a different theory. Consider: what is the Fed actually <em>doing</em> when
  1313. it does QE? According to the press release, it is buying "additional agency
  1314. mortgage-backed securities...and longer-term Treasury securities". In plain
  1315. English, the Fed is buying various securities from banks in order to drive
  1316. up their price and thereby drive down the interest rates on them. The hope
  1317. is that the lower interest rates will encourage people and businesses to
  1318. take out more loans and thereby start spending again.</p>
  1320. <p>But for that to happen, the banks, who are the direct recipients of the $2.8
  1321. trillion and counting from the Fed, have to <em>make</em> the loans. What if they
  1322. just choose to hold on to the cash instead? According to the Fed's
  1323. <a href="">accounting of bank reserves</a>,
  1324. that's exactly what they have been doing. As you can see from the link, at
  1325. the end of October, 2013, total bank reserve balances were about $2.4
  1326. trillion. Does that number sound familiar? And what's more, only $73 billion
  1327. of that is required to meet the banks' reserve requirements. According to the
  1328. Fed, they can lend out <em>all</em> the rest, i.e., about $2.3 trillion. But they
  1329. haven't. Why not?</p>
  1331. <p>Before answering this question, we should pause to observe how outlandish this
  1332. situation is, at least on the surface. As
  1333. <a href="">CNBC</a>
  1334. points out, banks can earn anywhere from 4 percent to 10 percent or more on
  1335. various types of loans; yet instead, they are leaving their money deposited at
  1336. the Fed earning 0.25 percent. What's going on here?</p>
  1338. <p>CNBC's answer is simple:</p>
  1340. <blockquote>
  1341.  <p>[B]anks now apparently consider that their risk-adjusted return on consumer
  1342.  loans are lower than the 0.25 percent deposit rate at the Fed...</p>
  1343. </blockquote>
  1345. <p>In other words, on this theory, banks are actually acting rationally: they
  1346. would make <em>less</em> than 0.25 percent, on net, if they actually loaned out that
  1347. $2.3 trillion. So the situation is actually even <em>more</em> outlandish than it
  1348. first appears; what kind of messed up economy do you have to have for banks
  1349. to think that their best available rate of return is 0.25 percent?</p>
  1351. <p>I've never seen a Keynesian economist even <em>ask</em> this kind of question, but
  1352. there are other kinds of economists, though you rarely see them as talking
  1353. heads on the news or blogging in venues like the New York Times. For example,
  1354. you could try
  1355. <a href="">this article</a>
  1356. from the Ludwig von Mises Institute:</p>
  1358. <blockquote>
  1359.  <p>[V]arious studies that supposedly show that the Fed's quantitative easing
  1360.  can grow the US economy are fallacious. To suggest that monetary pumping can
  1361.  grow an economy implies that increases in the money supply will result in
  1362.  increases in the pool of real wealth.</p>
  1364.  <p>This is however a fallacy since all that money does is serve as the medium
  1365.  of exchange. It enables the exchange of the produce of one specialist for
  1366.  the produce of another specialist and nothing more. If printing money could
  1367.  somehow generate wealth then world wide poverty would have been eliminated by
  1368.  now.</p>
  1370.  <p>On the contrary, monetary pumping sets in motion a process of economic
  1371.  impoverishment by activating an exchange of something for nothing. It diverts
  1372.  real wealth from wealth generating activities towards non-productive
  1373.  activities.</p>
  1374. </blockquote>
  1376. <p>First, a side note: many non-Austrian economists would object to the term
  1377. "monetary pumping" being used to describe QE. For example,
  1378. <a href="">this article</a>
  1379. on the "Pragmatic Capitalism" site claims that QE does not actually create
  1380. any new money. All it does is transfer money from one asset to another: the
  1381. Fed gains an asset such as a mortgage-backed security and the seller gains
  1382. a reserve balance at the Fed.</p>
  1384. <p>However, this analysis ignores the fact that a bank can make loans based on
  1385. its reserve balance at the Fed (as long as the balance doesn't go below the
  1386. "reserve limit", which as we saw above, is basically negligible right now),
  1387. whereas it cannot make loans based on its holdings in mortgage-backed
  1388. securities. And everyone agrees that banks making loans based on fractional
  1389. reserves <em>does</em> create new money; so the fact that QE itself is just an
  1390. "asset transfer" is a red herring: banks' ability to make loans <em>is</em> being
  1391. boosted, and that's what counts.</p>
  1393. <p>Having got that out of the way, let's look at the real point: what if banks
  1394. aren't making loans because, quite simply, there really is so little actual
  1395. productive activity going on that there are no useful loans they can make?
  1396. Of course it isn't quite right to say that there is <em>no</em> productive activity
  1397. going on. Obviously there are a lot of people still doing productive work,
  1398. because life is still going on: people still need food, clothing, and shelter,
  1399. they still need cars to go places, they still want new smartphones and wide
  1400. screen TVs and other gadgetry. But there is more than enough of all that stuff
  1401. already being produced; nobody needs a bank loan to make more of it.</p>
  1403. <p>What is <em>not</em> happening is the creation of <em>new</em> forms of wealth. Bright,
  1404. ambitious young people no longer want to be scientists or engineers or
  1405. explorers; they want to be investment bankers and hedge fund managers. But
  1406. these activities don't generate any wealth; all they do is transfer wealth
  1407. from one pocket to another. Of course investment bankers and hedge fund
  1408. managers will vehemently object to this, but there's an easy way to test
  1409. it. Just ask the question: when you trade a stock or a bond or some other
  1410. security, do you expect to make money? Of course the answer is yes, otherwise
  1411. you wouldn't be making the trade. But if that's the case, you must also
  1412. expect the other party to the trade to <em>lose</em> money.</p>
  1414. <p>It's worth taking a bit to unpack this. Normal transactions, involving money
  1415. on one side and something tangible on the other--a good or a service--are
  1416. positive sum: <em>both</em> parties are better off after the trade. When you buy
  1417. a DVD player, the player is worth more to you than the money you pay for it,
  1418. because you can't use money to play DVDs. But to the store, the money is
  1419. worth more than the player, because they don't need it to play DVDs; they
  1420. only have it in the first place in order to sell it. In other words, the
  1421. good or service that is traded has a <em>different</em> value to you than it does
  1422. to the store.</p>
  1424. <p>But when you trade stocks or bonds, that isn't the case, because the stock
  1425. or bond is no use for anything by itself; its only use is as a financial
  1426. instrument, entitling you to some series of cash flows in the future. Those
  1427. future cash flows will be the same regardless of who owns the stock or bond,
  1428. so it <em>must</em> be the case that, whenever the stock or bond is traded, one
  1429. party to the trade is worse off. If the stock or bond is going to do well,
  1430. the seller is worse off: the cash they receive is worth less than the net
  1431. present value of the future cash flows. If the stock or bond is going to do
  1432. poorly, the buyer is worse off: the net present value of the future cash
  1433. flows is worth less than the cash they paid. One or the other <em>must</em> be
  1434. true, as a matter of simple math, which means that, as above, one party to
  1435. the trade <em>must</em> lose money.</p>
  1437. <p>Now consider what must happen for bank loans to be profitable. The bank
  1438. gives the borrower present money in exchange for future money: a series of
  1439. future cash flows. The bank will not make the loan unless the net present
  1440. value of that series of future cash flows is greater than the amount of
  1441. present money they lend. But that means the borrower, in order to be able
  1442. to pay back the loan at all, <em>must</em> do something productive with it, i.e.,
  1443. something that is positive sum, something that creates enough new wealth
  1444. to be able to pay back the loan from the proceeds and still come out ahead.
  1445. Of course, it's quite possible for <em>some</em> people to take loaned money and
  1446. start investment banks or hedge funds (of course, they call the lenders
  1447. "clients" instead), and transfer enough wealth to themselves to pay back
  1448. the money. But it's impossible for <em>everybody</em> to do that; and if enough
  1449. people start trying to do it instead of productive activity, loans will
  1450. shut down, no matter how much QE you pump into the banks.</p>
  1452. <p>On this Austrian view, the solution is simple: stop QE, and in fact stop
  1453. <em>all</em> of the Fed's interventions into the economy. All they are doing is
  1454. masking the true state of the economy, and therefore preventing people
  1455. from adjusting to reality. True, the adjustment will be painful, but it
  1456. would have been less painful if we'd done it sooner.</p>
  1458. <p>So whose view is right? The Austrian view has at least this much going
  1459. for it: it offers an explanation for what is, on the mainstream Keynesian
  1460. view, the great mystery of why the economy continues to stagnate despite
  1461. all of the TLC lavished on it by the Fed. However, it looks like the TLC
  1462. is going to continue, though it seems to me to be a rather bitter twist
  1463. on the standard line about beatings and morale.</p>
  1464. </div>
  1465. </content>
  1466. </entry>
  1468. <entry>
  1469. <title type="html">Science, Heal Thyself</title>
  1470. <category term="/opinions" />
  1471. <id></id>
  1472. <updated>2013-10-26T01:38:00Z</updated>
  1473. <published>2013-10-26T01:38:00Z</published>
  1474. <link rel="alternate" type="text/html" href="" />
  1475. <content type="xhtml">
  1476. <div xmlns="">
  1477. <p>A while back, I advised climate change alarmists to
  1478. <a href="">get off the soapbox</a>.
  1479. Now it appears that I have to extend that advice to scientists more
  1480. generally.
  1481. In the course of wandering around the Intertubes, I came across
  1482. <a href="">this op-ed</a>
  1483. by Adam Frank, which appeared in the New York Times a couple of months ago.
  1484. (Hat tip:
  1485. <a href="">this post</a>
  1486. by Anthony Watts, which linked to Popular Science magazine's post explaining
  1487. why they were
  1488. <a href="">shutting down comments</a>,
  1489. which linked to the NYT op-ed.)
  1490. Frank laments the fact that the public doesn't have the confidence it
  1491. used to have in science:</p>
  1493. <blockquote>
  1494.  <p>The triumph of Western science led most of my professors to believe that
  1495.  progress was inevitable. While the bargain between science and political
  1496.  culture was at times challenged -- the nuclear power debate of the 1970s, for
  1497.  example -- the battles were fought using scientific evidence. Manufacturing
  1498.  doubt remained firmly off-limits.</p>
  1500.  <p>Today, however, it is politically effective, and socially acceptable, to
  1501.  deny scientific fact. Narrowly defined, "creationism" was a minor current in
  1502.  American thinking for much of the 20th century. But in the years since I was
  1503.  a student, a well-funded effort has skillfully rebranded that ideology as
  1504.  "creation science" and pushed it into classrooms across the country. Though
  1505.  transparently unscientific, denying evolution has become a litmus test for
  1506.  some conservative politicians, even at the highest levels.</p>
  1507. </blockquote>
  1509. <p>So far, so good; I've
  1510. <a href="">blogged before</a>
  1511. about the same problem. But then comes this:</p>
  1513. <blockquote>
  1514.  <p>Meanwhile, climate deniers, taking pages from the creationists' PR
  1515.  playbook, have manufactured doubt about fundamental issues in climate
  1516.  science that were decided scientifically decades ago.</p>
  1517. </blockquote>
  1519. <p>Sorry, Professor Frank, but you just illustrated <em>why</em> the tactic of denying
  1520. "scientific fact" has become politically effective and socially acceptable:
  1521. scientists themselves have misrepresented what is "scientific fact", using
  1522. that term not just in reference to fields like evolution that have massive
  1523. supporting data and a comprehensive theory to back them up, but to fields
  1524. like climate science that simply are not in the same category, but which
  1525. happen to fit the scientist's personal ideology.</p>
  1527. <p>For example: what, exactly, are these "fundamental issues in climate science
  1528. that were decided scientifically decades ago"? The fact that CO2 absorbs
  1529. infrared radiation? No reputable scientist denies this, and even climate
  1530. scientists who do not support the so-called "consensus" around climate change
  1531. alarmism, such as Richard Lindzen of MIT, will tell you that media hacks who
  1532. claim that CO2 is not a greenhouse gas are just that, media hacks.</p>
  1534. <p>But by juxtaposing climate science with evolution, Professor Frank is inviting us
  1535. to believe that climate change alarmism itself is based on "fundamental issues
  1536. in climate science that were decided scientifically decades ago": that the
  1537. climate science in, say, the IPCC's
  1538. <a href="">WG 1 Report in the AR5 draft</a>
  1539. has the same level of confidence and support behind it as the theory
  1540. of evolution. And that is, how can I put this delicately, <em>wrong</em>. I don't
  1541. want to make this a book-length post (though I will probably have more to
  1542. come on this subject in the near future), and the work of showing how the
  1543. so-called "consensus" trumpeted by the IPCC is, to put it bluntly, bogus,
  1544. has already been done, most recently by the
  1545. <a href="">NIPCC</a>,
  1546. a group of scientists who want to make clear that they are <em>not</em> part of
  1547. the "consensus" and are willing to do the grunt work of refuting it point
  1548. by point. But I'll take a brief detour to give just one illustration of what
  1549. I'm talking about.</p>
  1551. <p>A recent post on
  1552. <a href="">RealClimate</a>
  1553. about ocean heating contained this little gem:</p>
  1555. <blockquote>
  1556.  <p>The increase in the amount of heat in the oceans amounts to 17 x 10<sup>22</sup>
  1557.  Joules over the last 30 years.  That is so much energy it is equivalent to
  1558.  exploding a Hiroshima bomb every second in the ocean for thirty years.</p>
  1559. </blockquote>
  1561. <p>The hysteria factor alone should raise red flags with scientists in other
  1562. fields like Professor Frank; but let's put that aside and do some simple math to
  1563. see what these numbers really mean. (We're also putting aside, by the way,
  1564. any questions about whether the numbers quoted are accurate, which, since
  1565. we have only had reasonably comprehensive ocean coverage since 2003, is not
  1566. a trivial point; but that's another post.) This amount of heat is for the
  1567. upper 2000 meters of the world's oceans. How much water is that? The surface
  1568. area of Earth's oceans is 360 million square kilometers according to
  1569. <a href="">The Physics Factbook</a>.
  1570. That makes a total volume of water of 720 million cubic kilometers, if we
  1571. assume the entire ocean is at least 2000 meters deep. But of course it
  1572. isn't that deep everywhere, so we have to cut that number down some. Let's
  1573. say, just for a quick calculation, that the actual amount of water is 3/4
  1574. that, or 480 million cubic kilometers. (This is probably a substantial
  1575. underestimate, since the average depth of the oceans is more than 2000
  1576. meters; but we want to be conservative since we're just doing a rough
  1577. calculation to put the numbers in perspective.)</p>
  1579. <p>But that's cubic <em>kilometers</em>; each cubic kilometer is a billion cubic
  1580. meters, so we're talking about 480 million billion (4.8 x 10<sup>17</sup>)
  1581. cubic meters. Water weighs about 1000 kilograms per cubic meter (seawater is
  1582. actually somewhat heavier, but we're being conservative in our numbers),
  1583. so that's 4.8 x 10<sup>20</sup> kilograms of water.</p>
  1585. <p>How much will 17 x 10<sup>22</sup> Joules raise the temperature of
  1586. 4.8 x 10<sup>20</sup> kilograms of water? Water has a specific heat of 4180
  1587. Joules per kilogram per degree Celsius, so it takes 4.8 x 10<sup>20</sup>
  1588. x 4180 = 2.0 x 10<sup>24</sup> Joules to heat up the top 2000 meters of the
  1589. ocean by 1 degree C. That means the temperature rise over the last 30 years
  1590. is 17 x 10<sup>22</sup> divided by 2.0 x 10<sup>24</sup>, which comes out
  1591. to: <em>0.085 degrees Celsius</em>.</p>
  1593. <p>You may be fidgeting in your seat about now, thinking that I have pulled
  1594. a fast one. Surely the important quantity is heat, not temperature,
  1595. right? And that same 17 x 10<sup>22</sup> Joules can raise the temperature
  1596. of the <em>atmosphere</em> a lot more than the temperature of the oceans, right?</p>
  1598. <p>These statements are not false, but they are also not telling the whole
  1599. story. The first, obvious point left out is that direct heat transfer
  1600. only occurs if there is a <em>temperature</em> difference. So the heat will
  1601. <em>stay</em> in the ocean unless the ocean is <em>warmer</em> than the air above it;
  1602. and if the ocean's temperature has only changed by 0.085 degrees C,
  1603. it can't raise the temperature of the atmosphere by direct heat transfer
  1604. more than that.</p>
  1606. <p>The second point arises from a question you might have after reading the
  1607. last paragraph: what about evaporation? Even a small difference in ocean
  1608. temperature will increase the evaporation rate; and evaporation transfers
  1609. heat from the ocean to the atmosphere, right? Yes, that's right: but what
  1610. <em>that</em> statement leaves out is <em>how</em> the heat gets transferred. Evaporation
  1611. is part of the hydrologic cycle: heat gets carried by water vapor from the
  1612. surface to high altitudes, where the water vapor condenses to form clouds
  1613. or precipitation. When it condenses, the latent heat it was carrying gets
  1614. deposited in the atmosphere; but because that is happening at altitude,
  1615. it's easier for that heat to escape to space.</p>
  1617. <p>In other words, of the two possible ways that 17 x 10<sup>22</sup> Joules
  1618. of heat could get transferred from the oceans to the atmosphere, one will
  1619. not make much difference (because the ocean temperature has only risen by
  1620. 0.085 C), and the other makes it easier for that heat to escape back to
  1621. space, which provides a negative feedback. Either way, to just state the
  1622. amount of heat in the oceans, without talking about how it could possibly
  1623. get transferred to the atmosphere and what those mechanisms entail, is not
  1624. a fair presentation of this issue. (And the Hiroshima bomb thing only makes
  1625. it worse.)</p>
  1627. <p>If you read through the NIPCC's report that I linked to above and compare
  1628. it with the IPCC WG1 report (or if you have been following this issue for
  1629. any significant amount of time), you will see that every specific issue
  1630. you dig into suffers from the same disease: the actual science doesn't say
  1631. what the IPCC "summary" says it says, and it certainly doesn't say what the
  1632. hysterical rhetoric in the popular media says it says. Scientific theories
  1633. that deserve the confidence Mr. Frank wants us all to give to science are
  1634. not like that. There may be polemics on both sides; indeed, there are
  1635. scientists, for example Richard Dawkins, who are just as vituperative,
  1636. if you just look at the surface rhetoric, when defending the theory of
  1637. evolution as climate change alarmists are at defending their so-called
  1638. "consensus". But when you look at the actual substance behind the rhetoric,
  1639. with theories like evolution, you find that the main claims that the
  1640. rhetoric makes are justified. All living organisms on Earth <em>are</em>
  1641. descended from a common ancestor. Natural selection <em>does</em> cause changes
  1642. in the gene pools of populations of organisms.</p>
  1644. <p>Also, scientists in the field of evolution are quite willing to say just
  1645. where the limits of knowledge are; evolutionary biologists will readily
  1646. admit that many of our current beliefs about how specific species or
  1647. specific structures or traits evolved are tentative, and may well turn out
  1648. to be wrong when we learn more. And when biologists recommend policies to
  1649. the public that have huge consequences, they can back those recommendations
  1650. up, and they can demonstrate specific consequences that will happen if the
  1651. recommendations are not followed. For example, biologists predicted that
  1652. overuse of antibiotics would lead to the evolution of antibiotic-resistant
  1653. bacteria, and sure enough, it did.</p>
  1655. <p>In short, a well-supported scientific theory like evolution looks very
  1656. different, when you take the time to check into it, than climate science
  1657. does. Yet Professor Frank likens climate science to the theory of evolution; so
  1658. either he hasn't bothered to check into it, or he has decided that the
  1659. difference doesn't matter. Either way, he has demonstrated why ordinary
  1660. people don't trust science the way they used to: how can they, when
  1661. scientists themselves are either sloppy or disingenuous when talking to
  1662. the public?</p>
  1664. <p>Please note, by the way, that I said "scientists" just now, not "climate
  1665. scientists". Adam Frank is a physicist and astronomer; his scientific
  1666. work is far removed from climate science. I'm not saying that every
  1667. scientist in every field has to take the time to check up on every other
  1668. field; we all have plenty of demands on our time, I understand that. But
  1669. a scientist <em>does</em> have the responsibility, when talking to the public,
  1670. to not misrepresent science, in <em>any</em> field, not just his own. If he hasn't
  1671. checked up, he should say so, and should not present what he says about
  1672. the other field as fact. Being a scientist doesn't relieve you of the need
  1673. to have an informed opinion if you're going to have an opinion at all;
  1674. indeed, scientists are supposed to be <em>better</em> than the average person at
  1675. recognizing that need and taking action appropriately.</p>
  1677. <p>It's disappointing to see Frank, and many scientists like him, not doing
  1678. that. It's even more disappointing when you see that, when it suits him,
  1679. Frank is perfectly willing to draw the distinction he does <em>not</em> draw in
  1680. his op-ed. For example,
  1681. <a href="">here he is on life after death</a>:</p>
  1683. <blockquote>
  1684.  <p>For myself I remain fully and firmly agnostic on the question. If ever
  1685.  there was a place where firm convictions seem misplaced this is it.
  1686.  There simply is no controlled, experimental verifiable information to
  1687.  support either the "you rot" vs. "you go on" positions.</p>
  1689.  <p>In the absence of said information we are all free to believe as we
  1690.  like but, I would argue, it behooves us to remember that truly "public"
  1691.  knowledge on the subject - the kind science exemplifies - remains in
  1692.  short supply.</p>
  1693. </blockquote>
  1695. <p>Just to be clear: I am <em>not</em> saying that science can never give us
  1696. reliable knowledge, the kind of knowledge that <em>does</em> justify branding
  1697. those who refuse to accept it as "deniers". We have scientific theories,
  1698. such as relativity and quantum mechanics, that have been experimentally
  1699. verified to extremely high accuracy. We have others, such as the theory
  1700. of evolution, that, while their subject matter prevents them from being
  1701. verified experimentally to the same degree, still have a mountain of
  1702. evidence in their favor, with more coming in every day, and a
  1703. comprehensive theoretical structure that explains the evidence and
  1704. makes correct predictions. But we also have plenty of areas of science
  1705. where we do <em>not</em> have that same level of understanding: and the real
  1706. disservice to science is to fail to be honest with the public about
  1707. which is which. <em>That</em> is what needs to be fixed if scientists like
  1708. Frank want "science denial" to stop.</p>
  1709. </div>
  1710. </content>
  1711. </entry>
  1713. <entry>
  1714. <title type="html">Some Things Never Change</title>
  1715. <category term="/opinions" />
  1716. <id></id>
  1717. <updated>2013-09-13T15:47:00Z</updated>
  1718. <published>2013-09-13T15:47:00Z</published>
  1719. <link rel="alternate" type="text/html" href="" />
  1720. <content type="xhtml">
  1721. <div xmlns="">
  1722. <p>This
  1723. <a href="">news</a>
  1724. is several years old now, but I just came across the article today and
  1725. I can't resist a brief comment.</p>
  1727. <p>The good news: videos of Richard Feynman giving his famous lectures on
  1728. physics at Caltech in 1964 are available online, thanks to Bill Gates.</p>
  1730. <p>The bad news: if you think this means that a wonderful resource for
  1731. learning about science is now open and accessible to everyone, think again.
  1732. From the article:</p>
  1734. <blockquote>
  1735.  <p>Note you will need to download Microsoft's Silverlight to get around
  1736.  the site.</p>
  1737. </blockquote>
  1739. <p>You can take the boy out of Microsoft, but you can't take Microsoft out of
  1740. the boy.</p>
  1741. </div>
  1742. </content>
  1743. </entry>
  1745. <entry>
  1746. <title type="html">There Oughta Be A Law</title>
  1747. <category term="/opinions" />
  1748. <id></id>
  1749. <updated>2013-09-01T23:04:00Z</updated>
  1750. <published>2013-09-01T23:04:00Z</published>
  1751. <link rel="alternate" type="text/html" href="" />
  1752. <content type="xhtml">
  1753. <div xmlns="">
  1754. <p>A recent
  1755. <a href="">article</a>
  1756. (via
  1757. <a href="">Reuters</a>,
  1758. via
  1759. <a href="">Hacker News</a>)
  1760. says that the US Congress should spend more time working and less time
  1761. vacationing. I could go on and on about what Congress actually does when
  1762. it <em>is</em> working, but that would go in the rants section of this blog. Here
  1763. I just want to comment on one particular thing that struck me about the
  1764. article.</p>
  1766. <p>Here's the article's main point:</p>
  1768. <blockquote>
  1769.  <p>We have to let our representatives know that business as usual isn't
  1770.  acceptable and we expect them, above all else, to get stuff done. The
  1771.  number of laws passed by Congress last year was fewer than at any point
  1772.  since 1947.</p>
  1773. </blockquote>
  1775. <p>An accompanying graph shows laws passed by year from 1947 to 2012. (I
  1776. could digress by asking why they picked 1947, but that would be another
  1777. article.)</p>
  1779. <p>When I read this, I at once thought of a remark made by
  1780. <a href="">Edsger Dijkstra</a>
  1781. about measuring the effectiveness of programmers:</p>
  1783. <blockquote>
  1784.  <p>[I]f we wish to count lines of code, we should not regard them as "lines
  1785.  produced" but as "lines spent": the current conventional wisdom is so foolish
  1786.  as to book that count on the wrong side of the ledger.</p>
  1787. </blockquote>
  1789. <p>It seems to me that much the same thing applies to Congress and laws; we
  1790. should not count laws as "laws produced" but as "laws spent". If it takes
  1791. more and more laws to give us good government (leave aside, once again,
  1792. the question of how good it actually is), that means Congress is doing a
  1793. <em>worse</em> job, not a better job. A really competent Congress would figure out
  1794. how to accomplish the same goals for government with <em>fewer</em> laws, not more.</p>
  1795. </div>
  1796. </content>
  1797. </entry>
  1799. <entry>
  1800. <title type="html">"Your" Cloud Data Is Not Yours, Take 2</title>
  1801. <category term="/opinions" />
  1802. <id></id>
  1803. <updated>2013-08-12T23:56:00Z</updated>
  1804. <published>2013-08-12T23:56:00Z</published>
  1805. <link rel="alternate" type="text/html" href="" />
  1806. <content type="xhtml">
  1807. <div xmlns="">
  1808. <p>(Note: there is a discussion of this post on
  1809. <a href="">Hacker News</a>.)</p>
  1811. <p>I
  1812. <a href="">posted</a>
  1813. some time back that one drawback of the "cloud" is that you can't
  1814. control how data you post to a "cloud" service is used. Facebook has
  1815. now provided us with an even better example than the case (Instagram)
  1816. I talked about in that post.</p>
  1818. <p>According to
  1819. <a href="">groovyPost</a>
  1820. (via
  1821. <a href="">Hacker News</a>),
  1822. Facebook uses data in your contact list to create "shadow" accounts for
  1823. people who aren't even on Facebook. It isn't clear exactly how Facebook
  1824. uses the data in these "shadow" accounts, but their previous behavior
  1825. does not inspire confidence. There is no way for the user to turn off or
  1826. control this behavior; it's not even visible to you as a Facebook user.
  1827. In fact, based on a quote from a Facebook representative given in the
  1828. article, Facebook apparently believes that allowing users to control
  1829. this behavior would violate Facebook's freedom of speech!</p>
  1831. <p>(I should make clear that if you only sign in to Facebook through the
  1832. Facebook website, as far as I can tell, it doesn't access any of your
  1833. contacts. But if you sign in to any other site and connect that
  1834. account with your Facebook account--for example, if you use the Gmail
  1835. feature that lets you automatically log in to Facebook using your
  1836. Gmail account--then any contacts you have on the other site will get
  1837. harvested by Facebook. Or, if you use Facebook's smartphone app, all
  1838. of your contacts stored on the phone will get harvested.)</p>
  1840. <p>I don't want to draw out this post with a discussion of whether
  1841. corporations even <em>have</em> freedom of speech the way individuals do
  1842. (that's a whole other can of worms). My point here is simply that it's
  1843. one thing to decide that you don't mind your own personal information
  1844. being spread all over the Internet. I've said
  1845. <a href="">before</a>
  1846. that I personally don't choose to do that, but as long as it's just
  1847. your own information, it's your choice. But Facebook has decided to
  1848. put you, if you're a Facebook user, in a position where you might
  1849. be compromising <em>other</em> people's personal information, even if they
  1850. would much rather you didn't, and without giving you any choice in
  1851. the matter. If that isn't something you want to do, you should think
  1852. carefully about how you use Facebook.</p>
  1853. </div>
  1854. </content>
  1855. </entry>
  1857. <entry>
  1858. <title type="html">Linux Virus (Not) Causing Problems</title>
  1859. <category term="/opinions" />
  1860. <id></id>
  1861. <updated>2013-08-10T02:01:00Z</updated>
  1862. <published>2013-08-10T02:01:00Z</published>
  1863. <link rel="alternate" type="text/html" href="" />
  1864. <content type="xhtml">
  1865. <div xmlns="">
  1866. <p>A while back I
  1867. <a href="">blogged</a>
  1868. about the Linux kernel site (not) being cracked. That is, someone had
  1869. indeed cracked the server, but had not been able to do any damage because
  1870. all of the files stored there were cryptographically signed in a way that
  1871. could not be forged. Strictly speaking, that was not a story about how
  1872. Linux itself is more secure than other operating systems; but the fact
  1873. that the Linux kernel developers took such precautions certainly indicates
  1874. a mindset towards security that is different from that of certain other
  1875. operating systems.</p>
  1877. <p>Yesterday ZDNet
  1878. <a href="">reported</a>
  1879. on some more direct evidence of Linux's security as an operating system,
  1880. not just the security of its kernel repository.
  1881. There is a Linux virus out there called "Hand of Thief" which apparently
  1882. can do quite a bit of damage, <em>if</em> it gets installed on your Linux system.
  1883. (By the way, contrary to what the opening sentence of this article might
  1884. lead you to believe, this is <em>not</em> the first time such a thing has happened;
  1885. Linux viruses have been in the wild for years, doing negligible damage,
  1886. for precisely the same reasons as this one is doing negligible damage,
  1887. as we'll see in a moment.)</p>
  1889. <p>The problem (at least, it's a problem from the standpoint of whoever wrote
  1890. the virus) is that qualifier I put in: <em>if</em> it gets installed on your Linux
  1891. system. The article notes:</p>
  1893. <blockquote>
  1894.  <p>Fortunately, as Limor Kessem, one of RSA's top cyber Intelligence
  1895.  experts, wrote after a conversation with the Trojan's "sales agent,"
  1896.  Hand of Thief has no good ways of infecting Linux users. Instead, the
  1897.  cracker "suggested using email and social engineering as the infection
  1898.  vector."</p>
  1899. </blockquote>
  1901. <p>That probably doesn't sound as dramatic as it actually is. When a virus
  1902. author admits that he has "no good ways" of infecting a Linux computer,
  1903. that's like a bank robber admitting he has "no good ways" of getting into
  1904. Fort Knox. He's admitting defeat, pure and simple.</p>
  1906. <p>Evidence like this is nice because it cuts through all the opinions and
  1907. arguments among experts on a question like this. As you can see on
  1908. <a href="">Wikipedia</a>,
  1909. there are indeed experts on both sides of this question. But experts can
  1910. have plenty of reasons for promoting a particular opinion, particularly
  1911. if the experts happen to also sell anti-virus software. So it's refreshing
  1912. to see evidence that doesn't depend on anything like that.</p>
  1914. <p>You may be wondering about the last part of the above quote, that talks
  1915. about "email and social engineering". Does that mean Linux won't protect
  1916. you if you accidentally click on the wrong link or open the wrong email
  1917. attachment? And don't all those anti-virus programs for Windows advertise
  1918. email scanning, link scanning, etc.?</p>
  1920. <p>It's certainly true that no operating system can protect you from yourself;
  1921. if you try hard enough to run malicious code, your computer will run
  1922. malicious code. And that's true even if you're running all those anti-virus
  1923. programs with email scanning, link scanning, etc. At best, such programs
  1924. can remind users who need reminding that they shouldn't indiscriminately
  1925. click on links or open attachments; but these days, there aren't many users
  1926. left who even need such reminding. And no such scanning program can ever
  1927. spot <em>all</em> possible malware; at best such programs are an arms race, with
  1928. malware writers constantly finding new tricks and anti-virus writers trying
  1929. to update their programs to spot them. No program can replace human judgment
  1930. about whether something looks fishy.</p>
  1932. <p>But with Windows, even if you <em>do</em> practice good Internet hygiene, you can
  1933. still get infected, because there are just too many holes in the system.
  1934. Windows was not designed from the ground up to be secure; security has been
  1935. bolted on to it as an afterthought. The very existence of the anti-virus
  1936. industry is due to this fact. (And by the way, that's also true of the Linux
  1937. wing of the anti-virus industry; if you look at the Wikipedia article I
  1938. linked to above, you'll see that even the experts who advise running
  1939. anti-virus software on Linux do so only because it allows you to scrub
  1940. files that come from Windows systems.)</p>
  1942. <p>Some Windows users may be thinking, what about the popup that appears
  1943. whenever you try to install a new program, asking if it's OK to change
  1944. system files? Won't that protect you? Yes, <em>if</em> Windows spots the attempt
  1945. to modify system files. But on Windows, there are plenty of ways for malware
  1946. to get in <em>without</em> triggering the parts of Windows that monitor for such
  1947. attempts. On Linux systems, many of which now implement a similar prompt
  1948. since it's easier than having a completely separate administrator account,
  1949. there is no way to modify any system files without triggering it, since
  1950. unless you've responded "yes" to the prompt your user account has no
  1951. permissions to change anything except your user files.</p>
  1953. <p>And let's suppose you do slip up and malicious code manages to run on your
  1954. machine. There's still a big difference between a Linux system and a
  1955. Windows system. On a Linux system, malicious code can certainly mess up
  1956. your user files. But it can't corrupt the system unless you <em>really</em> slip
  1957. up; just clicking on the wrong link or opening the wrong email attachment
  1958. won't do it. So cleaning things up is easy, because you can still depend
  1959. on the system files to be clean. If you get malware on a Windows system,
  1960. you can't really trust anything, and most often the only remedy is to wipe
  1961. the hard drive and reinstall.</p>
  1963. <p>Of course now all the Mac users are thinking, doesn't OS X have the same
  1964. security features as Linux? After all, they're both variants of Unix,
  1965. which is the original source of the security model. That's quite true.
  1966. But then why is there anti-virus software for OS X?</p>
  1968. <p>There are experts on both sides of this question too; some say
  1969. <a href="">OS X doesn't need anti-virus software</a>,
  1970. for basically the same reasons that Linux doesn't. I'm inclined to
  1971. agree with this, and to interpret the fact that companies sell OS X
  1972. anti-virus software as saying more about those companies' ethics than
  1973. about OS X's security or lack thereof. But maybe that's just me.</p>
  1975. <p>However that may be, the article also contains an interesting tidbit:</p>
  1977. <blockquote>
  1978.  <p>The most recent version of OS X...includes the GateKeeper function
  1979.  that by default prevents Mac users from installing anything other than
  1980.  Apple-approved software. And the lack of Java and Flash plugins removes
  1981.  the temptation to install fake versions of both--previously the principal
  1982.  vectors of infection for Macs.</p>
  1983. </blockquote>
  1985. <p>In other words, Apple's approach to keeping OS X secure is to make it
  1986. less functional. I'm no fan of Java or Flash, but the fact remains that
  1987. a <em>lot</em> of Internet content is packaged that way, so just punting and
  1988. saying you can't install it isn't very helpful. It's not as though it
  1989. can't be done: Linux systems manage to run Java and Flash without
  1990. compromising security, by making sure that secure versions of them are
  1991. available in cryptographically signed repositories, so you can check
  1992. that they're the right versions when you install them. And although
  1993. third parties can write "Apple-approved software"--if they're willing
  1994. to pay Apple for the privilege--the quantity of such software available
  1995. is nothing like the quantity that's available for Windows or Linux.</p>
  1997. <p>All of which is just another reason for this:</p>
  1999. <div class="codehilite"><pre><span class="gp">peter@localhost:~$</span> uname
  2000. <span class="go">Linux</span>
  2001. </pre></div>
  2004. <p>(Update: there is a discussion of this post on
  2005. <a href="">Hacker News</a>.)</p>
  2006. </div>
  2007. </content>
  2008. </entry>
  2010. <entry>
  2011. <title type="html">Lawyer Humor</title>
  2012. <category term="/general" />
  2013. <id></id>
  2014. <updated>2013-06-20T03:25:00Z</updated>
  2015. <published>2013-06-20T03:25:00Z</published>
  2016. <link rel="alternate" type="text/html" href="" />
  2017. <content type="xhtml">
  2018. <div xmlns="">
  2019. <p>This is just a quick link to a
  2020. <a href="">hilarious response to a cease and desist letter</a>
  2021. (hat tip:
  2022. <a href="">Hacker News</a>).
  2023. If you ever need a lawyer, hopefully you will find one with a sense
  2024. of humor like the one that wrote this response.</p>
  2025. </div>
  2026. </content>
  2027. </entry>
  2029. <entry>
  2030. <title type="html">Be Careful What You Wish For</title>
  2031. <category term="/opinions" />
  2032. <id></id>
  2033. <updated>2013-04-25T22:12:00Z</updated>
  2034. <published>2013-04-25T22:12:00Z</published>
  2035. <link rel="alternate" type="text/html" href="" />
  2036. <content type="xhtml">
  2037. <div xmlns="">
  2038. <p><a href="">Fire Dog Lake</a>
  2039. is angry about the recent Senate vote that
  2040. <a href="">killed the Toomey-Manchin background check amendment</a>
  2041. to the latest gun control bill. However, the anger is not directed at
  2042. the Democrats that voted against the amendment, but at those who voted
  2043. <em>for</em> it.
  2044. What gives?</p>
  2046. <blockquote>
  2047.  <p>If the 51 Democrats who voted for the amendment really wanted to see
  2048.  it passed in the Senate, they could have passed it. They could have
  2049.  voted to eliminate the filibuster and than pass the bill with a simple
  2050.  majority vote like the Constitution specifies.</p>
  2051. </blockquote>
  2053. <p>Wait, what? Where, exactly, does the Constitution say that bills must
  2054. pass by simple majority vote? There is nothing in the Constitution about
  2055. what percentage of either house's vote is required to pass a bill.
  2056. <a href="">Article I</a>
  2057. says that a majority of each house constitutes a quorum to do business,
  2058. but that just means being able to debate bills or hold votes; it's not
  2059. the same as the standard for passing a bill. It also says each house can
  2060. determine its own rules of procedure, of which the filibuster rule is an
  2061. example; so, far from being against the Constitution, the filibuster looks
  2062. like an example of the Constitution being used as it was written.</p>
  2064. <p>Of course, nothing in constitutional law is ever that simple; if it were,
  2065. I would never have been able to get
  2066. <a href="">so</a>
  2067. <a href="">many</a>
  2068. <a href="">posts</a>
  2069. out of the subject. Nearly a year ago,
  2070. <a href="">Ezra Klein blogged</a>
  2071. about a suit that was submitted to the Supreme Court arguing that the
  2072. filibuster rule should be reviewed for possible unconstitutionality.
  2073. The argument was based on three main points. One is simply that, although
  2074. the Constitution says that each house of Congress determines its own rules
  2075. of procedure, those rules cannot violate other Constitutional provisions,
  2076. and the Supreme Court has in the past reviewed procedural rules for
  2077. constitutionality. The second, more interesting point is that there is
  2078. apparently an "established rule of construction" called "expressio unius
  2079. est exclusio alterius", which I won't bother translating literally from
  2080. the Latin, but which basically means that, since the Constitution does
  2081. explicitly state particular exceptions to majority rule (there are six
  2082. in total--for example, the two-thirds majority in each house that is
  2083. required to override a presendential veto), there is an implication that
  2084. there are no <em>other</em> exceptions, since if there were, it would have stated
  2085. them. (A contrast is drawn here with the Bill of Rights, which does
  2086. explicitly say that the rights enumerated in the Constitution are not
  2087. a complete list.) The third point is that the filibuster rule isn't what
  2088. the Founders intended; for example, by decreasing the number of Senators
  2089. required to vote down a bill, it upsets the "carefully crafted balance"
  2090. that the Founders set up between the large states and the small states.</p>
  2092. <p>The only thing these arguments prove to me is that constitutional lawyers
  2093. can do sophistry about as well as Supreme Court justices can. This will
  2094. come as no surprise if you have read the previous posts I linked to
  2095. above, since one of the main points I argued in those posts is that if
  2096. the standard were what the Founders intended, most of what passes for
  2097. constitutional law would be out the window. But even if we just consider
  2098. the merits of this particular case, the arguments don't hold up. For
  2099. example, the idea that if the Bill of Rights didn't include the Ninth
  2100. Amendment, there wouldn't be any other rights than those explicitly given,
  2101. doesn't hold water even with most constitutional lawyers; many of them
  2102. have argued that the Ninth Amendment is not really necessary, since
  2103. what it says is already implied by the rest of the Constitution and the
  2104. Bill of Rights.</p>
  2106. <p>As for the "carefully crafted balance" between large and small states, if
  2107. that's the objective, why should a 51 percent majority be the default?
  2108. There's no reason why that particular number must always magically be the
  2109. "right" balance. The brief for the lawsuit notes that at the time of the
  2110. founding, the smallest seven of the 13 states, representing 27 percent of
  2111. the population, could command a majority in the Senate; but now, with the
  2112. filibuster rule, the smallest 21 of the 50 states, representing only 11
  2113. percent of the population, can kill a bill. Clearly this is a major
  2114. change in the balance of power, but what makes the original balance any
  2115. more "right" than the current one--or than some point in between? The
  2116. brief makes no argument for what the right balance should be; it just
  2117. assumes that a simple majority must be right, because, well, that's what
  2118. "democracy" means, right? (We'll come back to this last point.)</p>
  2120. <p>Of course we know what the debate about filibuster reform is really about.
  2121. As I said in my
  2122. <a href="">post on the fiscal cliff</a>, we have
  2123. long since stopped thinking of the Constitution as a set of underlying
  2124. rules for a stable society, the preservation of which is more important
  2125. than the action we take on any particular issue. We no longer see our
  2126. government as a tool for arriving at consensus solutions to difficult
  2127. political problems; we now think of it as a vehicle for imposing our
  2128. particular beliefs on everybody else. And of course we get frustrated
  2129. when the rules seem to obstruct our efforts to do that, and we blame it
  2130. on the rules, instead of on ourselves. As the National Review recently
  2131. <a href="">noted</a>,
  2132. not so very long ago it was the Republicans who were complaining about the
  2133. filibuster rule, and the Democrats who were saying how important it was to
  2134. keep it as an aid to wise and mature government. But that same article
  2135. argues that the Republicans should <em>agree</em> to end the filibuster rule,
  2136. because, even though that will make it easier in the short run for the
  2137. Democrats to impose their policies, it will also make it easier in the
  2138. longer run for the Republicans to impose theirs.</p>
  2140. <p>To me this is a classic case of "be careful what you wish for". The
  2141. problem is not that it's too hard to pass laws; it's that it's too <em>easy</em>.
  2142. It's too easy for Congress to change the rules by which our society
  2143. operates, and as a result the rules are changed so often that nobody
  2144. can keep up. Or, rather, no ordinary citizen can keep up, which means
  2145. that laws are written by lobbyists and special interests, and are voted
  2146. on by legislators that for the most part have not even read them. Is
  2147. <em>that</em> what the Founders intended?</p>
  2149. <p>I said we would come back to that word, "democracy". That's another word
  2150. that does not appear in the Constitution. There is, indeed, only one
  2151. reference to any form of government in the entire document; Article IV,
  2152. Section 4, says: "The United States shall guarantee to every State in
  2153. this Union a Republican Form of Government". The original Constitution
  2154. had Senators elected by State legislatures; not until the 17th Amendment,
  2155. in 1913, were Senators elected by popular vote in their States. Even
  2156. the House of Representatives, which has always been elected by the people
  2157. directly, has never had each representative representing exactly the same
  2158. number of people. Indeed, the latest
  2159. <a href="">redistricting</a>
  2160. has the largest district containing almost twice the number of people as
  2161. the smallest. So even if we do have a democracy in the United States, it
  2162. is an imperfect one.</p>
  2164. <p>The Founders, if they knew anything, knew that any government must be
  2165. imperfect. Their goal was not to find the perfect set of rules, or even to
  2166. find the perfect framework for coming up with a set of rules. Their goal
  2167. was simply to come up with something that worked better than the Articles
  2168. of Confederation, and to include ways of amending it as time went on. If
  2169. simple majority rule was that important to them, they could have said it
  2170. explicitly; and for that matter, if it's really that important to us, we
  2171. could amend the Constitution now to say that all votes in Congress are
  2172. simple majority votes unless specified otherwise.</p>
  2174. <p>But the Founders could not protect us from ourselves. No system of
  2175. government, no set of rules, can protect us if our response when the rules
  2176. get in our way is to blame the rules. Abolishing the filibuster rule won't
  2177. fix our government; if anything, it will make it easier to mess things up
  2178. by making changes without a clear understanding of the consequences.</p>
  2179. </div>
  2180. </content>
  2181. </entry>
  2183. <entry>
  2184. <title type="html">In Defense Of Marriage?</title>
  2185. <category term="/opinions" />
  2186. <id></id>
  2187. <updated>2013-03-29T02:04:00Z</updated>
  2188. <published>2013-03-29T02:04:00Z</published>
  2189. <link rel="alternate" type="text/html" href="" />
  2190. <content type="xhtml">
  2191. <div xmlns="">
  2192. <p>Some interesting items have come out of yesterday's
  2193. <a href="">oral arguments</a>
  2194. before the Supreme Court on the Defense of Marriage Act case. Since
  2195. I've blogged about this case
  2196. <a href="">before</a>,
  2197. I wanted to take a look at the Court's handling of it.</p>
  2199. <p>The first thing I noted is that the Court appointed an amicus curiae to
  2200. present arguments for the position that the Court did not have jurisdiction
  2201. to hear the case. This is unusual since neither party to the case had taken
  2202. this position. However, the case is also unusual in that both parties
  2203. <em>agree</em> with the lower court decision; neither one is asking the Supreme
  2204. Court to reverse it. This is what raises the jurisdictional issue: since
  2205. there is no actual controversy between the parties, why should the Court
  2206. rule on the case at all?</p>
  2208. <p>The reason the case is before the Court is that the Obama administration
  2209. <em>wants</em> the Court to uphold the lower court decision, which would have
  2210. the effect of striking down DOMA, which the administration believes is
  2211. unconstitutional. Of course this raises the question: why is the
  2212. administration enforcing a law it thinks is unconstitutional? The
  2213. administration's position is that it is obligated to enforce laws even
  2214. if it disagrees with them, because Article II of the Constitution says
  2215. that the President shall "take care that the laws are faithfully
  2216. executed". But Chief Justice Roberts responded to that:</p>
  2218. <blockquote>
  2219.  <p>[T]he Executive's obligation to execute the law includes the obligation
  2220.  to execute the law consistent with the Constitution. And if he has made
  2221.  a determination that executing the law by enforcing the terms is
  2222.  unconstitutional, I don't see why he doesn't have the courage of his
  2223.  convictions and execute not only the statute, but do it consistent with
  2224.  his view of the Constitution, rather than saying, oh, we'll wait till the
  2225.  Supreme Court tells us we have no choice.</p>
  2226. </blockquote>
  2228. <p>This reminds me of Roberts' opinion on the
  2229. <a href="">individual mandate</a>
  2230. portion of Obamacare, where he basically said that the Court can only rule
  2231. on whether a law is constitutional, and a law that is constitutional may
  2232. still be a bad law; it's the job of elected representatives, not the Court,
  2233. to determine whether a law is good or bad. It will be interesting to see
  2234. if any of this comes through in the final opinion.</p>
  2236. <p>Later on, the point was raised that President Obama, who made a
  2237. determination that he believed the Act was unconstitutional, is not
  2238. the President who signed it into law--that was President Clinton, in
  2239. 1996--which could affect the decision on whether or not to enforce it.
  2240. If a President determines that a bill is unconstitutional when it reaches
  2241. his desk for signature, he can simply veto it. But if a President believes
  2242. that a law signed by a previous President is unconstitutional, he may
  2243. still choose to enforce it out of deference to the previous President who
  2244. signed the bill and the previous Congress that passed it. A ruling from
  2245. the Supreme Court would break this kind of deadlock. (Later on it was
  2246. pointed out that when DOMA was being considered by Congress, it asked
  2247. the Clinton Justice Department three times if the proposed Act was
  2248. constitutional, and all three times the response was that it was.)</p>
  2250. <p>Another interesting point arose during the next argument: since the
  2251. Executive Branch is basically asking the Court to declare a law passed by
  2252. Congress unconstitutional, shouldn't Congress be a party to the case as
  2253. well? One key aspect of this is that the Executive Branch is the one
  2254. litigating the case, making decisions as to how it's argued, whether or
  2255. not to appeal to the next level of courts, etc., even though the Executive
  2256. Branch doesn't believe the law is constitutional. As the lawyer making the
  2257. argument puts it:</p>
  2259. <blockquote>
  2260.  <p>It's a conflict of interest. They're the ones that are making litigation
  2261.  decisions to promote the defense of a statute they want to see invalidated.
  2262.  And if you want to see the problems with their position, look at Joint
  2263.  Appendix page 437. You will see the most anomalous motion to dismiss in
  2264.  the history of litigation: A motion to dismiss, filed by the United States,
  2265.  asking the district court not to dismiss the case.</p>
  2266. </blockquote>
  2268. <p>Justice Kennedy remarked in response that this "would give you intellectual
  2269. whiplash. I'm going to have to think about that."</p>
  2271. <p>But in the rebuttal argument of the amicus, a good response to this was
  2272. given:</p>
  2274. <blockquote>
  2275.  <p>[O]nce the litigation is enacted, Congress's authority to supervise it is
  2276.  at an end. It goes over to the Executive Branch. And whether the Executive
  2277.  Branch does it well or badly in the view of Congress, it's in its domain.
  2278.  And separation of powers will not be meaningful if all it means is the
  2279.  Congress has to stay out unless it thinks that the President is doing it
  2280.  badly.</p>
  2281. </blockquote>
  2283. <p>In other words, for better or worse, the Constitution says that the
  2284. President, not Congress, executes the laws, and that includes litigating
  2285. them when they are challenged in court.</p>
  2287. <p>I was pleased to see that the arguments addressed the issue of government
  2288. benefits given to married people, since I said
  2289. <a href="">previously</a>
  2290. that this was the key issue, because of the Constitution's guarantee of
  2291. equal protection of the laws to all citizens. But I was disappointed to
  2292. see that not much attention was paid to the fact that the 14th Amendment
  2293. applies the equal protection provision to the States, not just the Federal
  2294. government, so if it is unconstitutional for the Federal government to
  2295. restrict the definition of marriage on equal protection grounds, it must
  2296. be equally unconstitutional for a State to do so. In fact, the first
  2297. argument on the merits explicitly assumes the contrary:</p>
  2299. <blockquote>
  2300.  <p>[T]he legal question on the merits before this Court is actually quite
  2301.  narrow. On the assumption that States have the constitutional option either
  2302.  to define marriage in traditional terms or to recognize same-sex marriages
  2303.  or to adopt a compromise like civil unions, does the Federal Government
  2304.  have the same flexibility or must the Federal Government simply borrow the
  2305.  terms in State law?</p>
  2306. </blockquote>
  2308. <p>Justice Ginsburg did raise a question about what happens if Federal law
  2309. doesn't recognize a marriage that a State has recognized:</p>
  2311. <blockquote>
  2312.  <p>[I]f we are totally for the States' decision that there is a marriage
  2313.  between two people, for the Federal Government then to come in to say no
  2314.  joint return, no marital deduction, no Social Security benefits; your
  2315.  spouse is very sick but you can't get leave; people--if that set of
  2316.  attributes, one might well ask, what kind of marriage is this?</p>
  2317. </blockquote>
  2319. <p>This question eventually led to a good summation of the 1996 Congress's
  2320. motivation for passing DOMA in the first place:</p>
  2322. <blockquote>
  2323.  <p>Congress is worried that people are going to go there [to Hawaii, which
  2324.  was close to adopting same-sex marriage at the time through a state court
  2325.  ruling], go back to their home jurisdictions, insist on the recognition
  2326.  in their home jurisdictions of their same-sex marriage in Hawaii, and then
  2327.  the Federal Government will borrow that definition, and therefore, by the
  2328.  operation of one State's State judiciary, same-sex marriage is basically
  2329.  going to be recognized throughout the country. And what Congress says is,
  2330.  wait a minute. Let's take a timeout here. This is a redefinition of an
  2331.  age-old institution. Let's take a more cautious approach where every
  2332.  sovereign gets to do this for themselves.</p>
  2333. </blockquote>
  2335. <p>But, as the ensuing discussion made clear, DOMA doesn't actually do this.
  2336. It privileges the traditional definition of marriage, so States that uphold
  2337. that definition get to "decide for themselves", but States that adopt
  2338. same-sex marriage do not; their decision is constrained by the fact that
  2339. they know the Federal government under DOMA does not recognize same-sex
  2340. marriage. The lawyer tried to argue that DOMA simply codifies a uniform
  2341. definition of marriage for the purposes of Federal law, on the theory that
  2342. when previous Congresses passed laws that used the term "marriage", they
  2343. meant that term under its traditional definition. But Justice Kagan pointed
  2344. out:</p>
  2346. <blockquote>
  2347.  <p>[F]or the most part and historically, the only uniformity that the
  2348.  Federal Government has pursued is that it's uniformly recognized the
  2349.  marriages that are recognized by the State. So, this was a real difference
  2350.  in the uniformity that the Federal Government was pursuing. And it suggests
  2351.  that maybe something--maybe Congress had something different in mind than
  2352.  uniformity.</p>
  2353. </blockquote>
  2355. <p>Which of course it did; Congress explicitly said that the law was for the
  2356. purpose of protecting the traditional definition of marriage, and did not
  2357. recognize same-sex marriage because homosexuality was morally wrong. This
  2358. fact was referred to several times during the arguments.</p>
  2360. <p>The Solicitor General did make an equal protection argument when it was
  2361. his turn to speak:</p>
  2363. <blockquote>
  2364.  <p>The equal protection analysis in this case should focus on two fundamental
  2365.  points: First, what does Section 3 do; and second, to whom does Section 3 do
  2366.  it? What Section 3 does is exclude from an array of Federal benefits
  2367.  lawfully married couples.</p>
  2368. </blockquote>
  2370. <p>But this still applies the equal protection analysis only to the Federal
  2371. law. Later on, the question of State laws outlawing same-sex marriage was
  2372. brought up, and the Solicitor General did say that those would have to be
  2373. looked at on equal protection grounds as well. But then he basically said
  2374. that's not relevant to this case, which is only about what a Federal law
  2375. can properly exclude:</p>
  2377. <blockquote>
  2378.  <p>They [State laws] have to be analyzed under equal protections principles,
  2379.  but whatever is true about the other situations, in the situation in which
  2380.  the couple is lawfully married for purposes of State law and the exclusion
  2381.  is a result of DOMA itself, the exclusion has to be justified under this
  2382.  Court's equal protection analysis, and DOMA won't do it.</p>
  2383. </blockquote>
  2385. <p>In fact, towards the end of the Solicitor General's brief, Justice Sotomayor
  2386. asked him point blank:</p>
  2388. <blockquote>
  2389.  <p>[Y]our bottom line is, it's an equal protection violation for the Federal
  2390.  Government, and all States as well?</p>
  2391. </blockquote>
  2393. <p>And the Solicitor General responded yes; but then Justice Sotomayor asked:</p>
  2395. <blockquote>
  2396.  <p>Is there any argument you can make to limit this to this case, vis-a-vis
  2397.  the Federal Government and not the States?</p>
  2398. </blockquote>
  2400. <p>In other words, we don't want to hear about whether States outlawing
  2401. same-sex marriage is unconstitutional; that's not part of this case. In
  2402. fact, the ensuing discussion suggests that, not only is it not part of this
  2403. case, but that fact in itself gives the equal protection argument less
  2404. weight in this case, in the Justices' minds. It seems like the Court might
  2405. shy away from ruling DOMA unconstitutional on equal protection grounds,
  2406. precisely <em>because</em> that would imply that State laws outlawing same-sex
  2407. marriage are also unconstitutional on equal protection grounds.</p>
  2409. <p>During the next brief on the merits, Chief Justice Roberts posed an
  2410. interesting question to the lawyer for the woman who originally brought
  2411. the case in lower courts (she was married in Ontario to a same-sex partner,
  2412. but was forced to pay estate tax in the US when her spouse died because
  2413. her marriage was not recognized under DOMA): would there be an issue if
  2414. Congress passed a law <em>recognizing</em> same-sex couples as being married for
  2415. purposes of Federal law even if the State they lived in did not permit
  2416. same-sex marriage? The concern here is not equal protection but federalism:
  2417. can Congress <em>ever</em> adopt a different definition of marriage than the
  2418. States?</p>
  2420. <p>The Chief Justice commented that everyone kept "returning to the Equal
  2421. Protection Clause every time I ask a federalism question". But given the
  2422. discussion that followed, I can see why. The lawyer kept trying to say that
  2423. there isn't a single blanket answer to the federalism question because it
  2424. would depend on the circumstances, but the Justices kept pressing her to
  2425. give a blanket answer anyway. This makes one wonder whether the Court is
  2426. looking for a way to rule on DOMA's constitutionality without having to
  2427. do so on equal protection grounds, because of the broad implications that
  2428. the latter type of ruling would have.</p>
  2430. <p>Towards the end, the issue of benefits was brought up once more:</p>
  2432. <blockquote>
  2433.  <p>[W]hen somebody moves from New York to North Carolina, they can lose
  2434.  their benefits. The Federal Government uniquely, unlike the 50 States,
  2435.  can say, well, that doesn't make any sense, we are going to have the same
  2436.  rule. We don't want somebody, if they are going to be transferred in the
  2437.  military from West Point to Fort Sill in Oklahoma, to resist the transfer
  2438.  because they are going to lose some benefits.</p>
  2439. </blockquote>
  2441. <p>But nobody made the obvious rebuttal that, if DOMA is upheld, a same-sex
  2442. couple that moves from West Point to Fort Sill in Oklahoma won't <em>get</em> the
  2443. benefits in the first place; to say that well, then they won't have to
  2444. worry about losing them, is not much comfort.</p>
  2446. <p>But just after this, the most interesting argument is given, right at the
  2447. end of the transcript. I'll quote it at some length because I suspect it
  2448. is going to be referred to a lot:</p>
  2450. <blockquote>
  2451.  <p>Now the Solicitor General wants to say: Well, it [the passage of DOMA]
  2452.  was want of careful reflection? Well, where do we get careful reflection
  2453.  in our system? Generally, careful reflection comes in the democratic
  2454.  process. The democratic process requires people to persuade people.</p>
  2456.  <p>The reason there has been a sea change [in public opinion about same-sex
  2457.  marriage] is a combination of political power, as defined by this Court's
  2458.  cases as getting the attention of lawmakers; certainly they have that. But
  2459.  it's also persuasion. That's what the democratic process requires. You have
  2460.  to persuade somebody you're right. You don't label them a bigot. You don't
  2461.  label them as motivated by animus. You persuade them you are right.</p>
  2463.  <p>That's going on across the country. Colorado, the State that brought you
  2464.  Amendment 2, has just recognized civil unions. Maine, that was pointed to
  2465.  in the record in this case as being evidence of the persistence of
  2466.  discrimination because they voted down a statewide referendum, the next
  2467.  election cycle it came out the other way. And the Federal Congress is not
  2468.  immune. They repealed "Don't Ask, Don't Tell." Allow the democratic process
  2469.  to continue.</p>
  2470. </blockquote>
  2472. <p>What is interesting about this is that the lawyer invokes "the democratic
  2473. process", which is a pet phrase of Justice Scalia, and he invokes it to
  2474. make precisely the kind of argument that Scalia likes: the Federal
  2475. government shouldn't interfere in issues like this. But this lawyer is
  2476. arguing <em>for upholding DOMA</em>. In other words, he is arguing that the
  2477. Federal government should not interfere in the democratic process, and
  2478. his definition of "not interfering" is to uphold a Federal law that imposes
  2479. a definition of marriage which the "democratic process" in a number of
  2480. States has rejected.</p>
  2482. <p>Of course this kind of sophistry fits right in with the view I have taken in
  2483. <a href="">my</a>
  2484. <a href="">previous</a>
  2485. <a href="">posts</a>
  2486. on the Supreme Court. Compare, for example, the position I expect Justice
  2487. Scalia to take in this case, in line with the above argument, with the
  2488. position he has taken on abortion in a number of cases (most notably in
  2489. his dissent in Planned Parenthood v. Casey):</p>
  2491. <ul>
  2492. <li>
  2493. <p>Abortion: the Constitution says nothing about it, and long-standing
  2494. tradition of our society opposes it, so we should let the States decide;
  2495. the Supreme Court should get out of this area.</p>
  2497. </li>
  2498. <li>
  2499. <p>Same-sex marriage: the Constitution says nothing about it, and
  2500. long-standing tradition of our society opposes it, so we should <em>not</em>
  2501. let the States decide; the Supreme Court should uphold a Federal law
  2502. that imposes a uniform definition of marriage, for purposes of all
  2503. Federal laws and regulations, with which not all States agree.</p>
  2505. </li>
  2506. </ul>
  2507. <p>As far as the actual question at issue is concerned, I've made my position
  2508. clear before and I see no reason to change it: to me, the equal protection
  2509. argument is sufficient to strike down DOMA. It is true, as I noted above,
  2510. that this argument would also imply that State laws forbidding same-sex
  2511. marriage are unconstitutional. But note carefully <em>why</em> this would be. It
  2512. isn't because States can't define what "marriage" is; it's because States,
  2513. like the Federal government, attach lots of benefits to being "married".
  2514. If States are going to do that, then defining "marriage" in any way that
  2515. excludes a class of people violates equal protection. It's that simple.</p>
  2517. <p>The right solution to this problem would be for the Federal and State
  2518. benefits to attach, not to "marriage", but to some legal status that has
  2519. no social connotations. For example, people could make a legal commitment
  2520. to form a "household", or designate each other as "significant others",
  2521. in order to get the benefits. Of course such a legal commitment would
  2522. have to be something more than just saying so: it would have to involve
  2523. the same sort of signing of contracts and filing of paperwork and agreeing
  2524. to legal conditions that marriages and civil unions do now. But it would
  2525. be a separate thing from the social designation of a given couple as
  2526. "married", which could then be left up to whatever social circles the
  2527. couple belonged to.</p>
  2529. <p>Unfortunately, such a solution is almost certainly not politically viable
  2530. in the US today. We just can't help passing laws that help to increase
  2531. friction between different parts of our society instead of helping to
  2532. reduce it. We just can't help using political power, when we have it, to
  2533. try to entrench our particular view of how things should be, instead of
  2534. making it easier for people with different views to coexist. That's a
  2535. shame, since allowing people with different views to coexist is what the
  2536. United States of America is supposed to be about.</p>
  2537. </div>
  2538. </content>
  2539. </entry>
  2541. <entry>
  2542. <title type="html">Tolkien Redux</title>
  2543. <category term="/opinions" />
  2544. <id></id>
  2545. <updated>2013-02-11T01:55:00Z</updated>
  2546. <published>2013-02-11T01:55:00Z</published>
  2547. <link rel="alternate" type="text/html" href="" />
  2548. <content type="xhtml">
  2549. <div xmlns="">
  2550. <p>I hadn't intended to say any more about the Peter Jackson films after my
  2551. <a href="">last post</a>,
  2552. but then I came across a series of
  2553. <a href="">three</a>
  2554. <a href="">reviews</a>
  2555. <a href="">of the movies</a>
  2556. by Andrew Rilstone, and found that I have more to say after all. (This
  2557. will come as no surprise to those who know me, of course.)</p>
  2559. <p>The first review, of Fellowship of the Ring, makes a good general
  2560. observation:</p>
  2562. <blockquote>
  2563.  <p>This is not Lord of the Rings: it is only the story of Lord of the
  2564.  Rings. In movies, 'story' is all. The canons of script writing tell
  2565.  us that if a scene does not directly advance the plot, you must cut
  2566.  it out, and throw it away. But story is very rarely the most important
  2567.  thing in a novel. Name of the Rose is a rambling book about medieval
  2568.  church politics and semiology. The movie cut out nearly all the
  2569.  theology and all the philosophy, arguably missing the entire point
  2570.  of the book: but it turned out that the bit that was left over was
  2571.  still a rather engaging little whodunit. (Umberto Eco called it a
  2572.  palimpsest, but then he would, wouldn’t he.) If you cut all the
  2573.  elegant writing and ironic observations out of Pride and Prejudice,
  2574.  it turns out that you are still left with quite jolly little Barbara
  2575.  Cartland country house romances than that you can show in movie-houses
  2576.  and before the watershed on BBC 2. What you do not have is anything
  2577.  very much to do with Jane Austen. The point of Lord of the Rings is
  2578.  the Middle-earth setting: the history, the back-story, the languages,
  2579.  the little poetic asides. In filleting the book for the screen, and
  2580.  extracting the story, all this has be thrown out--but what is left,
  2581.  ring-fillet, is still plenty for a decent, entertaining fantasy film.</p>
  2582. </blockquote>
  2584. <p>As a lead-in to Rilstone's reviews of all three films, taken together,
  2585. this brings up an important point. I noted in my
  2586. <a href="">previous</a>
  2587. <a href="">posts</a>
  2588. that the films are all right if taken just as fantasy films, not as
  2589. adaptations of Tolkien's books for the screen. But Rilstone gives
  2590. plenty of ways in which the subsequent films, even taken just as
  2591. fantasy films, are not as good as the first one; in his view, the
  2592. films taken as a whole do not live up even to the standard set in
  2593. the quote above. I'll comment further on that after we've seen some
  2594. of the specific comments he makes.</p>
  2596. <p>There is at least one place where I think Rilstone gives the movies
  2597. more credit than they deserve:</p>
  2599. <blockquote>
  2600.  <p>Tolkien never makes it particularly clear why Aragorn has been
  2601.  wandering in wastelands when he could go home at any time and become
  2602.  king. Jackson’s elegant solution--that he is at some level afraid that
  2603.  he will become corrupt in the way that Isildur did--is true to the
  2604.  spirit of the book, if not to its letter.</p>
  2605. </blockquote>
  2607. <p>I have two problems with this. One is that it's quite clear in the
  2608. books why Aragorn doesn't just walk into Gondor and claim the Kingship.
  2609. First, Aragorn's ancestor, Arvedui, had already tried and failed;
  2610. Gondor rejected his claim. So it's clearly not as simple as Aragorn
  2611. simply showing up in Minas Tirith and saying, hey, I'm King now. Second,
  2612. if he did try to claim the Kingship, Sauron would destroy Gondor; whereas
  2613. by working in secret as he does, he can improve the odds without
  2614. provoking Sauron, as when he leads an attack against Umbar in disguise
  2615. (as told in the Tale of Aragorn and Arwen) to remove, or at least
  2616. postpone, the threat posed to Gondor by the Corsairs. Finally, it is
  2617. perfectly clear from Denethor's words to Gandalf and Pippin that Aragorn
  2618. would have caused a severe political upheaval in Gondor if he had tried
  2619. to claim the Kingship (and Aragorn knows this because he served Denethor's
  2620. father in disguise and so was familiar with Denethor's character and
  2621. views), and as he tells the other Captains after the Battle of the
  2622. Pelennor Fields, he has "no mind for strife except with the Enemy and
  2623. his counsellors". In short, Tolkien does give more than enough back
  2624. story to show that the course Aragorn actually takes is the best choice
  2625. for Gondor, all things considered.</p>
  2627. <p>The second, more important issue is that Jackson's "solution" is anything
  2628. but elegant. I've already
  2629. <a href="">gone into this in detail</a>,
  2630. so I won't belabor it here, but briefly, Aragorn in the books does
  2631. not show that kind of internal conflict because he, like all of Tolkien's
  2632. central characters, sees himself as a moral agent, able to make his own
  2633. choices, not controlled by external forces. What's more, Tolkien makes
  2634. clear <em>why</em> Aragorn has that sense of moral agency. When he finds out his
  2635. lineage from Elrond, he has just "returned to Rivendell after great deeds
  2636. in the company of the sons of Elrond". In other words, he has been given
  2637. the chance to prove himself, and to develop the self-knowledge and
  2638. self-confidence that enables him to <em>know</em> that he can make the right
  2639. choices. He has no reason to fear that he might fail at the test just
  2640. because his ancestor Isildur did, because he knows himself well enough
  2641. to know that he is his own person, regardless of what his ancestors did
  2642. or failed to do.</p>
  2644. <p>Jackson's "solution" takes away that crucial feature of Aragorn's
  2645. character, and it's not a positive change. We don't need another
  2646. angst-ridden post-modern antihero, and Tolkien certainly did not intend
  2647. Aragorn to be one. So Aragorn in the movies is <em>not</em> by any means "true
  2648. to the spirit of the book".</p>
  2650. <p>I also have a minor quibble with a comment Rilstone makes when he talks
  2651. about how the "battle between the wizards" was portrayed in the first
  2652. film:</p>
  2654. <blockquote>
  2655.  <p>But was it essential for the duel of the wizards (left off stage in
  2656.  the book) to resort quite so obviously to Star Wars Jedi trickery, with
  2657.  Gandalf and Saruman levitating each other all round the joint. One just
  2658.  hopes that McKellen will get through part 2 without having to say 'These
  2659.  aren't the hobbits you're looking for.'</p>
  2660. </blockquote>
  2662. <p>So far, so good (although, as we'll see in a moment, the criticism here
  2663. could have been even more pointed). But then Rilstone goes on:</p>
  2665. <blockquote>
  2666.  <p>(It is interesting, by the way, to speculate about how Tolkien would
  2667.  have visualized the battle, had be been required to do so. I think
  2668.  perhaps that Saruman and Gandalf would have stared at each other until
  2669.  Gandalf's 'Will' was overcome. Which would not, I grant you, have been
  2670.  a very visual moment.)</p>
  2671. </blockquote>
  2673. <p>We don't have to speculate; Tolkien <em>did</em> visualize the "battle", through
  2674. Gandalf, when he describes it to the Council of Elrond. And Gandalf's
  2675. account makes clear that there wasn't any "battle", and Gandalf's will
  2676. was <em>not</em> overcome. Saruman tried to convince Gandalf to help him find
  2677. and wield the Ring, and Gandalf made his rejection as plain as day:</p>
  2679. <blockquote>
  2680.  <p>'"Saruman," I said, standing away from him, "only one hand at a time
  2681.  can wield the One, and you know that well, so do not trouble to say <em>we</em>!
  2682.  But I would not give it, nay, I would not give even news of it to you,
  2683.  now that I learn your mind. You were head of the Council, but you have
  2684.  unmasked yourself at last. Well, the choices are, it seems, to submit to
  2685.  Sauron, or to yourself. I will take neither. Have you others to offer?"'</p>
  2686. </blockquote>
  2688. <p>A fair rendering of this into "movie-speak" might have shown Saruman
  2689. putting Gandalf in some kind of magical confinement, yes, but it would
  2690. certainly not show Gandalf's will being overcome. What overcomes him in
  2691. the book is main force: Saruman's servants take him and confine him at
  2692. the pinnacle of Orthanc. Saruman does <em>not</em> win any struggle of will with
  2693. Gandalf: quite the reverse, he <em>loses</em> that struggle, and that is why
  2694. he is forced to confine Gandalf by force. Saruman is trying to corrupt
  2695. Gandalf morally, and he fails. The "Jedi" portrayal in the movie makes
  2696. it seem as though the struggle isn't a moral struggle at all; it's just
  2697. a question of which one has stronger telekinetic powers. Rilstone's
  2698. suggested visualization has the same problem: it makes it seem like it's
  2699. just a question of who can stare harder.</p>
  2701. <p>But these are relatively minor objections to Rilstone's review of
  2702. Fellowship of the Ring. Let's get on to Rilstone's second review, of
  2703. The Two Towers, which says this early on:</p>
  2705. <blockquote>
  2706.  <p>I had a long list of quibbles with Fellowship of the Ring, but I was
  2707.  never in any doubt that it was a pretty successful attempt to translate
  2708.  books 1 and 2 of Lord of the Rings into a movie idiom. 'As good as could
  2709.  be expected under the circumstances' was about the rudest thing anyone
  2710.  sensible could say about it. My feelings towards the Two Towers, on the
  2711.  other hand, can best be summed up as 'Hey, what?'</p>
  2712. </blockquote>
  2714. <p>A key reason for this reaction is that this film is inconsistent in the
  2715. way the characters are portrayed; for example:</p>
  2717. <blockquote>
  2718.  <p>After the death of his son, Theodred, Theoden says: 'Alas that these
  2719.  evil days shall be mine. The young perish and the old linger. That I
  2720.  should live to see the last days of my house.' This isn't from the book,
  2721.  but it's the kind of thing that a chap like Theoden might be expected to
  2722.  say. But then he starts to blub and announces to the world that: 'No
  2723.  parent should have to bury their child.' (Note 'parent' and 'child'
  2724.  rather than 'father' and 'son'.) It is hard to imagine any sentiment
  2725.  less likely to come from the lips of a king in an honour-based warrior
  2726.  culture. The bathos comes, not just from the fact that we've shifted from
  2727.  'high' language to a vernacular, but because we've shifted from heroic
  2728.  sentiments to soap-operatic ones. You can't be expressing Tolkienesque
  2729.  ideas in Tolkienesque language in one sentence and Hollywood banalities
  2730.  the next and expect it to make sense.</p>
  2731. </blockquote>
  2733. <p>This is something that wasn't really present in the first film, and I
  2734. have to agree it is jarring. Rilstone also comments on how the morality
  2735. Tolkien tried to portray in the books is completely mangled in this film;
  2736. for example:</p>
  2738. <blockquote>
  2739.  <p>Jackson's simplifications are not limited to recasting complex speeches
  2740.  as Hollywood banalities. He simplifies the whole moral structure of the
  2741.  book, continuously re-casting it in terms of a two sided battle between
  2742.  'good' and 'evil'. Characters who Tolkien paints in darker or lighter
  2743.  shades of grey become, for Jackson, pure black or pure white. (Remind me
  2744.  to write an article one of these days on the Significance of the Colour
  2745.  Grey in Middle-earth: Grey Elves, Grey Havens, Grey Pilgrim, etc.)</p>
  2746. </blockquote>
  2748. <p>Another example is something I hadn't spotted in the interaction between
  2749. Frodo, Sam, and Gollum in the film:</p>
  2751. <blockquote>
  2752.  <p>Even the Frodo-Gollum-Sam triad is simplified, although it must be said
  2753.  with more reason and more dramatic success. In the book, Gollum talks with
  2754.  two voices. Sam imagines that there are two sides to him, 'Slinker' and
  2755.  'Stinker.' Smeagal-Gollum talks to himself in extended soliloquies, the
  2756.  dominance of the 'good' side often represented by a light in his eyes.
  2757.  Jackson extends this dual personality to the point where Smeagal (the
  2758.  good side) can consciously think of Gollum (the bad side) as 'he', and
  2759.  tell it to 'leave us and never come back'. In the book, Sam's cruelty to
  2760.  Gollum is treated consistently as a blemish on his character to be
  2761.  contrasted un-favourably with Frodo's kindness to him. Here, Smeagal
  2762.  is explicitly aware that Sam hates him because he sees--and Frodo does
  2763.  not--that Gollum intends to betray them. Sam, in fact, is in the right;
  2764.  Frodo's kindness really is a weakness. Frodo sees that Sam is cruel to
  2765.  Gollum and upbraids him for it; and in a magnificent example of
  2766.  Hollywoodisation, says that he pities Gollum 'because I have to believe
  2767.  that there is a way back.'</p>
  2769.  <p>This, I am afraid, will annihilate a psychological crux of the text.
  2770.  In Tolkien's story, Frodo's mercy to Gollum brings the good, Smeagal
  2771.  side to the fore. At one moment, Smeagal is on the point of repenting and
  2772.  becoming a wholly regenerate character, but Sam accuses him of 'sneaking'
  2773.  and destroys the moment. According to Tolkien this is the most poignant
  2774.  moment in the whole epic. But it is only because Gollum remains evil and
  2775.  seizes the Ring that the world is saved; the evil Gollum does what the
  2776.  good Frodo cannot do. In the long run Sam's cruelty rules the fate of
  2777.  many just as surely as Bilbo's mercy. (This is, presumably, the kind of
  2778.  thing which Phillip Pullman has in mind when he calls the book morally
  2779.  simplistic.) It will be interesting to see if any of this survives
  2780.  Jackson's Manichean reworking of the text.</p>
  2781. </blockquote>
  2783. <p>As we'll see when we get to Rilstone's review of the third film, below,
  2784. it doesn't.</p>
  2786. <p>Rilstone also notes some of the same things that griped me the most,
  2787. such as Aragorn's "death" (but, as Rilstone notes, "he's only <em>mostly</em>
  2788. dead", thus scoring serious points as a <em>Princess Bride</em> fan as well
  2789. as a good judge of movies), and the complete change in the way the
  2790. subplot between Arwen and Aragorn is handled.</p>
  2792. <p>The only point that I can see where Rilstone goes wrong in this review
  2793. is that he is confused about how Jackson, who knows Middle-earth lore so
  2794. well, can make so many bad decisions in making the film. My answer, of
  2795. course, is that while he may know the lore, he doesn't <em>understand</em>
  2796. Middle-earth.</p>
  2798. <p>Rilstone's review of Return of the King is even more negative than his
  2799. review of the Two Towers:</p>
  2801. <blockquote>
  2802.  <p>The end result is a movie which is uneven in tone, at crossed purposes
  2803.  with itself. Neither a successful adaptation of Lord of the Rings, nor a
  2804.  stand-alone fantasy movie.</p>
  2805. </blockquote>
  2807. <p>As I noted above, I was willing to give Jackson credit for making a
  2808. reasonable stand-alone fantasy movie, even if it wasn't a good adaptation
  2809. of Tolkien's Middle-earth. But I have to admit that Rilstone has found
  2810. flaws that I didn't spot. Some of them are the same ones he spotted in the
  2811. second movie:</p>
  2813. <blockquote>
  2814.  <p>It is pretty clear that Jackson the cinematographer wanted to make a
  2815.  movie where people spoke in modern English. But Jackson the Tolkien fan
  2816.  snuck into his office at night, scribbled lines from the book into the
  2817.  script, and hoped no-one spotted it.</p>
  2819.  <p>I think that the cinematographer resents the Tolkien geek's
  2820.  interventions, and starts retaliating. Often, when Tolkien-Fan-Jackson
  2821.  puts one of his "favourite" scenes into the scripts, Movie-Maker-Jackson
  2822.  deliberately spoils them, by adding a weak joke or making the characters
  2823.  appear more cynical and less noble than they do in the book.</p>
  2824. </blockquote>
  2826. <p>Also, Rilstone spots something I hadn't about the climactic scene at
  2827. Mount Doom:</p>
  2829. <blockquote>
  2830.  <p>I think that Cinematic Peter intended that Frodo really would fall
  2831.  over with Gollum: that Frodo would die in the closing minutes of the
  2832.  film. The only way for Frodo to destroy his shadow-self and evil
  2833.  reflection is to drag him down into the abyss with him.</p>
  2835.  <p>This ending was set up in Two Towers. Galadriel has told Elrond that
  2836.  "The Quest will claim Frodo. I have foreseen it. I know it to be true.
  2837.  It is your destiny". (Sorry, wrong movie again.) The scenes outside the
  2838.  Black Gate with everyone shouting "Frodo" and looking sad seem to have
  2839.  been filmed with this ending in mind. It would have been different in
  2840.  content from the book, but rather faithful thematically: Frodo sacrifices
  2841.  himself to save the Shire; one person gives something up so someone else
  2842.  can enjoy it. I think cinematic Jackson would have liked to end the movie
  2843.  with Frodo disappearing into the lava and the Dark Tower collapsing.
  2844.  (He could have thrust out his arms as he fell, thus conforming to another
  2845.  important cinematic rule: at least one character has to be Jesus.)</p>
  2847.  <p>But of course, Tolkien geek Jackson was aghast at the suggestion that
  2848.  someone might want to Change The Plot so radically, so Jackson has to
  2849.  splice in a terribly corny Flash Gordon get out clause in which Frodo
  2850.  grabs the edge of the cliff and is left hanging on by his fingers, and
  2851.  then does another love scene with Sam.</p>
  2853.  <p>Having had his ending messed up, Jackson now can't work out how to get
  2854.  out of the film.</p>
  2855. </blockquote>
  2857. <p>This may or may not be valid speculation about Jackson's actual mental
  2858. process here, but it certainly explains why Jackson said that everything
  2859. after the Mount Doom scene was "epilogue" to him, and why he does such a
  2860. bad job at the denouement of the story, in contrast to Tolkien, who was
  2861. careful to tell the <em>whole</em> story, not end it at the climactic moment.
  2862. (For example, the Scouring of the Shire, which Tolkien said several times
  2863. was an integral part of the story, is completely absent from the film;
  2864. but a detailed discussion of that will take yet another post, which I may
  2865. end up writing at some point. You have been warned.) Rilstone says that
  2866. the film has six endings, none of which are good ones, instead of the one
  2867. good ending it should have had, and it's hard to argue with him when he
  2868. presents the details.</p>
  2870. <p>On consideration, I have to agree with Rilstone's overall conclusion:
  2871. the first film was reasonably good as a fantasy film, but the second and
  2872. third don't even measure up to that standard. In the first film, Jackson
  2873. was able to keep a balance between making the movie recognizably about
  2874. Middle-Earth and making it a good fantasy movie in its own right. In the
  2875. second and third films, that balance is no longer there. Fortunately,
  2876. there are always the books.</p>
  2878. <h1>Postscript</h1>
  2880. <p>Rilstone also has a good
  2881. <a href="">post</a>
  2882. reviewing The Children of Hurin and discussing The Hobbit (the book, not
  2883. the movie). It's worth a read.</p>
  2884. </div>
  2885. </content>
  2886. </entry>
  2888. <entry>
  2889. <title type="html">Tolkien's Ring</title>
  2890. <category term="/opinions" />
  2891. <id></id>
  2892. <updated>2013-01-09T04:06:00Z</updated>
  2893. <published>2013-01-09T04:06:00Z</published>
  2894. <link rel="alternate" type="text/html" href="" />
  2895. <content type="xhtml">
  2896. <div xmlns="">
  2897. <p>I have a confession to make: I have not yet seen The Hobbit. This
  2898. may seem strange to you if you've read my
  2899. <a href="">previous post about Tolkien</a>,
  2900. since I made it plain that I have been a Tolkien fan for a long time;
  2901. but since I also said in the Postscript that I wasn't too happy with
  2902. the Peter Jackson films of Lord of the Rings, it may not seem so
  2903. strange after all that I haven't rushed out to see The Hobbit. But
  2904. I do have a report from a friend who has seen it, and who has been a
  2905. Tolkien fan as long as I have, and based on that report, I'm not in
  2906. any hurry to see it. This post explains why.</p>
  2908. <p><a href="">An article at the Huffington Post</a>
  2909. says that many of the negative reviews of the film are based on a
  2910. lack of understanding of Tolkien's world:</p>
  2912. <blockquote>
  2913.  <p>What these critics don't know, and what Jackson most certainly does,
  2914.  is the history of The Hobbit as a text, and of Middle Earth as a
  2915.  holistic construction.</p>
  2916. </blockquote>
  2918. <p>I don't dispute the fact that many reviewers clearly don't have such
  2919. an understanding, but I <em>do</em> dispute the claim that Jackson does. Or
  2920. at least, if he does have such an understanding, he hasn't put it to
  2921. very good use in the movies.</p>
  2923. <p>It's worth noting that Christopher Tolkien doesn't think so either. As
  2924. all Tolkien fans know, he has put in decades of hard work on Middle-earth,
  2925. first by reading his father's writings in draft and giving feedback, and
  2926. then by continuing to publish his father's writing posthumously, along
  2927. with voluminous editorial commentary on the history, development, and
  2928. meaning of the work. I think it's safe to say that no living person
  2929. understands Middle-earth better than Christopher Tolkien does, and he
  2930. was not very complimentary about the films in
  2931. <a href="">this interview in Le Monde</a>,
  2932. which is currently the subject of
  2933. <a href="">a long discussion on Hacker News</a>:</p>
  2935. <blockquote>
  2936.  <p>Invited to meet Peter Jackson, the Tolkien family preferred not to.
  2937.  Why? "They eviscerated the book by making it an action movie for young
  2938.  people aged 15 to 25," Christopher says regretfully. "And it seems that
  2939.  The Hobbit will be the same kind of film."</p>
  2941.  <p>This divorce has been systematically driven by the logic of Hollywood.
  2942.  "Tolkien has become a monster, devoured by his own popularity and absorbed
  2943.  into the absurdity of our time," Christopher Tolkien observes sadly. "The
  2944.  chasm between the beauty and seriousness of the work, and what it has
  2945.  become, has overwhelmed me. The commercialization has reduced the aesthetic
  2946.  and philosophical impact of the creation to nothing. There is only one
  2947.  solution for me: to turn my head away."</p>
  2948. </blockquote>
  2950. <p>The standard Hollywood rebuttal to this is that the films have generated
  2951. a lot of interest in the books. The Le Monde article notes that sales of
  2952. the trilogy went up by a factor of 10 after the release of the first
  2953. movie. On the Hollywood view, this can't be anything but good; the
  2954. measure of success is the number of viewers, after all. Whether those
  2955. viewers actually get a proper sense of what Tolkien was trying to convey
  2956. is beside the point. It's entertainment.</p>
  2958. <p>But, as I noted in my previous post, Tolkien's story of Middle-earth is
  2959. not just a fantasy story. Tolkien explicitly said in his foreword that
  2960. his story wasn't an allegory, but he didn't mean to imply that it had
  2961. nothing to say about the real world:</p>
  2963. <blockquote>
  2964.  <p>I cordially dislike allegory in all its manifestations, and always have
  2965.  done so since I grew old and wary enough to detect its presence. I much
  2966.  prefer history, true or feigned, with its varied applicability to the
  2967.  thought and experience of readers. I think that many confuse 'applicability'
  2968.  with 'allegory', but the one resides in the freedom of the reader, and the
  2969.  other in the purposed domination of the author.</p>
  2970. </blockquote>
  2972. <p>This passage, of course, has received much commentary. But I realized
  2973. recently that there is a clue in it, not just to the general way in which
  2974. Tolkien wanted readers to take his work, but specifically about its central
  2975. plot element and metaphor: the Ring. And the treatment of the Ring also
  2976. illustrates a key failing of the films: they give no sense of what the Ring
  2977. really stands for, and what lessons it has for the real world, not just
  2978. Middle-earth. I touched on this briefly in my previous post about Tolkien,
  2979. but it deserves a longer exposition.</p>
  2981. <p>In the films, the Ring is a magical object that's evil and needs to be
  2982. destroyed; that's pretty much all there is to it. It is portrayed as having
  2983. an attraction about it, and as changing the mental processes of those that
  2984. possess it; but the portrayal is standard Hollywood fare, with no hint of
  2985. any deeper meaning than "it's evil". In particular, no hint is given of
  2986. <em>why</em> any of the characters would <em>want</em> to wield the Ring, other than
  2987. "yes, it's evil, but it's powerful, too".</p>
  2989. <p>This is a great pity, because Tolkien gives us a lot more. He does not just
  2990. tell us that the Ring is evil; he also tells us <em>why</em>. What's more, he
  2991. doesn't tell us himself; he lets the characters themselves, the ones who are
  2992. tempted by the Ring, tell us. The first such temptation we see is that of
  2993. Gandalf:</p>
  2995. <blockquote>
  2996.  <p>'You are wise and powerful. Will you not take the Ring?'</p>
  2998.  <p>'No!' cried Gandalf, springing to his feet. 'With that power I should have
  2999.  power too great and terrible. And over me the Ring would gain a power still
  3000.  greater and more deadly.' His eyes flashed and his face was lit as by a fire
  3001.  within. 'Do not tempt me! For I do not wish to become like the Dark Lord
  3002.  himself. Yet the way of the Ring to my heart is by pity, pity for weakness
  3003.  and the desire of strength to do good. Do not tempt me! I dare not take it,
  3004.  not even to keep it safe, unused. The wish to wield it would be too great
  3005.  for my strength. I shall have such need of it. Great perils lie before me.'</p>
  3006. </blockquote>
  3008. <p>Note that Gandalf would desire the ring for strength to do <em>good</em>. If the
  3009. Ring is evil, how can this be? Clearly there is more going on here than
  3010. just "it's evil".</p>
  3012. <p>Another temptation that is drawn in greater detail than Gandalf's is that
  3013. of Galadriel. Her description of what would happen if Frodo gave her the
  3014. Ring, as he has offered to do, is one of the best passages in the whole
  3015. epic:</p>
  3017. <blockquote>
  3018.  <p>'In place of the Dark Lord you will set up a Queen. And I shall not be
  3019.  dark, but beautiful and terrible as the Morning and the Night! Fair as the
  3020.  Sea and the Sun and the Snow upon the Mountain! Dreadful as the Storm and
  3021.  the Lightning! Stronger than the foundations of the earth. All shall love
  3022.  me and despair!'</p>
  3023. </blockquote>
  3025. <p>To be fair, the key lines here are included in the movie as well; but the
  3026. scene is portrayed very differently. The movie's imagery is again standard
  3027. Hollywood fare for "being tempted by something evil", with the usual dark
  3028. clouds, lightning and thunder, and ominous background music. The scene in
  3029. the book is nothing of the sort; the description is almost austere in its
  3030. simplicity:</p>
  3032. <blockquote>
  3033.  <p>She lifted up her hand and from the ring that she wore there issued a
  3034.  great light that illumined her alone and left all else dark. She stood
  3035.  before Frodo seeming now tall beyond measurement, and beautiful beyond
  3036.  enduring, terrible and worshipful. Then she let her hand fall, and the
  3037.  light faded, and suddenly she laughed again, and lo! she was shrunken:
  3038.  a slender elf-woman, clad in simple white, whose gentle voice was soft
  3039.  and sad.</p>
  3041.  <p>'I pass the test,' she said. 'I will diminish, and go into the West, and
  3042.  remain Galadriel.'</p>
  3043. </blockquote>
  3045. <p>Note that Galadriel is <em>sad</em> that she cannot take the Ring. Once again, if
  3046. the Ring is evil, why should she be sad? You would expect her to be relieved
  3047. that she was able to pass the test, to resist the temptation and remain good
  3048. (and in the movie, that <em>is</em> how she reacts). What's going on here?</p>
  3050. <p>We get a further clue from Sam, just before the scene ends:</p>
  3052. <blockquote>
  3053.  <p>'But if you'll pardon my speaking out, I think my master was right. I wish
  3054.  you'd take his Ring. You'd put things to rights. You'd stop them digging up
  3055.  the gaffer and turning him adrift. You'd make some folk pay for their dirty
  3056.  work.'</p>
  3058.  <p>'I would,' she said. 'That is how it would begin. But it would not stop
  3059.  with that, alas!'</p>
  3060. </blockquote>
  3062. <p>In other words, a person who uses the Ring can start out by doing good. So
  3063. whatever power the Ring gives, it can't just be a stereotypical "evil"
  3064. power.</p>
  3066. <p>The obvious next step is to view the Ring as a metaphor for the old saying,
  3067. "power corrupts". People start out using power to do good things, but
  3068. gradually they become used to it, and start using it for questionable things,
  3069. and finally end up using it for outright evil things. A number of commentators
  3070. have taken this view. In fact, Peter Jackson himself appears to hold it,
  3071. according to
  3072. <a href=";topic=&amp;topic_set=">an article in Wired</a>:</p>
  3074. <blockquote>
  3075.  <p>"One of Tolkien's great themes is that power itself always corrupts,"
  3076.  explains Peter Jackson. "Ultimately there can never really be any good
  3077.  power."</p>
  3078. </blockquote>
  3080. <p>But if we take Tolkien's portrayal of Middle-earth seriously, this view
  3081. cannot be right, because Tolkien shows us plenty of uses of power that do
  3082. <em>not</em> lead to corruption. Gandalf has power, and uses it: he fights off
  3083. Wargs, and later the Balrog, and when he returns as Gandalf the White he
  3084. is invulnerable to weapons and able to fight off Ringwraiths. None of
  3085. this corrupts him; he ends up returning over the Sea to the Undying Lands
  3086. from which he came. Galadriel also has power: she has her own ring of
  3087. power, Nenya, and uses it to defend Lorien against Sauron and his minions.
  3088. And Aragorn <em>gains</em> power throughout the epic; he takes over the
  3089. leadership of the company after Gandalf's fall, he is recognized by the
  3090. Riders of Rohan as a leader in battle, and he ends up becoming King. None
  3091. of this corrupts him; as the Appendices tell us, he lives on for 120
  3092. years as a wise and just King, and ends up dying in peace.</p>
  3094. <p>What is the difference between all of these exercises of power, and the
  3095. power given by the Ring? As I noted in my previous post, the difference is
  3096. between power wielded justly, rightly, and power wielded unjustly, wrongly.
  3097. But how do we tell which uses of power are just and right? This is where
  3098. critics of Middle-earth often go astray. They claim that Tolkien cheated,
  3099. as it were; he declared by fiat, as the author, that certain people in
  3100. Middle-earth had a "right" to wield power, just because. Aragorn is the
  3101. rightful King because he is descended from Elendil; Gandalf has the right
  3102. to use power because he was given it by the Valar; Galadriel has the right
  3103. to wield Nenya because she is the last survivor in Middle-earth of one of
  3104. the noble houses of the elves. In the real world, we have nothing like
  3105. this; there is no magical method of knowing who has the right to wield
  3106. power, and so the only conclusion we can draw is that <em>no one</em> in the real
  3107. world can "rightfully" wield power.</p>
  3109. <p>It is true that there is nothing in the real world corresponding to the
  3110. notion of Aragorn as the "rightful King", for example, as it is portrayed
  3111. in Middle-earth. But Tolkien makes it clear that there is more to it than
  3112. just having the right descent, or the right blessing from the Valar. You
  3113. have to <em>use</em> the power you have in the right way. For every person who
  3114. wields power rightly, Tolkien shows us another person, with a similar grant
  3115. of power, who wields it wrongly, and goes astray as a result. Aragorn is
  3116. the rightful King, and lives up to that; Denethor is the rightful Steward,
  3117. but does <em>not</em> live up to it. Saruman starts out with the same blessing
  3118. from the Valar as Gandalf; indeed, he starts out with more, as Gandalf
  3119. himself recognizes when he says that as Gandalf the White he is "Saruman
  3120. as he should have been". But Saruman, unlike Gandalf, does not use what he
  3121. is given rightly. Even Galadriel and Elrond, who wield the elven rings as
  3122. best they can, are contrasted with the elven smiths who <em>made</em> the rings
  3123. because they were deceived by Sauron, and so tied their fates, and the
  3124. fate of the elves themselves, to the fate of the One.</p>
  3126. <p>But what, exactly, <em>is</em> it about the One that makes the difference? This
  3127. is where the clue I referred to earlier, in Tolkien's statement in the
  3128. foreword, comes in. He contrasts the "freedom of the reader" with the
  3129. "purposed domination of the author". And "purposed domination" is exactly
  3130. what the Ring is about. But in accordance with his own professed preference,
  3131. Tolkien does not shove this in our faces; he gives us hints and leaves us
  3132. to figure it out for ourselves.</p>
  3134. <p>It is notable that we are never shown explicitly exactly what the Ring
  3135. <em>does</em>. We are never shown anyone actually using the Ring for anything
  3136. except to become invisible. We are told that it would be very bad if
  3137. Sauron regained the Ring, but we are not given any details about what he
  3138. would do with it, except that the people of Middle-earth would become
  3139. "slaves", as the orcs already are. But we are given hints; for example,
  3140. when Frodo asks Galadriel:</p>
  3142. <blockquote>
  3143.  <p>'I am permitted to wear the One Ring: why cannot <em>I</em> see all the others
  3144.  and know the thoughts of those that wear them?'</p>
  3146.  <p>'You have not tried,' she said. 'Only thrice have you set the Ring upon
  3147.  your finger since you knew what you possessed. Do not try! It would
  3148.  destroy you. Did not Gandalf tell you that the rings give power according
  3149.  to the measure of each possessor? Before you could use that power you
  3150.  would need to become far stronger, and to train your will to the domination
  3151.  of others.'</p>
  3152. </blockquote>
  3154. <p>Of course we have already had a hint of the Ring's power in the verse
  3155. which Gandalf recites to Frodo, which includes the phrase "One Ring to
  3156. rule them all". As the verse is explained, by Gandalf and later by Elrond
  3157. at the Council, Sauron made the One Ring to rule over all the other Rings
  3158. of Power, so that their wearers would be controlled by him. But again,
  3159. we must not make the mistake of thinking of this as a stereotypical "evil"
  3160. power. As Elrond says, "Nothing is evil in the beginning. Even Sauron was
  3161. not so." Tolkien amplified this in a
  3162. <a href="">letter</a>:</p>
  3164. <blockquote>
  3165.  <p>[Sauron] was not indeed wholly evil, not unless all 'reformers' who
  3166.  want to hurry up with 'reconstruction' and 'reorganization' are wholly
  3167.  evil, even before pride and the lust to exert their will eat them up.</p>
  3168. </blockquote>
  3170. <p>In other words, even <em>Sauron</em> originally intended to use the Ring to do
  3171. "good", as he saw it. True, he saw "good" as a Middle-earth with himself
  3172. at the top, ruling over all as his master, Morgoth, had wanted to do. But
  3173. his purpose was not, originally, to rule over all just to "do evil". His
  3174. purpose was, essentially, the purpose that Saruman tried to persuade
  3175. Gandalf to follow: "to order all things as we will, for that good which
  3176. only the Wise can see." And the Ring was a tool to make it easier for
  3177. Sauron to achieve that purpose.</p>
  3179. <p>Let's try to put these clues together. Ordinarily, when you want someone
  3180. else to do something, you have to convince them; you have to give them a
  3181. <em>reason</em> to want to do it. But every person who has ever had an idea for
  3182. making the world better has felt a desire to skip that step. Convincing
  3183. people of things, particularly if they are things that require drastic
  3184. change, is hard work, and takes a long time, and often you fail anyway.
  3185. You have to spend all this time explaining to people why your idea is such
  3186. a good idea, over and over and over again, and many of them don't even
  3187. appreciate all the thought you've put into it and all the effort you've
  3188. made to consider everyone's needs and everyone's point of view. And all
  3189. the time, whatever problem you are trying to solve isn't getting solved.
  3190. Wouldn't it be nice if you could cut out all that fuss, and just <em>make</em>
  3191. people do what you want?</p>
  3193. <p><em>That</em> is what the Ring does. It lets you just <em>make</em> others do what you
  3194. want them to do. Of course you have to be strong enough to wield it, and
  3195. you have to "train your will to the domination of others". But given that,
  3196. you can use the Ring to cut out all the bother of convincing people, and
  3197. just command them instead. And at the start, the things you command them
  3198. to do might well be good things. Certainly if you are coming into a place
  3199. like the Shire under Saruman, which has been systematically plundered,
  3200. there are a lot of easy improvements to be made, and they may well get
  3201. made faster if the chain of command is very short, as it will be with the
  3202. Ring. (In the real world, of course, there are plenty of examples of
  3203. countries which improved for a while after a dictator was put in charge.)
  3204. But the improvements come at a terrible price.</p>
  3206. <p>In Middle-earth, we are shown the price in several ways, in addition to
  3207. the indirect hints we get from Gandalf, Elrond, Galadriel, and others in
  3208. passages like those quoted above. First, of course, we see the effect
  3209. that long possession of the Ring has had on Gollum. He never used it for
  3210. anything except becoming invisible to avoid relatives or catch fish; but
  3211. still it has made him into something quite unlike the hobbit-like
  3212. creature he originally was. Second, we see the Ringwraiths, who have
  3213. been commented on enough that I don't think I need to elaborate on them
  3214. here. And of course there is the price paid by Sauron himself when the
  3215. Ring is finally destroyed. But perhaps the most poignant price is paid
  3216. by the elves, who lose the power and benefits of the Three Rings when
  3217. the One is destroyed, and must then leave Middle-earth or "dwindle to
  3218. a rustic folk of dell and cave, slowly to forget and to be forgotten".</p>
  3220. <p>This last price is also the most relevant if we are trying to find a
  3221. parallel in the real world, because it shows that the making of the
  3222. Rings of Power in the first place was, in the long run, a mistake. It
  3223. gave the elves power, but it also made them vulnerable in a way they
  3224. would never have been had the Three never existed. This point is made
  3225. during the Council of Elrond:</p>
  3227. <blockquote>
  3228.  <p>'Those who made them [the Three Rings] did not desire strength or
  3229.  domination or hoarded wealth, but understanding, making, and healing,
  3230.  to preserve all things unstained. These things the Elves of Middle-earth
  3231.  have in some measure gained, though with sorrow. But all that has been
  3232.  wrought by those who wield the Three will turn to their undoing, and
  3233.  their minds and hearts will become revealed to Sauron, if he regains the
  3234.  One. It would be better if the Three had never been. That is his
  3235.  purpose.'</p>
  3237.  <p>'But what then would happen, if the Ruling Ring were destroyed, as
  3238.  you counsel?' asked Gloin.</p>
  3240.  <p>'We know not for certain,' answered Elrond sadly. 'Some hope that the
  3241.  Three Rings, which Sauron has never touched, would then become free,
  3242.  and their rulers might heal the hurts of the world that he has wrought.
  3243.  But maybe when the One has gone, the Three will fail, and many fair
  3244.  things will fade and be forgotten. That is my belief.'</p>
  3245. </blockquote>
  3247. <p>And Elrond turns out to be right. The specific way this plays out in
  3248. the story is tailored to Middle-earth, but the general point is
  3249. basically the one I made when I posted about
  3250. <a href="">my favorite Heinlein quote</a>:
  3251. any "shortcut" in the use of power turns out to be a net loss, not a
  3252. net gain. That is why the Ring is evil: there is <em>no</em> way to use it
  3253. that is <em>not</em> a shortcut. You can't use it to convince, only to command.</p>
  3255. <p>In the real world, of course, there is no One Ring; but there are lots
  3256. of ways to shortcut the use of power, and Tolkien was not a fan of any
  3257. of them. Certainly his attitude towards using power to command was
  3258. clear. In a
  3259. <a href="">letter to his son</a>,
  3260. he wrote:</p>
  3262. <blockquote>
  3263.  <p>The most improper job of any man, even saints (who at any rate were
  3264.  at least unwilling to take it on), is bossing other men. Not one in a
  3265.  million is fit for it, and least of all those who seek the
  3266.  opportunity...</p>
  3267. </blockquote>
  3269. <p>But Tolkien does not just show us how not to use power. He also, as I
  3270. have already noted, shows us what the just use of power looks like.
  3271. Critics often miss this point because they get hung up on the fact,
  3272. which I mentioned earlier, that in the real world there is no such
  3273. thing as "rightful" rule. But that does not mean that nobody in the
  3274. real world has power. And as I noted, the real test is how those with
  3275. power use it. In Tolkien's story, the common factor among all of those
  3276. who wield power rightly is that they don't use it to command. The
  3277. elves are the purest example of this; they are not only unwilling to
  3278. give orders, they are even reluctant to give advice, as Gildor tells
  3279. Frodo:</p>
  3281. <blockquote>
  3282.  <p>'Elves seldom give unguarded advice, for advice is a dangerous thing,
  3283.  even from the wise to the wise, and all courses may run ill.'</p>
  3284. </blockquote>
  3286. <p>Gandalf and Aragorn are quite willing to answer questions and give
  3287. advice, but they always make it clear that the choice lies with the one
  3288. asking for answers or advice. And even though they are obviously wise
  3289. and powerful, and recognized to be so by everyone, they almost never
  3290. give orders, and the ones they do give are obviously necessary to deal
  3291. with an immediate problem, such as Gandalf directing the company when the
  3292. Wargs attack, or Aragorn leading them out of Moria after Gandalf's fall.
  3293. At every critical point where there is a weighty decision to be made
  3294. that involves others, Gandalf and Aragorn never direct anyone; they
  3295. always leave the choice to them. Gandalf does not force anyone to go
  3296. to Moria; he only asks who will follow him if he leads them there.
  3297. Aragorn makes a suggestion about who should accompany Frodo to Mordor
  3298. if Frodo insists on going, but it is only a suggestion, and it gets an
  3299. immediate protest from the hobbits and Legolas (and of course it is
  3300. soon overtaken by events anyway). He insists that anyone who accompanies
  3301. him on the Paths of the Dead must do so willingly. And he does not force
  3302. anyone to go all the way to the Gates of Mordor for the final battle;
  3303. he gives those who are wavering an alternate task to choose if they
  3304. wish.</p>
  3306. <p>What Aragorn and Gandalf, and the others who wield power rightly, <em>do</em>
  3307. use their power for is to <em>make it possible</em> for others to make
  3308. important choices, and to give them the information they need to make
  3309. them. Gandalf, Aragorn, Elrond, and all the other enemies of Sauron
  3310. use whatever power they have to resist him, but nobody is ever <em>forced</em>
  3311. to resist; it is always their free choice. (I'll have quite a bit more
  3312. to say about <em>why</em> free choice is so important in Tolkien's conception
  3313. of Middle-earth below.) The hobbits are not commanded to take the Ring
  3314. to Rivendell; they freely choose to. But they are only able to make that
  3315. choice because of the efforts of Gandalf and Aragorn. So it is with the
  3316. other key choice points in the story: the choice is only made possible
  3317. because those who wield power in resisting Sauron, do so to enable the
  3318. free choices of others, not to command them. This happens throughout
  3319. the story, but the two pivotal points where it happens are the Council
  3320. of Elrond and the Last Debate.</p>
  3322. <p>At the Council of Elrond, almost all of the time is spent in sharing
  3323. information. The debate about what to do with the Ring, after all the
  3324. information is on the table, is brief. What's more, it is a real debate;
  3325. Elrond and Gandalf make clear that they believe the Ring must be
  3326. destroyed, but their position is not accepted blindly, without argument.
  3327. In fact, as the scene is presented, it seems quite possible that no
  3328. decision at all would have been made if Bilbo had not butted in:</p>
  3330. <blockquote>
  3331.  <p>'But tell me: what do you mean by <em>they</em>?'</p>
  3333.  <p>'The messengers who are sent with the Ring.'</p>
  3335.  <p>'Exactly. And who are they to be? That seems to me what this Council
  3336.  has to decide, and all that it has to decide. Elves may thrive on speech
  3337.  alone, and Dwarves endure great weariness; but I am only an old hobbit,
  3338.  and I miss my meal at noon. Can't you think of some names now?'</p>
  3339. </blockquote>
  3341. <p>At the Last Debate, Gandalf once again makes clear what he believes
  3342. must be done; but he does not order anyone to follow the strategy he
  3343. gives. Nor does Aragorn; he too makes clear that he will follow Gandalf's
  3344. strategy, but in the next breath he says:</p>
  3346. <blockquote>
  3347.  <p>'Nonetheless I do not yet claim to command any man. Let others choose
  3348.  as they will.'</p>
  3349. </blockquote>
  3351. <p>The other Captains choose to follow him, but even Imrahil, who takes
  3352. Aragorn's wish as a command, follows that up immediately with his own
  3353. concern, which none of the other Captains have thought of, and which is
  3354. immediately recognized as important by the others:</p>
  3356. <blockquote>
  3357.  <p>'Now, it may be that we shall triumph, and while there is any hope of
  3358.  this, Gondor must be protected. I would not have us return with victory
  3359.  to a City in ruins and a land ravaged behind us. And yet we learn from
  3360.  the Rohirrim that there is an army still unfought upon our Northern
  3361.  flank.'</p>
  3363.  <p>'That is true,' said Gandalf.</p>
  3364. </blockquote>
  3366. <p>And Imrahil's suggestion is forthwith incorporated into the strategy.</p>
  3368. <p>I have gone into quite a bit of detail about this in order to make my
  3369. two central points perfectly clear. First, the simplistic views of the
  3370. Ring and power, that the Ring is a simple "evil" item and that power
  3371. always corrupts, cannot be right; Tolkien's actual portrayal of these
  3372. central issues is much more complex than that. And second, <em>none</em> of
  3373. that complexity is there in the movies. And since these central themes,
  3374. and the complexity surrounding them, are a crucial part of "Middle-earth
  3375. as a holistic construction", the fact that they are not even so much as
  3376. hinted at in the movies tells me that, as I said at the start of this
  3377. post, Peter Jackson either doesn't understand Middle-earth, or at any
  3378. rate has failed to put such an understanding into his movies.</p>
  3380. <h1>Tolkien's Characters</h1>
  3382. <p>It is important to emphasize, once again, that Tolkien did not
  3383. put these ideas into his writings overtly. No one in Middle-earth
  3384. articulates just what I have articulated above about the Ring and the
  3385. use of power, and it is at least arguable that no one in Middle-earth
  3386. even conceptualizes things that way, at least not fully. Tolkien's
  3387. beliefs and convictions about how power should and should not be used
  3388. were "built in" to Middle-earth, as part of its "internal logic", as
  3389. Tolkien called it in his essay "On Fairy-Stories". As such, these ideas
  3390. would appear to those within Middle-earth, not as philosophical ideas,
  3391. but as something like "laws of nature".</p>
  3393. <p>And indeed, in the books, when we see the characters most likely to
  3394. understand the internal logic of their world, such as Gandalf, Elrond,
  3395. and Galadriel, explain it, they do so the way we would explain simple
  3396. facts about the world, like gravity. To people like us, who live in a
  3397. world whose laws are different, it may seem as though they are uttering
  3398. moral platitudes; but remember that, in the books, the understanding
  3399. these characters have comes from direct experience, not from reading
  3400. philosophical tomes. As Elrond tells Frodo:</p>
  3402. <blockquote>
  3403.  <p>'[M]y memory reaches back even to the Elder Days. Earendil was my
  3404.  sire, who was born in Gondolin before its fall; and my mother was
  3405.  Elwing, daughter of Dior, son of Luthien of Doriath. I have seen three
  3406.  ages in the West of the world, and many defeats, and many fruitless
  3407.  victories.'</p>
  3408. </blockquote>
  3410. <p>And Elrond's experience, great as it is, is still much less than that
  3411. of Galadriel. (Gandalf is something of a special case; he has been in
  3412. Middle-earth for a much shorter time, but he also is of the same order
  3413. as the Valar, so he can, at least in principle, draw on experiences
  3414. that not even Galadriel can. But a detailed comparison of these cases
  3415. would take us much too far afield.) Even Aragorn, who is only a mere
  3416. (by comparison) 88 years old when The Lord of the Rings begins, has
  3417. had the benefit of being raised by Elrond in Rivendell, and traveling
  3418. for many, many years all over Middle-earth learning about its lands
  3419. and peoples. He has also been friends with Gandalf during most of that
  3420. time, and has learned much from the wizard.</p>
  3422. <p>This brings up another key failing of the movies: none of these
  3423. characters display any of the maturity and understanding that they
  3424. have in the books. I mentioned this in my previous post, but again,
  3425. it deserves more discussion, because the understanding that the key
  3426. "wise" characters have is another crucial element in "Middle-earth as
  3427. a holistic construction", and its complete absence from the movies is
  3428. another illustration of Jackson's failure to display a real understanding
  3429. of Middle-earth.</p>
  3431. <p>The way in which these characters react to the temptation of the Ring,
  3432. already discussed, is only one aspect of the understanding that the
  3433. movies completely fail to convey. Another aspect is the way in which
  3434. the characters approach the struggle against Sauron, and how it interacts
  3435. with their personal lives. Let's start with the example I gave in my last
  3436. post, the desire of Aragorn and Arwen to wed and Elrond's reaction to it.
  3437. Just to be clear about what we're dealing with, keep in mind two key
  3438. things about the movies: Aragorn starts out <em>not</em> wanting to become King,
  3439. and Elrond lies to Arwen to try to get her to depart with the Elves
  3440. instead of waiting to marry Aragorn.</p>
  3442. <p>In the books, as many critics have commented, Tolkien gives us very
  3443. little to go on concerning the complex relationship between Elrond,
  3444. Arwen, and Aragorn. We get a hint at the merrymaking the night before
  3445. the Council of Elrond, when Frodo sees Arwen and Aragorn together; and
  3446. we get another hint when the Fellowship leaves Rivendell:</p>
  3448. <blockquote>
  3449.  <p>Aragorn sat with his head bowed to his knees; only Elrond knew fully
  3450.  what this hour meant to him.</p>
  3451. </blockquote>
  3453. <p>In Lorien we get another hint, when Frodo sees Aragorn at the foot of
  3454. Cerin Amroth recalling being there with Arwen; but there is nothing to
  3455. indicate what is in fact the truth, that the occasion he is recalling
  3456. is the plighting of their troth. Aragorn's words to Galadriel at the
  3457. Company's farewell to Lorien also mention Arwen, but this hint is likely
  3458. to be lost on the reader without the back story that is only given in the
  3459. Appendices. Another hint comes when the Rangers catch up with Aragorn
  3460. and the Riders of Rohan, and Halbarad tells Aragorn what he is carrying:</p>
  3462. <blockquote>
  3463.  <p>'It is a gift that I bring you from the Lady of Rivendell,' answered
  3464.  Halbarad. 'She wrought it in secret, and long was the making. But she
  3465.  also sends word to you: <em>The days now are short. Either our hope cometh,
  3466.  or all hopes end. Therefore I send thee what I have made for thee. Fare
  3467.  well, Elfstone!</em>'</p>
  3469.  <p>And Aragorn said: 'Now I know what you bear. Bear it still for me for
  3470.  a while!' And he turned and looked away to the North under the great
  3471.  stars, and then he fell silent and spoke no more while the night's
  3472.  journey lasted.</p>
  3473. </blockquote>
  3475. <p>And, of course, we get one more hint when Aragorn tells Eowyn:</p>
  3477. <blockquote>
  3478.  <p>'Were I to go where my heart dwells, far in the North I would now be
  3479.  wandering in the fair valley of Rivendell.'</p>
  3480. </blockquote>
  3482. <p>But these hints (and as far as I can tell the ones I have mentioned are
  3483. the <em>only</em> ones) are so thin that it is perfectly possible for the reader
  3484. to be as surprised as Frodo when he sees Arwen approaching Minas Tirith
  3485. and realizes that she is there to marry Aragorn. Only when we read The
  3486. Tale of Aragorn and Arwen in the Appendices do we finally see everything
  3487. that lies behind these hints; and it is nothing like what is portrayed
  3488. in the movies.</p>
  3490. <p>First of all, there is no hint in the movies of the gravity of the choice
  3491. Arwen has to make, which Aragorn makes clear to her in the book when they
  3492. plight their troth:</p>
  3494. <blockquote>
  3495.  <p>'Arwen said: "Dark is the Shadow, and yet my heart rejoices; for you,
  3496.  Estel, shall be among the great whose valour will destroy it."</p>
  3498.  <p>'But Aragorn answered: "Alas! I cannot foresee it, and how it may come
  3499.  to pass is hidden from me. Yet with your hope I will hope. And the Shadow
  3500.  I utterly reject. But neither, lady, is the Twilight for me; for I am
  3501.  mortal, and if you will cleave to me, Evenstar, then the Twilight you
  3502.  must also renounce."</p>
  3504.  <p>'And she stood then as still as a white tree, looking into the West,
  3505.  and at last she said: "I will cleave to you, Dunadan, and turn from the
  3506.  Twilight. Yet there lies the land of my people and the long home of all
  3507.  my kin." She loved her father dearly.'</p>
  3508. </blockquote>
  3510. <p>The Aragorn that is portrayed in the movies certainly does not start out
  3511. as the kind of man who could put everything on the line like that with
  3512. the woman he loves. By the end of the movies, true, Aragorn has grown;
  3513. and one might try to justify Jackson's treatment by observing that the
  3514. requirements of moviemaking force him to telescope a lot of character
  3515. development into a short time that, in the books, can take place over a
  3516. period of years. But that won't work, because in the books, even the
  3517. young Aragorn is not the kind of man that Aragorn in the movies is at
  3518. the start.</p>
  3520. <p>In the movie, Aragorn is unwilling to pursue the throne of Gondor
  3521. because he is afraid he will fail, because he is descended from Isildur
  3522. and Isildur failed by not destroying the Ring when he had the chance
  3523. (more on this below). In the Tale of Aragorn and Arwen, as soon as
  3524. Elrond tells the young Aragorn his lineage, Aragorn is committed to the
  3525. struggle against Sauron; and he also foresees that the climax of that
  3526. struggle might well come in his own lifetime. He tells Elrond, when the
  3527. latter has told him of the choice before his children:</p>
  3529. <blockquote>
  3530.  <p>'"But lo! Master Elrond, the years of your abiding run short at last,
  3531.  and the choice must soon be laid on your children, to part either with
  3532.  you or with Middle-earth."</p>
  3533. </blockquote>
  3535. <p>And Elrond confirms his foresight:</p>
  3537. <blockquote>
  3538.  <p>'"Truly," said Elrond. "Soon, as we account it, though many years of
  3539.  Men must still pass."'</p>
  3540. </blockquote>
  3542. <p>So Aragorn in the books simply does not undergo the kind of crisis of
  3543. faith that Aragorn in the movies does. But one might reasonably ask, why
  3544. not? He has chosen goals in life that would make any sane man stop and
  3545. think; and though he never loses hope, it is difficult to see anything
  3546. tangible for his hope to be based on. Indeed, Aragorn himself does not
  3547. see how his hope is to be fulfilled, as he admits to Arwen. Why, then,
  3548. does he keep it?</p>
  3550. <p>A hint at the answer is in Elrond's speech to Aragorn after he learns of
  3551. his betrothal to Arwen:</p>
  3553. <blockquote>
  3554.  <p>'"My son, years come when hope will fade, and beyond them little
  3555.  is clear to me. And now a shadow lies between us. Maybe, it has been
  3556.  appointed so, that by my loss the kingship of Men may be restored."'</p>
  3557. </blockquote>
  3559. <p>At many critical junctures in the story, characters like Elrond, Gandalf,
  3560. and Galadriel use language like this. Gandalf tells Frodo:</p>
  3562. <blockquote>
  3563.  <p>'I can put it no plainer than by saying that Bilbo was <em>meant</em> to find
  3564.  the Ring, and <em>not</em> by its maker. In which case you also were <em>meant</em> to
  3565.  have it. And that may be an encouraging thought.'</p>
  3566. </blockquote>
  3568. <p>When Frodo tells the Council of Elrond that he will take the Ring to
  3569. the Fire, this is Elrond's response:</p>
  3571. <blockquote>
  3572.  <p>Elrond raised his eyes and looked at him, and Frodo felt his heart
  3573.  pierced by the sudden keenness of the glance. 'If I understand aright
  3574.  all that I have heard,' he said, 'I think that this task is appointed
  3575.  for you, Frodo; and that if you do not find a way, no one will.'</p>
  3576. </blockquote>
  3578. <p>And before the Company's last night in Lorien, Galadriel tells them:</p>
  3580. <blockquote>
  3581.  <p>'Do not trouble your hearts overmuch with thought of the road tonight.
  3582.  Maybe the paths that you each shall tread are already laid before your
  3583.  feet, though you do not see them.'</p>
  3584. </blockquote>
  3586. <p>Many critics have commented on the fact that there is some sort of
  3587. "guiding power" at work behind the scenes in Middle-earth, and of course
  3588. this is not surprising given Tolkien's Catholic faith. But whatever we
  3589. might think about Tolkien's reasons, it is obvious that the "wise"
  3590. characters in the books believe that there is <em>something</em> at work that
  3591. "appoints" tasks to certain people, and to them, this something is like
  3592. a law of nature, part of the internal logic of Middle-earth, just like
  3593. the consequences of unjust use of power. The movies do not convey this
  3594. at all; and this failure means that one more key element of the
  3595. internal logic of Middle-earth is simply not there.</p>
  3597. <p>One might argue that, to modern sensibilities, the metaphysics of
  3598. Middle-earth as portrayed by Tolkien would simply be too alien to come
  3599. across in a movie. In the books, once again, the "wise" characters have
  3600. direct experience of a huge swath of history, so they have <em>seen</em> tasks
  3601. be "appointed" to people in the past, and seen how the response of the
  3602. people appointed makes a difference in how things come out. We don't
  3603. have anything like that kind of moral clarity in the real world. But it
  3604. would be one thing to try to tone down the metaphysics to make it more
  3605. believable to a modern audience; it is quite another to get rid of it
  3606. altogether and substitute a completely different set of "rules", which
  3607. is what the movies do.</p>
  3609. <p>I have already noted that Aragorn in the movies goes through a crisis
  3610. of faith, which is not there in the books. But more than that, in the
  3611. movies, he does not change his mind because of anything internal to
  3612. himself, any kind of personal growth; he changes it because Elrond
  3613. brings him the reforged sword Anduril, and he realizes that, oh, maybe
  3614. he ought to try and defeat Sauron and become King after all. In other
  3615. words, it is an external event that prompts him to change. Similarly,
  3616. Elrond's change of heart in the movies is brought about by an external
  3617. event, not by his own growth; Arwen finds out that he was lying to her
  3618. and confronts him with it. (Of course the lying itself is utterly unlike
  3619. the Elrond in the books; he is heavy of heart about the prospect of
  3620. losing his daughter, but even to think of lying to manipulate her would
  3621. be abhorrent to him. More on that below.) In general, what personal
  3622. growth the characters in the movies experience follows this pattern;
  3623. they react to external events instead of really trying to see their
  3624. choices and their lives as part of a larger whole, and choosing their
  3625. paths accordingly.</p>
  3627. <p>But the whole <em>point</em> of the faith that Tolkien's characters have in
  3628. the books is that it motivates them to make their own choices, and to
  3629. see those choices as fitting into something greater than themselves.
  3630. Aragorn's hope in the books is not just a Pollyanna belief that things
  3631. will work out in the end; it is what gives him the will and the
  3632. strength to <em>make</em> them work out, by doing his part. What's more,
  3633. this kind of hope is what makes it possible for the Council of Elrond
  3634. to even consider trying to send the Ring to Mordor to be destroyed,
  3635. which is the only way to really achieve victory:</p>
  3637. <blockquote>
  3638.  <p>'Thus we return once more to the destroying of the Ring,' said Erestor,
  3639.  'and yet we come no nearer. What strength have we for the finding of the
  3640.  Fire in which it was made? That is the path of despair. Of folly I would
  3641.  say, if the long wisdom of Elrond did not forbid me.'</p>
  3643.  <p>'Despair, or folly?' said Gandalf. 'It is not despair, for despair is
  3644.  only for those who see the end beyond all doubt. We do not. It is wisdom
  3645.  to recognize necessity, when all other courses have been weighed, though
  3646.  as folly it may appear to those who cling to false hope. Well, let folly
  3647.  be our cloak, a veil before the eyes of the Enemy! For he is very wise,
  3648.  and weighs all things to a nicety in the scales of his malice. But the
  3649.  only measure that he knows is desire, desire for power; and so he judges
  3650.  all hearts. Into his heart the thought will not enter that any will
  3651.  refuse it, that having the Ring we may seek to destroy it. If we seek
  3652.  this, we shall put him out of reckoning.'</p>
  3653. </blockquote>
  3655. <p>It is notable, once again, that this justification for the Ring-Bearer's
  3656. quest, which is a pivotal point in the story in the books, is not given
  3657. in the movies at all. In the movies the reasoning is basically, "the
  3658. Ring is evil and we can't use it, so let's destroy it". But there is no
  3659. real discussion of alternatives; the Council spends most of its time
  3660. bickering instead of debating. And we are given no reason why such a
  3661. mission would have any chance of succeeding. The key distinction Gandalf
  3662. makes in the book, between "false hope" and recognizing necessity, is
  3663. completely absent; for all we know, in the movie, Gandalf and Elrond might
  3664. just be rolling the dice, hoping that the Ring can get to Mordor somehow.</p>
  3666. <p>This illustrates that Jackson's portrayal of Middle-earth does not just
  3667. lack certain elements that are in the books: it <em>changes</em> critical
  3668. elements, and does so in a way that destroys the real meaning of
  3669. Tolkien's work. Jackson's characters are not just immature compared to
  3670. Tolkien's; they lack the moral agency that Tolkien took such pains to
  3671. give them. Tolkien's characters in the books do their best to make hard
  3672. choices in difficult situations; Jackson's characters in the movies just
  3673. react. For example, in the Council of Elrond scene in the movie, the
  3674. Ring itself controls the flow of events, by whispering to the various
  3675. Council members. There is no sense at all that the Council is trying to
  3676. <em>decide</em> what to do; they are just reacting to the Ring and each other.</p>
  3678. <p>In fact, the treatment of the Ring in general in the movies shows the
  3679. same failure to give the characters moral agency. In the books, the Ring
  3680. forces characters to make a choice: do I try to gain the Ring and use
  3681. it, or not? But in the movies, the Ring is just one more external thing
  3682. that the characters react to. Those who give in to it, such as Boromir,
  3683. are just doing what comes naturally; there is no sense that they are
  3684. faced with a difficult moral choice and choose wrong. Even those who
  3685. resist the Ring, like Gandalf and Galadriel, are not portrayed as
  3686. freely choosing to do so; they are portrayed simply as being lucky
  3687. enough not to give in.</p>
  3689. <p>Moreover, in the books, those who are tempted by the Ring and resist it
  3690. do so by <em>thinking through</em> what the consequences would be. Some, like
  3691. Gandalf and Galadriel, have evidently thought it through beforehand; in
  3692. Galadriel's case, she tells us so herself:</p>
  3694. <blockquote>
  3695.  <p>'For many long years I had pondered what I might do, should the Great
  3696.  Ring come into my hands...'</p>
  3697. </blockquote>
  3699. <p>Faramir, too, has evidently considered the issues involved; perhaps not
  3700. specifically with regard to the Ring, but considered them nonetheless:</p>
  3702. <blockquote>
  3703.  <p>'I would not take this thing, if it lay by the highway. Not were Minas
  3704.  Tirith falling in ruin and I alone could save her, so, using the weapon
  3705.  of the Dark Lord for her good and my glory. I do not wish for such
  3706.  triumphs, Frodo son of Drogo.'</p>
  3707. </blockquote>
  3709. <p>It is notable, by the way, that Faramir says this <em>before</em> he knows
  3710. what Frodo's burden actually is, and before he knows what it did to his
  3711. brother Boromir. In the movies, Faramir diverts Frodo, Sam, and Gollum
  3712. towards Minas Tirith, and only after the Nazgul attack Osgiliath (the
  3713. timing of these events is significantly changed from the books) does
  3714. he change his mind and let them go on with the quest.</p>
  3716. <p>But even Sam, who has <em>not</em> thought it through beforehand, is able to
  3717. resist the temptation of the Ring in the book by thinking it through:</p>
  3719. <blockquote>
  3720.  <p>In that hour of trial it was the love of his master that helped most
  3721.  to hold him firm; but also deep down in him lived still unconquered his
  3722.  plain hobbit-sense: he knew in the core of his heart that he was not
  3723.  large enough to bear such a burden, even if such visions were not a mere
  3724.  cheat to betray him. The one small garden of a free gardener was all his
  3725.  need and due, not a garden swollen to a realm; his own hands to use, not
  3726.  the hands of others to command.</p>
  3727. </blockquote>
  3729. <p>The movie changes the whole sequence of events at Cirith Ungol, and
  3730. again, the characters, instead of doing their best to make hard choices,
  3731. react to external forces; in this case, Frodo and Sam react to Gollum's
  3732. manipulations, Sam by offering to take the Ring, Frodo by telling Sam
  3733. to go home and leaving him (!!). Sam comes back in time to drive off
  3734. Shelob, but he thinks Frodo is dead (as in the book); but before he has
  3735. any chance to consider what to do next, the orcs come along and Sam,
  3736. overhearing them, realizes that Frodo is not dead after all, and follows
  3737. them. So once again, Sam does not really make any choice, he simply
  3738. reacts to events.</p>
  3740. <p>But the most significant change in this scene occurs after Sam has found
  3741. Frodo, and gives him back the Ring. In the movie, Sam feels the Ring's
  3742. temptation here, and it seems like Frodo only gets the Ring back because
  3743. Sam is too slow to react and Frodo, who is also drawn by the Ring, is
  3744. able to snatch it from him. Both characters are being controlled by the
  3745. Ring. And after Frodo has it, the only thing he says is, "It will destroy
  3746. you, Sam."</p>
  3748. <p>The scene in the book is nothing like that, and it's worth quoting it at
  3749. length to see why:</p>
  3751. <blockquote>
  3752.  <p>'I took it, Mr. Frodo, begging your pardon. And I've kept it safe. It's
  3753.  round my neck now, and a terrible burden it is, too.' Sam fumbled for the
  3754.  Ring and its chain. 'But I suppose you must take it back.' Now it had
  3755.  come to it, Sam felt reluctant to give up the Ring and burden his master
  3756.  with it again.</p>
  3758.  <p>'You've got it?' gasped Frodo. 'You've got it here? Sam, you're a
  3759.  marvel!' Then quickly and strangely his tone changed. 'Give it to me!'
  3760.  he cried, standing up, holding out a trembling hand. 'Give it me at once!
  3761.  You can't have it!'</p>
  3763.  <p>'All right, Mr. Frodo,' said Sam, rather startled. 'Here it is!' Slowly
  3764.  he drew the Ring out and passed the chain over his head. 'But you're in
  3765.  the land of Mordor now, sir; and when you get out, you'll see the Fiery
  3766.  Mountain and all. You'll find the Ring very dangerous now, and very hard
  3767.  to bear. If it's too hard a job, I could share it with you, maybe?'</p>
  3769.  <p>'No, no!' cried Frodo, snatching the Ring and chain from Sam's hands.
  3770.  'No you won't, you thief!' He panted, staring at Sam with eyes wide with
  3771.  fear and enmity. Then suddenly, clasping the Ring in one clenched fist,
  3772.  he stood aghast. A mist seemed to clear from his eyes, and he passed a
  3773.  hand over his aching brow. The hideous vision had seemed so real to him,
  3774.  half bemused as he was still with wound and fear. Sam had changed before
  3775.  his very eyes into an orc again, leering and pawing at his treasure, a
  3776.  foul little creature with greedy eyes and slobbering mouth. But now the
  3777.  vision had passed. There was Sam kneeling before him, his face wrung
  3778.  with pain, as if he had been stabbed in the heart; tears welled from his
  3779.  eyes.</p>
  3781.  <p>'O Sam!' cried Frodo. 'What have I said? What have I done? Forgive me!
  3782.  After all you have done. It is the horrible power of the Ring. I wish it
  3783.  had never, never been found. But don't mind me, Sam. I must carry the
  3784.  burden to the end. It can't be altered. You can't come between me and
  3785.  this doom.'</p>
  3786. </blockquote>
  3788. <p>It is worth noting, not only that the effect of the Ring is treated
  3789. very differently, but that Frodo and Sam are both <em>aware</em> of that effect
  3790. in a way that they are not in the movie. They both realize that the
  3791. power of the Ring can alter their perceptions, so that they need to
  3792. take extra precautions before believing them; and they both realize that
  3793. they <em>can</em> take such precautions, that they do not <em>have</em> to let the Ring
  3794. control them. There is nothing like this in the movies.</p>
  3796. <p>Other pivotal situations in the movies show similar differences from
  3797. their treatment in the books. We have already seen two others: Elrond's
  3798. lying to Arwen, and Aragorn's fear that he will fail as Isildur did. I
  3799. have already observed that to lie about <em>anything</em> so important, most
  3800. of all to his beloved daughter, would be abhorrent to the Elrond of the
  3801. books. But <em>why</em>? Here is how Elrond deals with the situation in the
  3802. book; I quoted the first part of this above, but now let's take a look
  3803. at all of it:</p>
  3805. <blockquote>
  3806.  <p>'"Maybe, it has been appointed so, that by my loss the kingship of Men
  3807.  may be restored. Therefore, though I love you, I say to you: Arwen
  3808.  Undomiel shall not diminish her life's grace for less cause. She shall
  3809.  not be the bride of any Man less than the King of both Gondor and Arnor.
  3810.  To me then even our victory can bring only sorrow and parting -- but to
  3811.  you hope of joy for a while. Alas, my son! I fear that to Arwen the Doom
  3812.  of Men may seem hard at the ending."</p>
  3814.  <p>'So it stood afterwards between Elrond and Aragorn, and they spoke no
  3815.  more of this matter; but Aragorn went forth again to danger and toil.'</p>
  3816. </blockquote>
  3818. <p>Elrond in the books does not even think of lying, to either Arwen or
  3819. Aragorn, because he understands that it isn't about him; all three of
  3820. them are part of something much larger, the possible restoration of the
  3821. kingdoms of Arnor and Gondor. But note well that he does not simply let
  3822. Aragorn off the hook because there's something greater involved. He
  3823. makes it clear that Aragorn will have to live up to his heritage if he
  3824. wants to win Arwen.</p>
  3826. <p>Elrond in the movies is too caught up in his own feelings to realize
  3827. that, by lying to Arwen, he is jeopardizing something much bigger than
  3828. any of them. Only when it starts affecting Arwen herself does he wise
  3829. up and start thinking of the consequences of what he is doing. Reviewers
  3830. have commented that Elrond in the movies is a much darker character
  3831. than he is in the books; but as the above shows, he is not just dark,
  3832. but small, mean, and petty, quite unlike anything that Tolkien conceived
  3833. for the Master of Rivendell.</p>
  3835. <p>Aragorn in the movies is at least not mean or petty, but he is, as
  3836. already noted, conflicted in a way that Aragorn in the books is not, and
  3837. afraid to fulfill his birthright because his ancestor, Isildur, failed
  3838. at a critical moment, and he fears that he might fail too. This is quite
  3839. in keeping with Jackson's failure to give his characters moral agency as
  3840. Tolkien does; the idea that Isildur might just have made a wrong choice,
  3841. and that Aragorn could choose differently, is not even contemplated. In
  3842. the books, the possibility that Aragorn might have inherited whatever
  3843. quality in Isildur caused him to refuse to destroy the Ring is never
  3844. mentioned; but it is clear, nonetheless, that Aragorn is aware that
  3845. Isildur made a mistake, for he tells the Council of Elrond that he helped
  3846. Gandalf to search for Gollum because "it seemed fit that Isildur's heir
  3847. should labour to repair Isildur's fault". In other words, he sees his
  3848. inheritance simply as imposing a duty on him, not as anything to fear.
  3849. And why should he fear it? By that time he has already had his chance to
  3850. give in to the temptation of the Ring (it is notable that this scene is
  3851. completely absent in the movie) and has chosen not to.</p>
  3853. <p>So just as with Jackson's treatment of the Ring, his treatment of the
  3854. characters completely fails to show any understanding of Middle-earth.
  3855. Tolkien made his characters moral agents, doing their best to make hard
  3856. choices in difficult situations. Sometimes they fail, but they never
  3857. simply react to events, as Jackson's characters do. The movies convey
  3858. no sense of the importance of free choice in Tolkien's world.</p>
  3860. <h1>Tolkien's World</h1>
  3862. <p>The above should make it abundantly clear why I am in no hurry to see
  3863. The Hobbit. But it is worth making an already long post a little longer
  3864. in order to comment on one more feature of Tolkien's world that is
  3865. absent in the movies.</p>
  3867. <p>A fairly common reaction to the basic plot of The Lord of the Rings is
  3868. that the plan which is adopted, and which ultimately leads to victory,
  3869. to send the Ring to Mordor to be destroyed, is not a plan that anyone
  3870. would expect to work in the real world. In fact, Tolkien has Denethor
  3871. voice a similar opinion to Gandalf and Faramir and Pippin:</p>
  3873. <blockquote>
  3874.  <p>'What then is your wisdom?' said Gandalf.</p>
  3876.  <p>'Enough to perceive that there are two follies to avoid. To use this
  3877.  thing is perilous. At this hour, to send it in the hands of a witless
  3878.  halfling into the land of the Enemy himself, as you have done, and this
  3879.  son of mine, that is madness.'</p>
  3880. </blockquote>
  3882. <p>(Denethor, by the way, is another character who is much diminished in
  3883. the movies compared to the books, for no real purpose that I can see.
  3884. In the books, even though he ends badly, Denethor does his best to make
  3885. hard choices, as the other characters do. In the movies he is the next
  3886. thing to a barbarian. But I've said enough on that theme here.)</p>
  3888. <p>The plan only works, according to this line of criticism, because of
  3889. the "guiding power" behind the scenes, which I referred to above, and
  3890. which Tolkien obviously meant to play a similar role to what is called
  3891. Divine Providence in Christian literature. In other words, the main
  3892. plot line is basically a <em>deus ex machina</em>. I have to admit that when
  3893. <em>I</em> first read The Lord of the Rings, I had a similar reaction. Tolkien
  3894. sets up his world so that, ultimately, right choices will be rewarded
  3895. and wrong choices will be punished. Yes, "ultimately" may be a long
  3896. while: it takes more than 4700 years for Sauron's wrong choice in
  3897. making the One Ring to be punished. But sooner or later the appropriate
  3898. consequences will occur.</p>
  3900. <p>Tolkien, as a Catholic, may have believed that the real world is like
  3901. that; but I, as an agnostic, do not. Yet I can read about Middle-earth
  3902. again and again without ever getting tired of it. Why is that? Because
  3903. the question of whether our real world has a "guiding power" behind the
  3904. scenes or not is beside the point. Tolkien's wanted to <em>create</em> an
  3905. imaginary world, with rules that he believed <em>ought</em> to be true of our
  3906. world, whether or not they actually are. And many, like me, who read
  3907. and re-read his stories of Middle-earth do so, at least in large part,
  3908. because we enjoy being in that world, even if it is only imaginary. We
  3909. enjoy reading about and imagining a world where people, not just wise
  3910. and powerful ones like Gandalf and Aragorn, but ordinary ones like
  3911. Frodo and Sam, can make free choices, and make the <em>right</em> choices,
  3912. and have those choices rewarded.</p>
  3914. <p>Not only that, but imagining such a world helps to motivate us to do
  3915. what we can to make our own real world more just, more fair than we
  3916. found it. And there is no need to disregard Tolkien and treat his work
  3917. as an allegory to do that. "Applicability" is more than enough, and
  3918. Tolkien's books are a rich source of material to apply; I have discussed
  3919. some of it here, but of course there is a lot more. Jackson's movies
  3920. give none of that, and <em>that</em> is why I seeing The Hobbit is not going
  3921. to be high on my list of things to do any time soon.</p>
  3922. </div>
  3923. </content>
  3924. </entry>
  3926. <entry>
  3927. <title type="html">The Media Industry Is Officially Lame</title>
  3928. <category term="/opinions" />
  3929. <id></id>
  3930. <updated>2013-01-03T05:52:00Z</updated>
  3931. <published>2013-01-03T05:52:00Z</published>
  3932. <link rel="alternate" type="text/html" href="" />
  3933. <content type="xhtml">
  3934. <div xmlns="">
  3935. <p>This is just a quick note to confirm that it's official: the media
  3936. industry is lame. YouTube recently deleted more that 2 billion
  3937. <a href=",news-16514.html#xtor=RSS-980">fake video views</a>
  3938. that were created by Sony, Universal, RCA, and other media companies.
  3939. This violates YouTube's terms of service, of course, which is why the
  3940. fake views were deleted. But that's a minor point compared to the big
  3941. question: how lame do you have to be to generate fake views to make
  3942. your videos appear to be more popular than they actually are? Remember
  3943. we're not talking about a few teenagers shooting home videos; we're
  3944. talking about the biggest media companies in the world.</p>
  3946. <p>But even that isn't the full extent of the lameness. Remember that these
  3947. are the same companies that complain loudly about "pirated" videos being
  3948. posted on sites like...YouTube. As I have blogged
  3949. <a href="">a number</a>
  3950. <a href="">of times</a>
  3951. <a href="">before</a>,
  3952. the reason these companies are having these problems is that they are
  3953. either unwilling or unable to change their business models to give their
  3954. customers what they actually want. If this is their attempt to try and
  3955. fix that, they need to think again.</p>
  3956. </div>
  3957. </content>
  3958. </entry>
  3960. <entry>
  3961. <title type="html">Watch Out For That First Step, It's A Lulu</title>
  3962. <category term="/opinions" />
  3963. <id></id>
  3964. <updated>2012-12-31T23:55:00Z</updated>
  3965. <published>2012-12-31T23:55:00Z</published>
  3966. <link rel="alternate" type="text/html" href="" />
  3967. <content type="xhtml">
  3968. <div xmlns="">
  3969. <p>This is my obligatory blog post about the "fiscal cliff". One can't
  3970. expect to maintain one's blogging credentials without making some
  3971. comment on an issue like this, but I have been hesitant even so
  3972. because there didn't seem to be anything worth saying that hadn't
  3973. already been said many, many times. Then I came across
  3974. <a href=";ref=general&amp;src=me&amp;pagewanted=all">this op-ed</a>
  3975. from yesterday's New York Times:</p>
  3977. <blockquote>
  3978.  <p>As the nation teeters at the edge of fiscal chaos, observers are
  3979.  reaching the conclusion that the American system of government is broken.
  3980.  But almost no one blames the culprit: our insistence on obedience to the
  3981.  Constitution, with all its archaic, idiosyncratic and downright evil
  3982.  provisions.</p>
  3983. </blockquote>
  3985. <p>My take is exactly the opposite: our government is broken because we
  3986. <em>don't</em> obey the Constitution, or indeed <em>any</em> coherent system of rules,
  3987. if we think we can get our way by breaking them. And the fiscal cliff
  3988. gives a perfect illustration of how this works and why it's a problem.</p>
  3990. <p>Of course the op-ed's lead, quoted above, did its job by impelling me to
  3991. read further, hoping to see examples of these "evil provisions" in the
  3992. Constitution. Well, it does give one:</p>
  3994. <blockquote>
  3995.  <p>Consider, for example, the assertion by the Senate minority leader
  3996.  last week that the House could not take up a plan by Senate Democrats
  3997.  to extend tax cuts on households making $250,000 or less because the
  3998.  Constitution requires that revenue measures originate in the lower
  3999.  chamber. Why should anyone care? Why should a lame-duck House, 27
  4000.  members of which were defeated for re-election, have a stranglehold
  4001.  on our economy? Why does a grotesquely malapportioned Senate get to
  4002.  decide the nation’s fate?</p>
  4003. </blockquote>
  4005. <p>Apparently this writer can't recognize political posturing when he sees
  4006. it. Congress has been more than happy to work around that Constitutional
  4007. provision in the past; in at least one
  4008. <a href="">fairly recent case</a>,
  4009. the Senate replaced the entire text of a bill that had originated in the
  4010. House so that they could address a revenue issue without violating the
  4011. Constitutional provision. Much of the rest of the op-ed is devoted
  4012. to listing similar instances of adhering to the letter of the
  4013. Constitution while violating its spirit, so the writer is clearly
  4014. aware that it can be done. Why should the problem suddenly be, in
  4015. this case, that the Constitution is getting in the way, instead of
  4016. simply that it's not being used as intended?</p>
  4018. <p>It may seem hard to believe, but this is the <em>only</em> specific example
  4019. in the entire op-ed of an "evil provision" of the Constitution that
  4020. makes our government dysfunctional. We do get another example, but it's
  4021. a hypothetical one:</p>
  4023. <blockquote>
  4024.  <p>Imagine that after careful study a government official -- say, the
  4025.  president or one of the party leaders in Congress -- reaches a considered
  4026.  judgment that a particular course of action is best for the country.
  4027.  Suddenly, someone bursts into the room with new information: a group of
  4028.  white propertied men who have been dead for two centuries, knew nothing
  4029.  of our present situation, acted illegally under existing law and thought
  4030.  it was fine to own slaves might have disagreed with this course of action.
  4031.  Is it even remotely rational that the official should change his or her
  4032.  mind because of this divination?</p>
  4033. </blockquote>
  4035. <p>A more pertinent question might be, has this ever actually happened?
  4036. Once again, the rest of the op-ed is a survey of the many different ways
  4037. in which our government has ignored the Constitution, or adhered to its
  4038. letter while violating its spirit. A reasonable conclusion from all this
  4039. would be that we are not paying <em>enough</em> attention to the Constitution;
  4040. yet the writer offers it as support for the claim that we are paying
  4041. <em>too much</em> attention to it?</p>
  4043. <p>It's worth taking a brief detour here to note this comment in the op-ed
  4044. on Supreme Court jurisprudence:</p>
  4046. <blockquote>
  4047.  <p>The fact that dissenting justices regularly, publicly and vociferously
  4048.  assert that their colleagues have ignored the Constitution -- in landmark
  4049.  cases from Miranda v. Arizona to Roe v. Wade to Romer v. Evans to Bush v.
  4050.  Gore -- should give us pause. The two main rival interpretive methods,
  4051.  "originalism" (divining the framers' intent) and "living constitutionalism"
  4052.  (reinterpreting the text in light of modern demands), cannot be reconciled.
  4053.  Some decisions have been grounded in one school of thought, and some in the
  4054.  other. Whichever your philosophy, many of the results -- by definition --
  4055.  must be wrong.</p>
  4056. </blockquote>
  4058. <p>I have
  4059. <a href="">covered</a>
  4060. <a href="">this territory</a>
  4061. <a href="">before</a>,
  4062. and I certainly agree that there has been a lot of creative interpretation
  4063. of the Constitution in Supreme Court decisions, much of it inconsistent
  4064. when taken as a whole. But again, while this may give some support to the
  4065. claim (which I'll get to in a moment) that ignoring the Constitution is
  4066. not going to cause our society to collapse, it does not at all support the
  4067. claim that we have been paying too much attention to it up to now.</p>
  4069. <p>But we are getting further away from the fiscal cliff, which is what I
  4070. said this post was going to be about. Let's start heading back by looking
  4071. at the central claim of the op-ed:</p>
  4073. <blockquote>
  4074.  <p>Our sometimes flagrant disregard of the Constitution has not produced
  4075.  chaos or totalitarianism; on the contrary, it has helped us to grow and
  4076.  prosper.</p>
  4077. </blockquote>
  4079. <p>Set aside (for the moment) the question of whether we have grown and
  4080. prospered <em>because</em> of ignoring the Constitution, or <em>in spite of</em> that.
  4081. The claim appears to be that basically, ignoring it has worked all right
  4082. up to now, so why not make it official? But once again, that's not at all
  4083. the same as claiming that we have paid <em>too much</em> attention to it, and
  4084. we'd be better off not doing that. Later on, the writer gives more
  4085. details about how our key institutions -- Congress, the Presidency,
  4086. the Supreme Court -- might work if we did this:</p>
  4088. <blockquote>
  4089.  <p>What would change is not the existence of these institutions, but the
  4090.  basis on which they claim legitimacy. The president would have to justify
  4091.  military action against Iran solely on the merits, without shutting down
  4092.  the debate with a claim of unchallengeable constitutional power as
  4093.  commander in chief. Congress might well retain the power of the purse,
  4094.  but this power would have to be defended on contemporary policy grounds,
  4095.  not abstruse constitutional doctrine. The Supreme Court could stop
  4096.  pretending that its decisions protecting same-sex intimacy or limiting
  4097.  affirmative action were rooted in constitutional text.</p>
  4098. </blockquote>
  4100. <p>Going back to the example at the beginning of the op-ed, the Senate
  4101. minority leader would no longer be able to hide behind a Constitutional
  4102. provision to justify inaction. He might still try other ways of political
  4103. posturing, but that particular way would be closed to him.</p>
  4105. <p>There is a point that could be taken away from this, but unfortunately
  4106. it's not the one the writer intended. Giving up the Constitution would
  4107. not stop political posturing; it would only change its form. Giving up
  4108. the Constitution would amount to admitting that <em>no</em> written document,
  4109. <em>no</em> set of rules, can work if people refuse to follow them when it
  4110. serves their interests to break them instead.</p>
  4112. <p>This throws rather a different light on the conclusion of the op-ed:</p>
  4114. <blockquote>
  4115.  <p>If we are not to abandon constitutionalism entirely, then we might at
  4116.  least understand it as a place for discussion, a demand that we make a
  4117.  good-faith effort to understand the views of others, rather than as a
  4118.  tool to force others to give up their moral and political judgments.</p>
  4120.  <p>[P]erhaps the dream of a country ruled by "We the people" is impossibly
  4121.  utopian. If so, we have to give up on the claim that we are a self-governing
  4122.  people who can settle our disagreements through mature and tolerant debate.
  4123.  But before abandoning our heritage of self-government, we ought to try
  4124.  extricating ourselves from constitutional bondage so that we can give real
  4125.  freedom a chance.</p>
  4126. </blockquote>
  4128. <p>But if we can't even make a good faith effort to understand others'
  4129. views when we have a written document setting out the rules for how to
  4130. do so, how will it work any better when we don't? The Constitution, and
  4131. written law in general, is by no means the only tool people have to force
  4132. others to do things they don't want to do. But written laws, like written
  4133. contracts, have at least the advantage of being written: there is a text,
  4134. however imperfect, whose words are a matter of objective fact, rather than
  4135. just vague ideas in people's heads. (Of course, the whole issue of strict
  4136. vs. loose construction of the Constitution, or "originalism" vs. "living
  4137. constitutionalism" as the op-ed has it, shows that even written words
  4138. don't always establish an objective meaning. But having <em>no</em> written
  4139. words would be worse still.)</p>
  4141. <p>You may think I'm still on a detour here; the fiscal cliff is not a matter
  4142. of Constitutional law. Indeed it isn't, and that's the point. Consider what
  4143. got us here: back in the summer of 2011, to deal with the debt ceiling
  4144. crisis, a law was passed that imposed a deadline on Congress and the
  4145. President to deal with budget deficits, or else spending cuts would be
  4146. imposed willy-nilly. But what caused the debt ceiling crisis? Well, another
  4147. law was passed, quite some time ago, that imposed a ceiling on the debt and
  4148. required Congress and the President to periodically revisit it. Why was
  4149. <em>that</em> law passed? Well, because it had become evident that the national
  4150. debt was continuing to grow despite all efforts to control it. And why was
  4151. <em>that</em>? Well, because the government couldn't stop spending more money
  4152. than it was taking in in revenues.</p>
  4154. <p>Where, in <em>any</em> of this, was the Constitution a factor? The Constitution
  4155. doesn't say anything about a debt ceiling, or how the national debt is to
  4156. be controlled. It doesn't say anything about how the government is supposed
  4157. to control its spending. (It does put limits on the <em>kinds</em> of things
  4158. Congress can spend money on, which, as I've argued before, have been
  4159. interpreted out of all recognition by the Supreme Court. But that's just
  4160. another example of us <em>not</em> paying attention to the Constitution.) Blaming
  4161. any of this on the Constitution, or on our supposed adherence to the
  4162. Constitution instead of to practical solutions to problems, is simply a
  4163. misdiagnosis. The problem is that our government refuses to be restrained
  4164. by <em>any</em> set of rules if politicians think that breaking them will help
  4165. them reach their goals.</p>
  4167. <p>It's important to note that this applies to <em>both</em> sides of the aisle.
  4168. The Republicans are the ones currently backed into a corner, but the tactic
  4169. of bending or breaking the rules to one's advantage is used all the time by
  4170. both parties. As an example on the other side, consider all the maneuvering
  4171. done by the Democratic leadership of both houses of Congress to get
  4172. Obamacare passed before the 2010 midterm elections gave control of the
  4173. House to the Republicans.</p>
  4175. <p>But, I hear the op-ed writer protesting, Obamacare was a <em>good</em> thing! Of
  4176. course it's all right to bend (or even break) the rules if you're doing a
  4177. good thing, right? If you have reached "a considered judgment that a
  4178. particular course of action is best for the country", why should you let
  4179. yourself be stopped by a few pesky rules? Of course this is exactly what
  4180. one expects to hear from people of
  4181. <a href="">Heinlein's class one</a>
  4182. when they want to tell other people what to do. But my point here is even
  4183. more drastic than the one I made in that earlier post about my favorite
  4184. Heinlein quote. Here I'm saying that even if a particular course of action
  4185. <em>is</em> best for the country, considered by itself, getting it done by bending
  4186. or breaking the rules is <em>still</em> a net loss, because it destroys people's
  4187. faith in the stability of our society, and that faith is more important
  4188. than the particular course of action we take on any single issue.</p>
  4190. <p>It's worth taking another brief detour here to note that even the premise
  4191. of the argument I just sketched is usually not valid. It is actually
  4192. extremely rare for <em>anyone</em>, in government or out, to have a "considered
  4193. judgment that a particular course of action is best for the country" which
  4194. ends up being correct. The track record of such "considered judgments" is
  4195. extremely poor. Very often doing nothing would be better than whatever the
  4196. government does. When the government passed huge bailouts and a stimulus
  4197. package in 2008 and 2009, the "considered judgment" was that it would fix
  4198. things without too much delay; nobody then was contemplating an economic
  4199. situation in 2012 like the one we actually have.</p>
  4201. <p>In the case of the fiscal cliff (that last detour was actually a shortcut),
  4202. both sides are basing their positions on predictions of the future that
  4203. are no more reliable than the ones that were used four years ago. The
  4204. Democratic position is basically the one described by Paul Krugman in
  4205. <a href=";ref=general">another op-ed</a>
  4206. from yesterday's New York Times: if we let the "sequestration" spending
  4207. cuts happen, <em>and</em> we let the tax rates go up for everyone, we will reduce
  4208. the deficit too fast, leading to another recession. But if we really had
  4209. that much of a handle on how government spending and tax rates impact the
  4210. economy as a whole, we would be in better shape now than we are. It's not
  4211. as though trying to regulate the economy through government spending and
  4212. tax rates is a new idea. The argument seems plausible, but "plausible"
  4213. isn't a very high bar, particularly when Nobel-Prize-winning economists
  4214. like Krugman are spending a lot of time and (electronic) ink on this.</p>
  4216. <p>The Republican position is hard to discern (they have, perhaps wisely,
  4217. refrained from being too explicit about exactly what their goals are),
  4218. but based on comments made over the past couple of days by several
  4219. Republican Senators about "sequestration", they may well think, behind
  4220. closed doors, that going over the cliff might cause short-term pain, but
  4221. could ultimately prove to be a good thing. It would <em>force</em> the
  4222. government to cut spending whether it wants to or not, and the tax hikes
  4223. would make a start at addressing the deficit. But such a strategy, while
  4224. it would be refreshing coming from a party that claims to support
  4225. conservative values but hasn't done much to actually support them, would
  4226. only work if they stuck to it, not just for the next couple of days, but
  4227. for the next couple of <em>years</em> -- the ultimate object in view being, of
  4228. course, to shift the balance in Congress again in the 2014 election by
  4229. pointing to the benefits gained by standing firm on this issue.</p>
  4231. <p>I seriously doubt that the Republican party is capable of holding to <em>any</em>
  4232. strategy for that long. Even if we go over the "cliff" tomorrow, there is
  4233. still plenty of time for a deal to happen before the effects actually
  4234. build up, and such a deal would mean that any supposed benefits of the
  4235. spending cuts and the tax hikes would not actually happen. Even assuming
  4236. that going over the cliff would, in the long run, be a net positive, to
  4237. achieve that, we would have to go over it, and then start climbing back
  4238. up from the bottom of the canyon; stopping the fall part way down won't
  4239. do it. I don't think the Republican party has the stomach for that.</p>
  4241. <p>In the end, I suspect we will, as usual, end up somewhere in the middle.
  4242. And that is what really makes the willingness of both sides to break the
  4243. rules when it suits them so maddening: it doesn't even <em>accomplish</em>
  4244. anything. After all of the grandstanding, we will probably end up with
  4245. much the same deal that would have happened if it had been done a while
  4246. ago, when it should have been. After all of the bending of the rules, we
  4247. will be no better off than if everything had been done the way it should
  4248. have been, within the rules. In fact, we'll be worse off, because we will
  4249. have yet another proof that the rules have no force, and <em>that</em> is the
  4250. real problem.</p>
  4252. <p>The same goes for all of the ignoring of the Constitution that the op-ed
  4253. talks about. What has it really gained us? Consider some of the examples
  4254. the op-ed gives. The Alien and Sedition Acts were indeed contrary to the
  4255. First Amendment; perhaps that's why the main practical impact they had
  4256. was to remove the Federalists, who passed them, from power in the next
  4257. election. The New Deal legislation, which led FDR to pack the Supreme
  4258. Court with Justices favorable to his views, started the very trend of
  4259. increasing spending that has led to our current situation. And nearly
  4260. 60 years after the Brown v. Board of Education decision, which Justice
  4261. Jackson said had no basis in the Constitution (that's debatable, but
  4262. let's concede it to the op-ed writer for the sake of argument), our
  4263. schools, while they may no longer be segregated, are
  4264. <a href="">not exactly paragons</a>
  4265. <a href=",31813,2043378,00.html">of educational achievement</a>.</p>
  4267. <p>Given all this, it seems to me that the problem with the Constitution is
  4268. not that we pay too much attention to it; it's that we never gave it a
  4269. chance at all. The op-ed admits that "No sooner was the Constitution in
  4270. place than our leaders began ignoring it." This is a bug, not a feature.
  4271. Perhaps it's impossible for us to actually abide by a set of rules;
  4272. perhaps the temptation to gain a temporary advantage by bending or
  4273. breaking them is always going to be too strong. But if so, then <em>that</em>
  4274. is what will end up destroying "our heritage of self-government".</p>
  4276. <p>If we really want to preserve that heritage, we ought to try placing a
  4277. higher value on the fundamental stability of our society, and the trust
  4278. and integrity that it requires, even if that means giving up some
  4279. temporary political advantage, or even giving up a quicker solution to
  4280. a specific problem. And if there's one thing the history surveyed by
  4281. the op-ed writer teaches, it's that we need explicit rules to do that.
  4282. It's all very well to talk about "aspirations that, at the broadest
  4283. level, everyone can embrace"; but the same people who talk about sharing
  4284. such broad aspirations will, in the next breath, proceed to disagree on
  4285. every specific point of any substance. The fiscal cliff has shown us
  4286. this in spades: one minute everybody is talking about how no one wants
  4287. to go over the cliff, but the next minute everybody is talking about how
  4288. they can't come to agreement on any specifics for avoiding that.</p>
  4290. <p>The Framers of the Constitution believed that people could not be trusted
  4291. to work things out based on nothing more than shared aspirations. They
  4292. knew that solutions that are worked out under pressure, solutions that are
  4293. pushed by one faction over the objections of another based on the political
  4294. climate of the moment, are likely to be bad in the long run for the country
  4295. as a whole. So they gave us a specific structure that was designed to force
  4296. us to take a step back instead of charging ahead whenever we saw something
  4297. that needed fixing. Our government is broken today because we have lost
  4298. sight of the basic truths that the Constitution was built on. Maybe we
  4299. ought to give the Constitution a chance.</p>
  4300. </div>
  4301. </content>
  4302. </entry>
  4304. <entry>
  4305. <title type="html">"Your" Cloud Data Is Not Yours</title>
  4306. <category term="/opinions" />
  4307. <id></id>
  4308. <updated>2012-12-19T00:48:00Z</updated>
  4309. <published>2012-12-19T00:48:00Z</published>
  4310. <link rel="alternate" type="text/html" href="" />
  4311. <content type="xhtml">
  4312. <div xmlns="">
  4313. <p>A while ago I explained
  4314. <a href="">why I'm not crazy about the cloud</a>.
  4315. In that post I stressed that, since you're not a paying customer to "cloud"
  4316. services like Facebook and Google, you don't get to decide how they're run.
  4317. Now I want to talk about another aspect of the cloud that seems risky to
  4318. me: you don't get to decide how the data you post to a "cloud" service is
  4319. used.</p>
  4321. <p>Yesterday,
  4322. <a href="">Instagram</a>,
  4323. which was recently
  4324. <a href="">acquired by Facebook</a>,
  4325. released updated Terms of Service which were widely interpreted as
  4326. <a href="">claiming the right to sell your photos</a>
  4327. without giving you a penny of compensation. Of course this caused much outrage
  4328. all over the Internet, and Instagram
  4329. <a href="">responded</a>
  4330. by clarifying why they changed their Terms of Service:</p>
  4332. <blockquote>
  4333.  <p>Our intention in updating the terms was to communicate that we'd like to
  4334.  experiment with innovative advertising that feels appropriate on Instagram.
  4335.  Instead it was interpreted by many that we were going to sell your photos to
  4336.  others without any compensation. This is not true and it is our mistake that
  4337.  this language is confusing. To be clear: it is not our intention to sell your
  4338.  photos. We are working on updated language in the terms to make sure this is
  4339.  clear.</p>
  4340. </blockquote>
  4342. <p>This seems clear enough, and further on there is more clarification:</p>
  4344. <blockquote>
  4345.  <p>The language we proposed also raised question about whether your photos can
  4346.  be part of an advertisement. We do not have plans for anything like this and
  4347.  because of that we're going to remove the language that raised the question.</p>
  4348. </blockquote>
  4350. <p>So this looks, on its face, like a good story about a cloud service. The
  4351. service proposed new terms (it's important to note that the new Terms of
  4352. Service will not take effect until January 16, 2013, so Instagram was not
  4353. trying to slip anything by), people raised concerns, and the service responded
  4354. to those concerns. There is even an explicit recognition that Instagram users
  4355. own their photos, not Instagram itself:</p>
  4357. <blockquote>
  4358.  <p>Instagram users own their content and Instagram does not claim any ownership
  4359.  rights over your photos. Nothing about this has changed. We respect that there
  4360.  are creative artists and hobbyists alike that pour their heart into creating
  4361.  beautiful photos, and we respect that your photos are your photos. Period.</p>
  4363.  <p>I always want you to feel comfortable sharing your photos on Instagram and we
  4364.  will always work hard to foster and respect our community and go out of our way
  4365.  to support its rights.</p>
  4366. </blockquote>
  4368. <p>All of this sounds good. But it begs the question: if Instagram is so
  4369. concerned about its users, and if it's so valuable to them, why doesn't it
  4370. let them <em>pay</em> for the service?</p>
  4372. <p>The more you look at what Instagram says, the more this question comes to the
  4373. fore. For example:</p>
  4375. <blockquote>
  4376.  <p>From the start, Instagram was created to become a business. Advertising is
  4377.  one of many ways that Instagram can become a self-sustaining business, but not
  4378.  the only one.</p>
  4379. </blockquote>
  4381. <p>Yes, indeed. So why is it the only one that Instagram appears to be pursuing?
  4382. And why is it continuing to pursue it when it has already led to one near miss?
  4383. Instagram responded well this time, but if they weren't depending on ads for
  4384. their business model, they wouldn't have had to respond at all.</p>
  4386. <p>What's more, the "context" they provide about their business plans makes you
  4387. wonder where "innovative advertising" fits in:</p>
  4389. <blockquote>
  4390.  <p>To provide context, we envision a future where both users and brands alike
  4391.  may promote their photos &amp; accounts to increase engagement and to build a
  4392.  more meaningful following. Let’s say a business wanted to promote their account
  4393.  to gain more followers and Instagram was able to feature them in some way. In
  4394.  order to help make a more relevant and useful promotion, it would be helpful
  4395.  to see which of the people you follow also follow this business. In this way,
  4396.  some of the data you produce — like the actions you take (eg, following the
  4397.  account) and your profile photo — might show up if you are following this
  4398.  business.</p>
  4399. </blockquote>
  4401. <p>What does any of this have to do with ads? It's just straightforward social
  4402. networking.</p>
  4404. <p>I should make clear that I am not accusing Instagram of being engaged in a
  4405. deep conspiracy to hoodwink users, as some comments on the
  4406. <a href="">Hacker News thread</a>
  4407. have implied. I am quite ready to believe that they are perfectly sincere.
  4408. That's the problem: they sincerely believe that pursuing "innovative
  4409. advertising" is a good business model, but just charging users for a service
  4410. that is obviously valuable is not.</p>
  4412. <p>This blind spot is not limited to Instagram, of course. Facebook itself has
  4413. the same problem, although to be fair, Facebook's users fit a very different
  4414. profile from Instagram's users. Google has the problem too; in fact, the
  4415. problem is worse with Google, because their core search service is at a more
  4416. fundamental level than social networking. And yet those core search results
  4417. are now less useful because they can be
  4418. <a href="">skewed by personalization</a>,
  4419. and can
  4420. <a href="">preferentially show results from Google services over competitors</a>.
  4421. In fact, the case of Google is probably worth its own post all by itself.</p>
  4423. <p>Of course, the direct impact of this blind spot will be felt by the cloud
  4424. services themselves, not users; but how will the services respond? It would
  4425. be nice if they would respond in the obvious way, by finding ways for users
  4426. to pay them directly for the value they receive. But I don't see any big push
  4427. in that direction. Instead, I see cloud services looking for more and more
  4428. creative ways to monetize their users' data while keeping the service free.
  4429. Unless the service gets driven out of business by some competitor that <em>does</em>
  4430. let users pay directly, there is only one way this trend can end, as far as
  4431. you the user are concerned: "your" data will ultimately not be yours. It's
  4432. great that Instagram wants to protect its users' rights, but it's not up
  4433. against a hard choice (yet) between cashing in on its users' data and going
  4434. out of business. What will happen when (not if) it is?</p>
  4435. </div>
  4436. </content>
  4437. </entry>
  4439. <entry>
  4440. <title type="html">Strict Constructionist?</title>
  4441. <category term="/opinions" />
  4442. <id></id>
  4443. <updated>2012-12-09T02:34:00Z</updated>
  4444. <published>2012-12-09T02:34:00Z</published>
  4445. <link rel="alternate" type="text/html" href="" />
  4446. <content type="xhtml">
  4447. <div xmlns="">
  4448. <p>I've posted a
  4449. <a href="">few</a>
  4450. <a href="">times</a>
  4451. <a href="">now</a>
  4452. about the Supreme Court, and at one point I noted that I had labeled
  4453. myself a "strict constructionist". Now that the Defense of Marriage
  4454. Act (DOMA) and California's Proposition 8 are
  4455. <a href="">going to the Supreme Court for review</a>,
  4456. having been found unconstitutional in a number of lower court cases,
  4457. I have a chance to swing the pendulum back the other way somewhat.</p>
  4459. <p>The terms "strict constructionist" and "loose constructionist", as they
  4460. are usually used, are actually rather ironic since each really means
  4461. its opposite when you look closely. A strict constructionist like
  4462. Justice Scalia says that "the words of the Constitution say what they
  4463. say and there is no fiddling with them", but he also believes that the
  4464. "traditions" of our society are what we should turn to when the
  4465. Constitution doesn't address something, rather than looking at the
  4466. words in more general terms. But those traditions change over time, as
  4467. he acknowledges; he just thinks that's okay because the changes happen
  4468. by democratic processes.</p>
  4470. <p>Furthermore, the traditions are not always consistent. For example, in
  4471. 1996, the Supreme Court struck down VMI's males-only admissions policy in
  4472. <a href="">United States v. Virginia</a>.
  4473. Justice Scalia dissented, arguing that the standard applied by the
  4474. Court was stricter than it had been in similar cases in the past.
  4475. However, the service academies had been
  4476. <a href=";C/History/milacad.html">admitting women since 1976</a>,
  4477. so Scalia was essentially arguing that it was Constitutional for VMI to
  4478. exclude women, but equally Constitutional for the service academies to
  4479. include them.</p>
  4481. <p>A loose constructionist, on the other hand, believes that what changes
  4482. over time is only our understanding of the Constitution and the
  4483. principles it is based on, not the principles themselves. To a loose
  4484. constructionist, it was always against the principles of the Constitution
  4485. for the federal service academies to exclude women; we just didn't
  4486. understand that until the 1976 laws were passed. Similarly, the Virginia
  4487. law allowing VMI to exclude women was always, strictly speaking,
  4488. unconstitutional; we just didn't acknowledge it until the Supreme Court
  4489. said so in 1996. And if the 1976 federal laws were repealed, a woman
  4490. could bring suit against the service academies for excluding her and the
  4491. Supreme Court ought to support her and declare the new law
  4492. unconstitutional. So we have the ironic situation of a "strict"
  4493. constructionist having to take a position that implies that what is
  4494. just changes over time, while the "loose" constructionist is the one
  4495. arguing that justice itself does not change, although our understanding
  4496. of it does.</p>
  4498. <p>The question the strict constructionist always asks, of course, is what
  4499. justification the loose constructionist can give for finding all these
  4500. "new rights" in the Constitution that aren't mentioned explicitly there.
  4501. I think there is a good argument for this, which I wish would be made
  4502. more explicit in court opinions on these issues. The Constitution was
  4503. not meant to enumerate all rights, powers, or duties explicitly; it
  4504. clearly envisions that people will use common sense in interpreting
  4505. what it says. This is shown in the original document itself by such
  4506. clauses as the "necessary and proper" clause, and in the Bill of Rights
  4507. by the Ninth and Tenth Amendments, which clearly show an expectation
  4508. that the document is not expected to explicitly cover all cases. This
  4509. much is admitted by everyone, but strict constructionists appear to
  4510. believe that there is no middle ground: either the Constitution says
  4511. it explicitly, or it doesn't cover it at all.</p>
  4513. <p>The problem with that strict viewpoint is that it views the Constitution
  4514. as stating laws. Laws either apply or don't apply; a law concerning
  4515. theft doesn't apply to a case of assault. But the Constitution is not
  4516. supposed to be law in this sense; it is concerned with setting up the
  4517. governmental structure that will make, execute, and interpret laws. As
  4518. such, it obviously will have to be applied to cases that were not
  4519. envisioned by the framers, and the only way to do that is to read it,
  4520. where possible, as giving general principles, not specific rules for
  4521. specific cases. That is what the loose constructionist does. For
  4522. example, the Constitution says nothing specifically about marriage,
  4523. but the Fourteenth Amendment does give general guarantees of due
  4524. process and "equal protection of the laws". That means that, if the
  4525. law provides benefits attached to a certain status, such as marriage,
  4526. all citizens must have equal access to that status; the law cannot
  4527. arbitrarily exclude a class of persons (such as gays) from access.
  4528. (This does not mean that the law cannot impose any limitations on
  4529. access; for example, it can specify a minimum age requirement for
  4530. marriage, or require some other legal condition, as long as everyone
  4531. can potentially meet it. Everyone eventually arrives at adult age,
  4532. barring tragedy.) So laws banning gay marriage are unconstitutional,
  4533. and the courts should overturn them.</p>
  4535. <p>As you can see, this issue is actually pretty simple from the loose
  4536. constructionist's point of view. The Constitution doesn't specifically
  4537. say that "equal protection" applies to gay marriage, but it doesn't
  4538. specifically say it applies to anything. In fact, the Constitution
  4539. says nothing whatever about marriage specifically. But "equal protection
  4540. of the laws" is a general principle, not a specific rule for specific
  4541. cases. It is true that long-standing tradition in our society says that
  4542. marriage is between one man and one woman, but there are plenty of ways
  4543. of acknowledging that without violating the equal protection guarantee.
  4544. One obvious way would be to separate the legal aspects of marriage from
  4545. the social ones: find some neutral legal term (like, oh, say, "civil
  4546. union", or "household") and use that to define the legal benefits
  4547. available, and let various social groups decide for themselves what
  4548. they will count as a "marriage". You are perfectly within your rights
  4549. not to invite the gay couple down the street to dinner because you
  4550. don't acknowledge their union, but you are not within your rights to
  4551. say they can't file a joint income tax return, make medical decisions
  4552. for each other if one is incapacitated, buy a house together, or
  4553. inherit from each other without being taxed.</p>
  4555. <p>This brings up the key point that opponents of gay marriage seem
  4556. unable to talk about: married couples get <em>lots</em> of legal benefits.
  4557. As
  4558. <a href="">the Wikipedia page on same-sex marriage</a>
  4559. notes, the GAO has identified well over a thousand benefits conferred
  4560. on married couples, so the ones I listed above, as important as they
  4561. are, barely scratch the surface. To pretend that gay marriage is only
  4562. about whose unions deserve to be recognized socially is to ignore the
  4563. huge network of advantages that "traditional" married couples take for
  4564. granted. Once those advantages are recognized, of course the equal
  4565. protection implications are obvious.</p>
  4567. <p>In the previous post where I said I had labeled myself as a strict
  4568. constructionist, I was making the point that the courts should only
  4569. "say what the law is" in the sense of determining which law should
  4570. govern when different laws conflict, not in the sense of making new
  4571. laws. But the loose constructionist also has a valid point: it is
  4572. perfectly possible for courts to uphold the principles embodied in the
  4573. Constitution without making up new laws and legal frameworks out of
  4574. whole cloth. In the case of gay marriage, as I said above, and also
  4575. back when New York State
  4576. <a href="">passed its law permitting same-sex marriage</a>,
  4577. the issue is pretty simple: does "equal protection" mean what it says,
  4578. or not? In the Supreme Court's opinion in Brown v. Board of Education,
  4579. <a href="">Chief Justice Warren wrote</a>
  4580. that "Such an opportunity, where the State has undertaken to provide
  4581. it, is a right that must be made available to all on equal terms". He
  4582. was talking about education, but I see no reason why the same logic
  4583. should not apply to the legal benefits that the State attaches to
  4584. marriage. If that means I have to turn in my strict constructionist
  4585. membership card, well, so be it.</p>
  4586. </div>
  4587. </content>
  4588. </entry>
  4590. <entry>
  4591. <title type="html">Vote Early, Not Often</title>
  4592. <category term="/opinions" />
  4593. <id></id>
  4594. <updated>2012-11-11T19:17:00Z</updated>
  4595. <published>2012-11-11T19:17:00Z</published>
  4596. <link rel="alternate" type="text/html" href="" />
  4597. <content type="xhtml">
  4598. <div xmlns="">
  4599. <p>The New York Times' "Bits" blog has a
  4600. <a href="">post</a>
  4601. today arguing in favor of digital voting.
  4602. The main argument is that allowing people to vote via the Internet would
  4603. increase turnout:</p>
  4605. <blockquote>
  4606.  <p>According to a report released by the Census Bureau this year, nearly
  4607.  50 million Americans didn’t vote in the 2008 election. Millions of people
  4608.  said this was because they were out of town, had transportation problems
  4609.  or were too busy to get to the polls. Internet voting could let millions
  4610.  more people take part.</p>
  4611. </blockquote>
  4613. <p>The Times also quotes President Obama, who said regarding the long lines
  4614. at the polls in the 2012 election, "We have to fix that."</p>
  4616. <p>However, the post fails to mention that the President, most likely, was
  4617. referring to <em>early</em> voting, not Internet voting, as the fix. Perhaps
  4618. the "Bits" blogger doesn't read other Times blogs and missed
  4619. <a href="">this post</a>
  4620. about President Obama voting early (the first time a President has done
  4621. so), and encouraging others to do so as well. (My wife and I have early
  4622. voted in the past three elections; each time there have been more people
  4623. doing so.)</p>
  4625. <p>Furthermore, the Times says that the inherent security issues with online
  4626. voting are "not impossible" to fix; but it also quotes Ronald Rivest, one
  4627. of the three inventors of RSA, the most widely used strong encryption
  4628. scheme today, as saying the opposite:</p>
  4630. <blockquote>
  4631.  <p>“One of the main goals of the election is to produce credible evidence
  4632.  to the loser that he’s really lost,” he said. "When you have complicated
  4633.  technology, you really do have to worry about election fraud."</p>
  4634. </blockquote>
  4636. <p>No further details are given, but it's pretty easy to fill them in. The
  4637. RSA algorithm works like this: say you want to send me a message in such
  4638. a way that I can prove that it came from you. You generate two encryption
  4639. keys, a public one, that you give to me (and anyone else who wants to get
  4640. digitally signed messages from me), and a private one that you keep secret.
  4641. The two keys are the inverses of each other: each one decrypts what the
  4642. other one encrypts. So if you want to send me a digitally signed message,
  4643. you encrypt it with your private key and send me the encrypted version. I
  4644. decrypt it with your public key. The fact that I get readable text instead
  4645. of gibberish proves that it must have been encrypted with your private key.</p>
  4647. <p>This system works fine <em>as long as your private key stays private</em>. But
  4648. it has to be stored somewhere; the most likely place is on your computer
  4649. (or smartphone, or tablet, or whatever device you want to use to vote).
  4650. What if that device gets infected with a virus that is specifically
  4651. designed to change your vote, and do nothing else? You would have no way
  4652. of knowing it was there; you would use your voting app, cast your vote for
  4653. Candidate A (you think), and the virus encrypts a vote for Candidate B and
  4654. sends it to the voting server. To the server, it looks like a valid vote;
  4655. it's encrypted with your private key. Only you know that you intended to
  4656. vote for Candidate A, not Candidate B; but you don't see the vote that the
  4657. server actually counted.</p>
  4659. <p>Of course, in principle, a recount could be done by looking at every single
  4660. vote counted by the server and asking the corresponding voter if it matched
  4661. his intent, which would show that something fishy was going on. But recounts
  4662. are not currently done that way; they only look at the ballot itself. And
  4663. changing voting law to permit recounts to ask voters about their intent
  4664. would destroy the secrecy of your ballot, not to mention that it would be
  4665. a <em>huge</em> increase in the time required for a recount. (It would also be
  4666. tantamount to admitting that online voting was not secure.) The
  4667. whole point of paper ballots is that you can leave a record of your vote
  4668. that can be verified without raising all those issues.</p>
  4670. <p>By the way, the same issues apply to electronic voting machines at the
  4671. polling place. I've had the option of choosing electronic or paper ballots
  4672. in a number of elections now, and I've always chosen paper. The Times
  4673. mentions that Estonia has more than a million voters who are registered to
  4674. cast their ballots online, apparently to make the point that the United
  4675. States should be a technological leader. But leadership means making the
  4676. <em>right</em> choices, not following every new trend. The United States would do
  4677. better, in my opinion, to set an example of restraint and proper setting
  4678. of priorities for voting, not technological faddishness.</p>
  4680. <p>As a commenter on the Times blog post said, the voting process is not
  4681. supposed to be fast; it's supposed to be <em>accurate</em>, to properly capture the
  4682. vote that <em>you</em> want to cast. We already have a solution for the genuine
  4683. issue of making it easier for more people to cast their votes: early voting.
  4684. But it still has to be <em>secure</em> early voting, and that, I submit, means paper
  4685. ballots, now and for the foreseeable future. If that means we have to wait
  4686. longer for the results, so be it. For one thing, a lot of pundits would be
  4687. able to go to bed at a normal hour on election night.</p>
  4688. </div>
  4689. </content>
  4690. </entry>
  4692. <entry>
  4693. <title type="html">A Proposal for Campaign Finance Reform</title>
  4694. <category term="/opinions" />
  4695. <id></id>
  4696. <updated>2012-11-08T16:27:00Z</updated>
  4697. <published>2012-11-08T16:27:00Z</published>
  4698. <link rel="alternate" type="text/html" href="" />
  4699. <content type="xhtml">
  4700. <div xmlns="">
  4701. <p>Now that the 2016 campaign has officially started, I thought it would
  4702. be a good time to take another look at campaign finance reform. This
  4703. is a very frustrating subject for me, as I'm sure it is for many; every
  4704. scheme I've seen so far, from what I can tell, is just an attempt by
  4705. some special interest groups to give an advantage to their method of
  4706. buying politicians over other methods of buying politicians. But I have
  4707. a proposal to cut through all the posturing and get to the root of the
  4708. problem:</p>
  4710. <blockquote>
  4711.  <p>The only entities who can make political contributions are those who
  4712.  can vote. In other words, only individual voters.</p>
  4713. </blockquote>
  4715. <p>We have a fundamental imbalance in the United States between political
  4716. power and political rights. Special interest groups don't vote.
  4717. Corporations don't vote. Religious organizations don't vote. Think tanks
  4718. don't vote. Lobbying organizations don't vote. Individual people are the
  4719. only ones who vote, and yet their political voices are the easiest ones
  4720. to lose amidst all the noise from everywhere else. There's no single
  4721. change that will fix that overnight, but limiting political contributions
  4722. to individual voters seems like a good start.</p>
  4724. <p>I realize that this scheme is not perfect; the most obvious problem is
  4725. that it gives an advantage to the rich, who have more money to contribute.
  4726. Perhaps limits on individual contributions could be retained, similar to
  4727. or maybe somewhat larger than the
  4728. <a href="">current ones</a>.
  4729. This would mean, for example, that no individual could contribute more
  4730. than a few thousand dollars to any single candidate for a given election.</p>
  4732. <p>What?! I hear the politicians crying. A few <em>thousand</em> dollars per
  4733. person? How are we going to amass our huge war chests at that rate?
  4734. Well, maybe those huge war chests are part of the problem. Maybe
  4735. politicians who don't have those huge pots of money coming from
  4736. non-voting special interests will have to get creative in getting their
  4737. message across. Like maybe actually talking about issues, or maybe even
  4738. writing actual essays about their views and posting them on the web
  4739. where everybody can read them, and having open discussions online about
  4740. them so people can actually weigh the pros and cons. Putting up a web
  4741. site costs next to nothing; even the Green party can afford one.</p>
  4743. <p>But if that doesn't satisfy the politicians, they would have another
  4744. option: actually convince enough individual voters to contribute enough
  4745. money to allow them to buy television airtime, ads in newspapers and
  4746. magazines, and trips around the country to glad-hand the voters. I know
  4747. that sounds like hard work compared to attending gala fund-raisers at
  4748. hundreds or thousands of bucks a plate and getting free food and
  4749. entertainment. But nobody said political life had to be easy.</p>
  4750. </div>
  4751. </content>
  4752. </entry>
  4754. <entry>
  4755. <title type="html">Economics Is Not Optional</title>
  4756. <category term="/opinions" />
  4757. <id></id>
  4758. <updated>2012-10-18T02:16:00Z</updated>
  4759. <published>2012-10-18T02:16:00Z</published>
  4760. <link rel="alternate" type="text/html" href="" />
  4761. <content type="xhtml">
  4762. <div xmlns="">
  4763. <p>I recently came across
  4764. <a href="">a blog post</a>
  4765. proposing a rather novel solution to what it calls "the shortage of
  4766. technology talent in the United States".
  4767. (I got there by following a link trail from
  4768. <a href="">this post</a>
  4769. at
  4770. <a href="">CSpace</a>,
  4771. a sysadmin's blog which I recommend for general reading if you're at
  4772. all interested in tech issues; I read it regularly and I'm not even a
  4773. sysadmin.)</p>
  4775. <p>The blog in question only allows comments if you have a Facebook,
  4776. Twitter, Yahoo, or AOL account, and for reasons which I have blogged
  4777. about
  4778. <a href="">before</a>,
  4779. I have none of the above and don't intend to any time soon, so this
  4780. post will have to do. I have no comment on most of the post, which
  4781. seems to be fairly standard fare in this genre; what got my attention
  4782. was this towards the end:</p>
  4784. <blockquote>
  4785.  <p>I am proud of, and impressed by, Craigslist's ability to serve hundreds
  4786.  of millions of users with a few dozen employees. But I want the next
  4787.  Craigslist to optimize for providing dozens of jobs in each of the towns
  4788.  it serves, and I want educators in those cities to prepare young people
  4789.  to step into those jobs.</p>
  4790. </blockquote>
  4792. <p>In other words, the author wants tech companies like Craigslist to be
  4793. economically inefficient on purpose, in order to provide more tech jobs.</p>
  4795. <p>As
  4796. <a href="">Sherlock Holmes</a>
  4797. would say, this plan seems to have only one drawback, and that is that it is
  4798. intrinsically impossible. Economic inefficiency is economic inefficiency; it
  4799. means you are expending excess resources that could be put to better use
  4800. elsewhere. And that means that, sooner or later, those resources <em>will</em> be
  4801. put to better use elsewhere. How long that process takes, and how painful
  4802. it is, depends on how soon you recognize that it's inevitable and let it
  4803. happen. (As an example of what happens when you <em>don't</em> recognize this early,
  4804. I give you the U.S. economy, and indeed the world economy, over the last few
  4805. years. But that's another post.) Propping up economically inefficient jobs
  4806. just isn't a viable long-term strategy; that's what "economically inefficient"
  4807. <em>means</em>, and it doesn't magically stop being true just because you're doing
  4808. it in what you consider to be a noble cause.</p>
  4810. <p>But what about all those venture capitalists and startup founders that are
  4811. raking in tons of wealth and not sharing it, leading to what the blog post
  4812. author calls "continued income inequality"? Isn't that unfair? Paul Graham
  4813. wrote an essay years ago entitled
  4814. <a href="">Mind the Gap</a>
  4815. that gave a good rebuttal to that argument: the prospect of all that
  4816. wealth is what <em>drives</em> wealth creation. If startup founders didn't have the
  4817. prospect of a huge payoff if they succeed, they wouldn't even try; and if
  4818. the venture capitalists didn't have the prospect of a huge payoff if the
  4819. startup they fund succeeds, they wouldn't fund it. The wealth those startups
  4820. create is not taken from anyone else; it's <em>created</em>, out of nothing, and
  4821. the people who create it get the first dibs on it. That's how it works.</p>
  4823. <p>Furthermore, even if it were somehow true that startup founders, once they
  4824. have gotten the big payoff, have some sort of "duty" to share it by creating
  4825. lots of jobs that their companies don't need, that wouldn't fix what the
  4826. author says is the underlying problem. In
  4827. <a href="">an earlier post</a>
  4828. entitled "To Less Efficient Startups", the author expands on what he thinks is
  4829. wrong with the "standard" startup wealth creation model:</p>
  4831. <blockquote>
  4832.  <p>Instead of generating tens of thousands of middle-class jobs as industrial-age
  4833.  titans did, these companies make a few dozen people truly extraordinarily
  4834.  wealthy, and then give generous payouts to a few hundred people who were already
  4835.  on a path to success by having been privileged enough to go to top universities
  4836.  and by having the identities that tech and engineering cultures are biased toward
  4837.  today.</p>
  4838. </blockquote>
  4840. <p>There are certainly well-known startups and founders that fit this profile
  4841. (Mark Zuckerberg, for example, or even Bill Gates). But the one the author chose
  4842. as his primary example doesn't.
  4843. <a href="">Craig Newmark</a>,
  4844. the founder of Craigslist, came from a normal middle-class background,
  4845. and he founded Craigslist 18 years after he graduated from college. If anything,
  4846. Craigslist <em>illustrates</em> the sort of behavior the author wants
  4847. from startups; the
  4848. <a href="">Wikipedia page</a>
  4849. quotes its CEO as saying that</p>
  4851. <blockquote>
  4852.  <p>Craigslist has little interest in maximizing profit, instead it prefers to help
  4853.  users find cars, apartments, jobs, and dates.</p>
  4854. </blockquote>
  4856. <p>So if Craigslist is operating with a small number of employees, that's because
  4857. it can; it can meet the needs of its owners and its customers with what it has,
  4858. so why should it try to grab more resources and wealth? Instead, it leaves those
  4859. resources and wealth for others to use. If only Wall Street investment banks
  4860. showed that kind of restraint. The author should be applauding Craigslist, not
  4861. telling them to create unnecessary jobs. And in any case, creating unnecessary
  4862. jobs wouldn't do a thing to fix what the author is really complaining about:</p>
  4864. <blockquote>
  4865.  <p>There is effectively no blue collar path to success, notwithstanding the
  4866.  much-vaunted stories of tech company chefs entering these companies in the
  4867.  kitchen and exiting as millionaires.</p>
  4868. </blockquote>
  4870. <p>The way to fix this would be to try to create more blue-collar startup
  4871. founders, not to create more blue-collar employees. But they will still be
  4872. startup founders, facing the same set of potential outcomes and economic
  4873. constraints.</p>
  4875. <p>In fact, the author completely misses a real issue that <em>does</em> contribute to
  4876. excess income inequality ("excess" meaning over and above the amount that's
  4877. needed to drive innovation, per Graham's essay). He says:</p>
  4879. <blockquote>
  4880.  <p>It is a triumph...for Facebook to serve a billion users with just a few thousand
  4881.  employees.</p>
  4882. </blockquote>
  4884. <p>Which is true, but misses a key point: Facebook doesn't get paid by users. It
  4885. gets paid by advertisers, and many if not most of those few thousand employees
  4886. spend their time, not on meeting the needs of users, but on figuring out ever
  4887. more creative and complex ways to get those users to click on buttons that track
  4888. their web usage for advertisers and marketers. If Facebook were paid by its users,
  4889. it might well be able to operate with <em>fewer</em> employees; certainly its employees
  4890. would be spending time doing far more interesting things, not to mention avoiding
  4891. the constant stream of privacy issues and other debacles for users, like
  4892. <a href="">losing "unknown amounts of email"</a>
  4893. with no way to recover them even after the bug that caused it was fixed.</p>
  4895. <p>So the huge concentration of wealth in the hands of Facebook's early investors
  4896. (from the looks of things after the IPO, "early" means "before it went
  4897. public", but that's another post) is more a sign of an economic bubble in
  4898. the ad-driven model of web applications, than a sign of lasting value. Many
  4899. of the Internet startups the author talks about (even Google, at least to an
  4900. extent) are the same way. But that's not something that programmers and
  4901. startup founders can easily fix, because it requires changing user behavior,
  4902. not just programmer and founder behavior. Facebook has a billion users because
  4903. it's easy, centralized, and free. Sooner or later that model may well backfire
  4904. on enough users to make a difference (for the reasons why, see the earlier post
  4905. of mine about the cloud that I linked to above), but even that won't matter
  4906. unless enough people take enough time and expend enough effort to build an
  4907. alternative.</p>
  4909. <p>And there's the real rub: <em>any</em> handicap in building the alternative may well
  4910. kill it. At least one company the author mentions, Kickstarter, is trying to
  4911. do something along these lines, building a distributed way for people with
  4912. ideas and people with spare cash to hook up. Does the author really think that
  4913. such companies will be able to compete with the likes of Facebook by being
  4914. economically inefficient? Because ultimately, that's what it's going to come
  4915. down to.</p>
  4917. <p>I'm all for a more distributed economy, where creation of wealth is open to
  4918. everybody, not just those who can get the attention of venture capitalists.
  4919. But we won't get there by being less efficient on purpose.</p>
  4920. </div>
  4921. </content>
  4922. </entry>
  4924. <entry>
  4925. <title type="html">Python v. Go Redux: Error Handling</title>
  4926. <category term="/rants" />
  4927. <id></id>
  4928. <updated>2012-09-26T01:01:00Z</updated>
  4929. <published>2012-09-26T01:01:00Z</published>
  4930. <link rel="alternate" type="text/html" href="" />
  4931. <content type="xhtml">
  4932. <div xmlns="">
  4933. <p>Some time ago I posted about Go vs. Python with regard to
  4934. <a href="">delimiters</a>.
  4935. I now have another reason to prefer Python to Go:
  4936. <a href="">Go's error handling</a>
  4937. (hat tip:
  4938. <a href="">Hacker News</a>).</p>
  4940. <p>To briefly see the issue, consider the following snippet of idiomatic
  4941. Python:</p>
  4943. <div class="codehilite"><pre><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
  4944.    <span class="c1"># do something with f</span>
  4945. </pre></div>
  4948. <p>What happens if the <code>open</code> call fails? An exception is thrown. If we
  4949. want to deal with it, we wrap the call in a <code>try/except</code> block:</p>
  4951. <div class="codehilite"><pre><span class="k">try</span><span class="p">:</span>
  4952.    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
  4953.        <span class="c1"># do something with f</span>
  4954. <span class="k">except</span> <span class="ne">IOError</span><span class="p">:</span>
  4955.    <span class="c1"># handle failure to open the file</span>
  4956. </pre></div>
  4959. <p>Now consider the corresponding snippet of Go, taken from the blog post
  4960. linked to above:</p>
  4962. <div class="codehilite"><pre><span class="nx">f</span><span class="p">,</span> <span class="nx">err</span> <span class="o">:=</span> <span class="nx">os</span><span class="p">.</span><span class="nx">Open</span><span class="p">(</span><span class="nx">filename</span><span class="p">)</span>
  4963. <span class="k">if</span> <span class="nx">err</span> <span class="o">!=</span> <span class="kc">nil</span> <span class="p">{</span>
  4964.    <span class="c1">// handle failure to open the file</span>
  4965. <span class="p">}</span>
  4966. <span class="c1">// do something with f</span>
  4967. </pre></div>
  4970. <p>Two things jump out at me by comparing the above snippets (leaving aside
  4971. all the stuff about delimiters, etc. that I ranted about last time).
  4972. First, if the file open fails, Python guarantees that the "do something
  4973. with f" code will not execute; Go depends on the programmer putting
  4974. something in the "handle failure to open file" code that does that. Of
  4975. course, fixing that particular wart is easy:</p>
  4977. <div class="codehilite"><pre><span class="nx">f</span><span class="p">,</span> <span class="nx">err</span> <span class="o">:=</span> <span class="nx">os</span><span class="p">.</span><span class="nx">Open</span><span class="p">(</span><span class="nx">filename</span><span class="p">)</span>
  4978. <span class="k">if</span> <span class="nx">err</span> <span class="o">!=</span> <span class="kc">nil</span> <span class="p">{</span>
  4979.    <span class="c1">// handle failure to open the file</span>
  4980. <span class="p">}</span> <span class="k">else</span> <span class="p">{</span>
  4981.    <span class="c1">// do something with f</span>
  4982. <span class="p">}</span>
  4983. </pre></div>
  4986. <p>Which of course begs the question, why didn't the blog post write it
  4987. that way? Perhaps because the poster expected the "do something with f"
  4988. code to test for a valid file object? In other words, they really
  4989. intended to write this:</p>
  4991. <div class="codehilite"><pre><span class="nx">f</span><span class="p">,</span> <span class="nx">err</span> <span class="o">:=</span> <span class="nx">os</span><span class="p">.</span><span class="nx">Open</span><span class="p">(</span><span class="nx">filename</span><span class="p">)</span>
  4992. <span class="k">if</span> <span class="nx">err</span> <span class="o">!=</span> <span class="kc">nil</span> <span class="p">{</span>
  4993.    <span class="c1">// handle failure to open the file</span>
  4994. <span class="p">}</span>
  4995. <span class="k">if</span> <span class="nx">f</span> <span class="o">!=</span> <span class="kc">nil</span> <span class="p">{</span>
  4996.    <span class="c1">// do something with f</span>
  4997. <span class="p">}</span>
  4998. </pre></div>
  5001. <p>(I'm assuming that testing for a non-nil <code>f</code> is sufficient; if it isn't,
  5002. changing the if statement appropriately is straightforward.) In Python,
  5003. no such testing of <code>f</code> is required:; Python guarantees that inside
  5004. the <code>with</code> statement block (the "do something with f" code), <code>f</code> is
  5005. a valid open file object. It can make that guarantee, of course, because
  5006. it can guarantee that the block will not execute if the file open fails.
  5007. In other words, Go forces the programmer to do things by hand that Python
  5008. takes care of automatically, and since those things are, at least IMHO,
  5009. "boilerplate" things that programmers shouldn't have to worry about,
  5010. Python's method is preferable.</p>
  5012. <p>This by itself may not be a huge issue; but now consider the second thing
  5013. that jumped out at me. In Python, I only need to wrap the <code>with</code> block
  5014. in a <code>try/except block</code> if I want to handle the failure condition in that
  5015. particular section of code. Otherwise, I just let the exception propagate
  5016. until something catches it. This is, of course, the whole point of having
  5017. exceptions as your error handling mechanism: it uncouples handling of
  5018. errors from handling of normal conditions. Some people, apparently including
  5019. the Go designers, consider this to be Very Bad Juju, and as you can see,
  5020. in Go you have no choice about where you handle errors; you <em>have</em> to test
  5021. for them and handle them locally, whether you want to or not.</p>
  5023. <p>Why might you <em>not</em> want to? Suppose I'm writing a library to open and parse
  5024. a particular type of file. This library might be used by a variety of
  5025. applications; some might be end-user apps for editing the file, while others
  5026. might be server-side apps that just want a parsed object they can use to
  5027. read attributes from. The way that a failure to open the file should be
  5028. handled is very different for these two types of apps: the end-user app
  5029. needs to display a message to the user (at least if it wants to be usable),
  5030. while the server-side app probably should just log the error and go on,
  5031. or perhaps send an urgent page to a sysadmin.</p>
  5033. <p>If I'm writing this library in Python, handling all this is simple, because
  5034. error handling is uncoupled from normal functionality. Each app's code simply
  5035. catches the <code>IOError</code> exception in the appropriate place and deals with it.
  5036. In Go, what do I do? Either my library gets overgrown with error-handling
  5037. code for all manner of possible use cases, even though I know far less about
  5038. those use cases than the app writers do, or else I have to put together an
  5039. elaborate system of callbacks, plugins, or what-have-you to deal with what
  5040. is fundamentally a simple problem.</p>
  5042. <p>So once again, while it's great that people are trying new things with
  5043. programming languages, I'm still sticking with Python.</p>
  5044. </div>
  5045. </content>
  5046. </entry>
  5048. <entry>
  5049. <title type="html">An Interesting Twist in the Obamacare Debate</title>
  5050. <category term="/opinions" />
  5051. <id></id>
  5052. <updated>2012-09-14T18:06:00Z</updated>
  5053. <published>2012-09-14T18:06:00Z</published>
  5054. <link rel="alternate" type="text/html" href="" />
  5055. <content type="xhtml">
  5056. <div xmlns="">
  5057. <p>As reported by
  5058. <a href="">the Volokh conspiracy</a>,
  5059. the Pacific Legal Foundation is now asking a Federal court to rule
  5060. that Obamacare violates the
  5061. <a href="">Origination Clause of the Constitution</a>.
  5062. Article I, Section 7 says:</p>
  5064. <blockquote>
  5065.  <p>All Bills for raising Revenue shall originate in the House of
  5066.  Representatives; but the Senate may propose or concur with Amendments
  5067.  as on other Bills.</p>
  5068. </blockquote>
  5070. <p>Since the Supreme Court has ruled that
  5071. <a href="">the Individual Mandate is a tax</a>,
  5072. Obamacare qualifies as a bill for raising revenue; and the text of
  5073. the bill originated in the Senate, not the House, so the challenge
  5074. appears to be straightforward.</p>
  5076. <p>There's a twist, though: the text of the bill did originate in the
  5077. Senate, but the actual bill, on paper, originated in the House. The
  5078. Senate took a bill that had been passed by the House, struck out all
  5079. of its language, and replaced it in its entirety with the language
  5080. that was passed by Congress and signed into law by President Obama.
  5081. So technically, the bill did "originate" in the House; it's just
  5082. that none of the language that originated in the House actually
  5083. became part of the final law that was passed.</p>
  5085. <p>Does that make the law unconstitutional? At this point, nobody knows.
  5086. The Supreme Court has ruled in the past that revenue bills which
  5087. originated in the House but were amended by the Senate were
  5088. constitutional, but that is explicitly allowed for by the Constitution,
  5089. as the above quote from Article I, Section 7 makes clear. There has
  5090. never been a ruling before on a bill in which the entire text was
  5091. replaced by the Senate. To me, that seems like an obvious attempt
  5092. to circumvent the constitutional provision, and therefore should not
  5093. be upheld; but of course I'm not a judge.</p>
  5095. <p>It seems inevitable that this issue will reach the Supreme Court
  5096. one way or the other, since whichever side loses at each lower level
  5097. of the Federal courts will certainly appeal. It will be interesting
  5098. to see what the final outcome is, particularly because it looks to
  5099. me like Chief Justice Roberts is likely to be the swing vote again,
  5100. just as he was for the Individual Mandate ruling. It will be still
  5101. more interesting to see how quickly the case is handled, given the
  5102. upcoming election; it would be <em>extremely</em> quick work to even get it
  5103. onto the Supreme Court's dockets by November, but the people bringing
  5104. the challenge are, of course, not unaware of the political implications
  5105. (to put it mildly). We'll see.</p>
  5106. </div>
  5107. </content>
  5108. </entry>
  5110. <entry>
  5111. <title type="html">Announcing Simpleblog</title>
  5112. <category term="/general" />
  5113. <id></id>
  5114. <updated>2012-09-10T22:55:00Z</updated>
  5115. <published>2012-09-10T22:55:00Z</published>
  5116. <link rel="alternate" type="text/html" href="" />
  5117. <content type="xhtml">
  5118. <div xmlns="">
  5119. <p>Well, it's happened: I have yielded to the temptation to write my
  5120. own blogging system, and I am now using it to write and publish
  5121. this blog. It's called simpleblog, and you can find the release
  5122. version on the Python Package Index
  5123. <a href="">here</a>,
  5124. or look at the bleeding-edge development git repository
  5125. <a href="">here</a>.</p>
  5127. <p>I should probably explain that when I say "blogging system" I
  5128. don't mean a complete package that I'm planning on offering as
  5129. a web service for blog publishing any time soon. It's just a
  5130. Python package that lets me write entries for this blog and
  5131. publish them here with an absolute minimum of effort. I wrote
  5132. it because it was getting to be too much effort to get things
  5133. exactly the way I wanted them with any existing blogging system.
  5134. If you've visited here before, you will have seen the "powered
  5135. by Pyblosxom" image on the right, but Pyblosxom was not the only
  5136. package I experimented with while I was trying to make an existing
  5137. system fit my needs. (And I should also emphasize that I'm talking
  5138. about <em>my</em> needs; plenty of other people use existing blogging
  5139. systems and it doesn't seem to get in their way.)</p>
  5141. <p>Just in case you're curious, the two things which pushed me over
  5142. the edge were the two main things that have changed here as of
  5143. this post. First, entries on index pages now only show the
  5144. portion "above the fold", with a link to the full entry. This
  5145. makes index pages a lot shorter and more manageable. Second,
  5146. the individual entry pages now have links at the bottom to the
  5147. next and previous entries in the blog as a whole, within the
  5148. entry's category, and within each of the entry's tags. This makes
  5149. it easier to browse the blog post by post without having to flip
  5150. back and forth between an index page and entry pages. There are
  5151. plenty of other features I've added, of course, but most of them
  5152. are behind the scenes; they make it easier for me, but they aren't
  5153. visible to you, reading this.</p>
  5155. <p>Of course I could have tried to write plugins for Pyblosxom to
  5156. do the above (I couldn't find any existing ones or I would have
  5157. used them). In fact, I had already written several home-brewed
  5158. plugins to add other small features to Pyblosxom. But that was
  5159. part of the problem: writing those plugins had shown me that it,
  5160. for whatever reason, Pyblosxom's architecture was not making it
  5161. easy for me to do what I wanted to do. Designing simpleblog, and
  5162. architecting it so that modifying it would be easy for me, has
  5163. gone very smoothly. (Some of the issues I was having were probably
  5164. because this blog is statically generated, whereas Pyblosxom is
  5165. mainly intended for dynamically serving blogs; it has static
  5166. rendering, but that's not its primary purpose. Simpleblog, right
  5167. now, <em>only</em> does static rendering, so that's what it's optimized
  5168. for.)</p>
  5170. <p>If you want to try simpleblog yourself, please do; the release
  5171. source on PyPI is stable (though of course I've only tested it
  5172. with this blog and the "example" blog that comes with the source).
  5173. I make no guarantees about the development source on github, but
  5174. I will try not to push to it unless this blog and the example
  5175. blog render properly (for some value of "properly"). If you do
  5176. try it, please let me know what you think, either by email or
  5177. through github's issues and pull request system.</p>
  5178. </div>
  5179. </content>
  5180. </entry>
  5182. <entry>
  5183. <title type="html">I Miss Konqueror</title>
  5184. <category term="/rants" />
  5185. <id></id>
  5186. <updated>2012-08-23T01:17:00Z</updated>
  5187. <published>2012-08-23T01:17:00Z</published>
  5188. <link rel="alternate" type="text/html" href="" />
  5189. <content type="xhtml">
  5190. <div xmlns="">
  5191. <p>I recently came across
  5192. <a href="">this</a>
  5193. from Jamie Zawinski, and one of his gripes with Firefox struck a <em>huge</em>
  5194. chord with me:</p>
  5196. <blockquote>
  5197.  <p>The Firefox UI is a moving target. It is under constant "improvement",
  5198.  which means "change" which means every few months I'm forced to upgrade
  5199.  it and shit has moved around and I need to re-learn how to do a task
  5200.  that I was happily doing before.</p>
  5201. </blockquote>
  5203. <p>Not only that, but each "upgrade" seems to give the Linux version of
  5204. Firefox, at least, a new crop of bugs. The latest Firefox "upgrade" from
  5205. Ubuntu was so crappy that I was forced to build my own copy of an
  5206. earlier version from source, so I could delete the standard package.
  5207. (Yes, I know that means I won't get "security fixes" and so forth any
  5208. more. I'll deal with it. See below for some further comments.) To pick
  5209. just one example: when Firefox re-drew its window, menu headings and menu
  5210. item titles would often disappear, so that I would see blank spaces where
  5211. words like "File", "Edit", "Copy", "Options", etc. were supposed to be.
  5212. What the f---?</p>
  5214. <p>And the thing is, it isn't just Firefox. It's practically <em>everything</em> on
  5215. the desktop. It may be more noticeable on Firefox these days, but it's
  5216. everywhere. For example, a while back I ranted about how
  5217. <a href="">KDE 4 sucks</a>.
  5218. Most of my complaints can be filed under this heading: they changed stuff
  5219. that worked just fine as it was. And the other Linux desktop environments
  5220. are no better; here's a good quick example of
  5221. <a href="">jwz on gnome</a>.
  5222. And don't even get me started on Ubuntu's latest eye candy (over and above
  5223. the KDE 4 suckage that I've already mentioned).</p>
  5225. <p>To be fair, some of the changes on the desktop are responses to changes
  5226. elsewhere. For example, as the title of this post tells you, I miss the
  5227. old days when I could use
  5228. <a href="">Konqueror</a>
  5229. (the
  5230. <a href="">KDE 3 Trinity</a>
  5231. version, thank you very much) to surf the web with no issues. The Konq was
  5232. not a very popular browser, even among people who were otherwise solidly
  5233. hooked on KDE; but it just worked, at least for me, and more importantly,
  5234. it <em>worked the same</em> through a lot of KDE 3 releases. I never had to re-learn
  5235. the browser interface just because my Linux distro had decided to do an
  5236. update. Not only that, but its rendering was more consistent than other
  5237. browsers. (This was because Konqueror, unlike other browsers, actually
  5238. paid attention to actual standards.)</p>
  5240. <p>But over the years, the web changed, and more and more sites would cause
  5241. Konqueror to hiccup or even crash, simply because it was not being kept
  5242. up to date with the latest Web 2.0 fads like other browsers. This
  5243. eventually pushed me to the point of having to switch to Firefox for most
  5244. of my surfing, because back then, Firefox was still reasonably clean and
  5245. fast instead of the bloatware it is now. Which meant that I now had to,
  5246. as jwz says, re-learn how the browser works every time a version upgrade
  5247. came out. Which eventually meant (combined with the bloatware issue) that
  5248. I had to switch to Chrome for most of my surfing, because Chrome, at least
  5249. for the time being (but I'm not expecting it to last), is reasonably clean
  5250. and fast.</p>
  5252. <p>(Of course, now I get to wonder how much additional data Google is
  5253. collecting on me since I'm using their browser. I do <em>not</em> use it to surf
  5254. to sensitive sites, like my bank, btw; I use my built-from-source copy of
  5255. Firefox, the earlier version, for that, with all privacy settings cranked
  5256. up to maximum. But that's a different issue, which I've blogged about
  5257. <a href="">before</a>;
  5258. I would do it even if none of the stuff I'm complaining about in this post
  5259. were a problem.)</p>
  5261. <p>Linux distributions do the same thing, and with the same annoying
  5262. results. Ubuntu, thank goodness, still believes in long-term support
  5263. releases, which means I can continue to run 10.04 while I wait for all
  5264. the bugs to be sorted out of 12.04, without worrying that my current
  5265. version will drop out from under me and force me into an unwanted
  5266. upgrade (which happened a number of times with other distros). But one
  5267. of the things I am waiting for with 12.04 is for the KDE 3 Trinity
  5268. project to do a build based on it, which hasn't happened yet, and
  5269. doesn't look like happening any time soon. Which is now making me look
  5270. around to see what else is out there in case I have to switch desktops
  5271. (again) so that at least I can do it on my own initiative instead of
  5272. being forced into it.</p>
  5274. <p>Zawinski is right that the Mac interface has been a lot more stable over
  5275. the years than Linux desktops have. Unfortunately, I can't stand the Mac
  5276. interface, which in addition to all the other reasons
  5277. <a href="">why I run Linux</a>,
  5278. keeps me from considering switching to a Mac for ordinary use. If there
  5279. were a version of Safari for Linux, I might consider using it (for one
  5280. thing, they still, last I checked, use the KHTML rendering engine that
  5281. Konqueror used), but of course I don't expect that to happen any time
  5282. soon. :-)</p>
  5284. <p>But even Apple is not immune to the "force users to upgrade whether they
  5285. need to or not" problem. (I take it the fact that Microsoft was not, is not,
  5286. and never shall be so immune is too obvious to need mentioning.) We have a
  5287. MacBook that is six years old or so, and still runs OS X 10.4. It works
  5288. fine, if a little slow for some things (but that's a much a matter of
  5289. Internet bandwidth as anything); but there are a number of apps out there
  5290. now that require 10.5 or later (or in some cases even 10.6 or later).
  5291. Upgrade? Oh, sure--if we pay Apple for the privilege. Or, of course, we
  5292. could pay them even more for the privilege of buying a <em>new</em> MacBook to
  5293. replace a perfectly good older one. Sigh.</p>
  5295. <p>As I said in my earlier post about KDE 4, in the grand scheme of things,
  5296. this isn't all that big a deal. But it makes you wonder what all these
  5297. people are thinking. At the end of the day, we're talking about drawing
  5298. text, rectangles, and little images on the screen. This is not rocket
  5299. science, and it shouldn't require the level of incessant design effort that
  5300. goes into, say, nuclear reactors. This should be a solved problem by now.
  5301. Of course, that's not to say that if it were, all the people currently
  5302. working feverishly on it would switch to something useful as opposed to,
  5303. say, figuring out ways to get users to click on ads. But one could hope.</p>
  5304. </div>
  5305. </content>
  5306. </entry>
  5308. <entry>
  5309. <title type="html">Still Another Nerd Interlude</title>
  5310. <category term="/opinions" />
  5311. <id></id>
  5312. <updated>2012-07-28T03:59:00Z</updated>
  5313. <published>2012-07-28T03:59:00Z</published>
  5314. <link rel="alternate" type="text/html" href="" />
  5315. <content type="xhtml">
  5316. <div xmlns="">
  5317. <p>(Note: there is a good discussion of this post and the Knuth-McIlroy
  5318. exchange on
  5319. <a href="">Hacker News</a>.)</p>
  5321. <p>The other day I came across
  5322. <a href="">a blog post</a>
  5323. about an interesting exchange between two world-class programmers,
  5324. <a href="">Donald Knuth</a>
  5325. and
  5326. <a href="">Doug McIlroy</a>.
  5327. I talked about McIlroy in
  5328. <a href="">my last nerd interlude</a>. To
  5329. briefly summarize what happened (the article goes into more detail and
  5330. is worth reading, at least if you're into this sort of thing), Knuth
  5331. was asked by a computer magazine to write a program that illustrated
  5332. "literate programming", a technique that Knuth had spent much time
  5333. developing, and McIlroy was asked to critique what Knuth did.
  5334. The program was supposed to solve the following problem, as described
  5335. in the blog post:</p>
  5337. <blockquote>
  5338.  <p>Read a file of text, determine the n most frequently used words, and
  5339.  print out a sorted list of those words along with their frequencies.</p>
  5340. </blockquote>
  5342. <p>As it turned out, the heart of McIlroy's critique was simple but
  5343. devastating: a six-line Unix shell script that accomplished exactly
  5344. the same task as Knuth's 10-page Pascal program. McIlroy then went on
  5345. to explain why this was a better method of solving the problem than
  5346. Knuth's, his main argument being that his method re-used general purpose
  5347. utilities instead of writing custom code. Just to lay my cards on the
  5348. table at once, I'm with McIlroy on this one, as the author of the blog
  5349. post appears to be, and I'll spend most of the rest of this post giving
  5350. some reasons why. Unfortunately it's too late to comment on the blog post
  5351. itself, or I'd be doing it there, since I'll be responding to comments
  5352. that were made there.</p>
  5354. <p>A number of commenters on the post said that McIlroy's critique was
  5355. unfair, because Knuth's program was intended to illustrate his
  5356. literate programming techniques, not to serve as an actual real-world
  5357. solution of the problem he was given. This misses the point. McIlroy
  5358. specifically points out, not just that reusable code is a good thing in
  5359. general, but that Knuth's program, specifically, <em>does not use or produce
  5360. any</em>. It is a one-off program to solve a one-off problem, and it doesn't
  5361. break the problem down into generic sub-problems, as the shell pipeline
  5362. does. It implements everything specifically for this particular problem.
  5363. As McIlroy notes:</p>
  5365. <blockquote>
  5366.  <p>Everything there--even input conversion and sorting--is programmed
  5367.  monolithically and from scratch. In particular the isolation of words,
  5368.  the handling of punctuation, and the treatment of case distinctions are
  5369.  built in. Even if data-filtering programs for these exact purposes were
  5370.  not at hand, these operations would well be implemented separately: for
  5371.  separation of concerns, for easier development, for piecewise debugging,
  5372.  and for potential reuse.</p>
  5373. </blockquote>
  5375. <p>It's quite true that this criticism is orthogonal to the question
  5376. of whether literate programming, in general, is a good thing. But it is
  5377. certainly <em>not</em> orthogonal to the question of <em>how to use</em> any
  5378. programming methodology, literate or otherwise. McIlroy is not saying
  5379. that Knuth's program is bad; he's saying it could have been a lot better,
  5380. and served as a much better showpiece for literate programming, if it
  5381. had taken a different approach to solving the actual problem.</p>
  5383. <p>Other commenters noted that you can have reusable code with libraries just
  5384. as easily as with separate Unix utilities called by shell scripts. That
  5385. is quite true. (For example, see my Python solution to the problem, linked
  5386. to below, which is entirely built out of built-in Python commands and
  5387. routines from the Python standard library.) But either way, you have to
  5388. have the reusable code available. What if you don't? That was McIlroy's
  5389. point: if you don't <em>have</em> any reusable code yet (which Knuth didn't in
  5390. the Pascal he was working with), why not build some as part of building
  5391. your program? Instead of writing a one-off program to solve a one-off
  5392. problem, why not write a library, or set of libraries, that provides a
  5393. set of generic tools that you can then use to compose the solution to
  5394. your specific problem?</p>
  5396. <p>Of course McIlroy already had the generic tools, the Unix utilities. But
  5397. that's because <em>he helped build them</em>. He <em>invented</em> Unix pipes,
  5398. remember? Knuth had an opportunity to do for Pascal what McIlroy and
  5399. those he worked with did for Unix: Knuth could have emerged from his
  5400. effort with a bunch of Pascal libraries that did a lot of the same general
  5401. tasks as the Unix utilities. But he didn't.</p>
  5403. <p>This is also why the commenters who talked about "portability" are
  5404. missing the point. Yes, McIlroy's specific solution would only work on a
  5405. Unix system where those utilities are available. But again, they're there
  5406. because he helped build them. Knuth's solution is "portable" to any OS
  5407. that has a Pascal compiler, but so what? Once you've compiled it,
  5408. you still have a one-off program to solve a one-off problem. Why not
  5409. build the generic utilities in Pascal instead, and have <em>them</em> be
  5410. portable? (And in any case the Unix utilities are written in C and so are
  5411. portable to any OS that has a C compiler, i.e, to a superset of those
  5412. that have Pascal compilers. Yes, some of the system calls would have to
  5413. be changed, but that would be true of the Pascal version as well.)</p>
  5415. <p>Finally, another commenter noted that Knuth's program is easily extensible
  5416. and claimed that McIlroy's shell script is not. He gave as an example
  5417. handling contractions. Now, I am certainly no McIlroy at shell script (or
  5418. programming in general, for that matter--I noted in my
  5419. <a href="">previous nerd interlude</a>
  5420. that McIlroy had written a set of Haskell one-liners that do the same job
  5421. as the multiple files of Python code I wrote to implement power series
  5422. as generators), but even I can see how to handle contractions in shell
  5423. script: just add four lines to the beginning of McIlroy's pipeline, and
  5424. modify one of his lines:</p>
  5426. <div class="codehilite"><pre>sed s/<span class="o">[</span>^A-Za-z<span class="o">]</span><span class="se">\&#39;</span>/<span class="se">\ </span>/ <span class="p">|</span>
  5427. sed s/^<span class="se">\&#39;</span>/<span class="se">\ </span>/ <span class="p">|</span>
  5428. sed s/<span class="se">\&#39;</span><span class="o">[</span>^A-Za-z<span class="o">]</span>/<span class="se">\ </span>/ <span class="p">|</span>
  5429. sed s/<span class="se">\&#39;</span>$/<span class="se">\ </span>/ <span class="p">|</span>
  5430. tr -cs <span class="se">\&#39;</span>A-Za-z <span class="s1">&#39;\n&#39;</span> <span class="p">|</span>
  5431. tr A-Z a-z <span class="p">|</span>
  5432. sort <span class="p">|</span>
  5433. uniq -c <span class="p">|</span>
  5434. sort -rn <span class="p">|</span>
  5435. sed <span class="si">${</span><span class="nv">1</span><span class="si">}</span>q
  5436. </pre></div>
  5439. <p>The first four lines just change quotes that don't lie between two
  5440. letters--i.e., that aren't part of contractions--to spaces. (There's
  5441. probably a slicker way to do this with <code>sed</code>, but what's there works,
  5442. and as I said, I'm not McIlroy.) Then we just modify the <code>tr</code> line to
  5443. include the remaining quotes in the set of characters we don't change to
  5444. newlines. Boom. Done.</p>
  5446. <p>The commenter gives further challenges: count multiple spellings (for
  5447. example, "color" vs. "colour") as single words, and hyphenated words.
  5448. He seems to think those would be difficult in shell script. But having
  5449. seen how contractions were handled just now, it should be obvious that
  5450. those other cases could be handled as well. Hyphenated words can be
  5451. treated the same way contractions were--in fact just adding the hyphen
  5452. character to the regular expressions in the first and third <code>sed</code> lines
  5453. above, as an alternate to the quote, should do it. Multiple spellings
  5454. require a little more, since the script would have to take as input a
  5455. table of what different spellings count as the same word. But given such
  5456. a table, using it to canonicalize the spellings of words prior to sorting
  5457. them is straightforward. True, it might actually require writing a small
  5458. separate filter script to do the canonicalization--but such a script
  5459. would just be another generic, reusable tool in the same spirit as the
  5460. Unix utilities. (The actual programming is left as an exercise for the
  5461. reader--though at some point I may post my own solutions.)</p>
  5463. <p>Of course all these cases could be handled in Pascal (or Python--my
  5464. Python solution includes an extended version that handles contractions)
  5465. as well. But once again, debating whether they would be harder to handle
  5466. in shell is missing the point. Knuth's program could no doubt have been
  5467. modified to handle contractions, multiple spellings, hyphenation, etc.,
  5468. etc. But <em>how</em> would he have modified it? Would he just have written
  5469. more custom code? In shell, you just keep composing the same simple
  5470. utilities in new ways. That's what makes them so powerful. (And as I
  5471. noted above, I have tried to take the same approach in my Python
  5472. solutions.)</p>
  5474. <p>So with all due respect to Donald Knuth (which is a lot, make no
  5475. mistake), I have to agree with McIlroy's final assessment:</p>
  5477. <blockquote>
  5478.  <p>Knuth has shown us here how to program intelligibly, but not wisely.
  5479.  I buy the discipline. I do not buy the result.</p>
  5480. </blockquote>
  5482. <h1>Postscript</h1>
  5484. <p>As I mentioned above, I couldn't resist the temptation to program a
  5485. solution to this problem in Python. It's quite a bit more than six
  5486. lines. :-) You can see it on github
  5487. <a href="">here</a>,
  5488. along with an enhanced version that handles contractions, and McIlroy's
  5489. shell pipeline along with my enhanced version above, for comparison.
  5490. There's also a test text file on which you can run the different versions
  5491. to see the output, along with the expected output for each version.</p>
  5492. </div>
  5493. </content>
  5494. </entry>
  5496. <entry>
  5497. <title type="html">The Supreme Court Does It Again</title>
  5498. <category term="/opinions" />
  5499. <id></id>
  5500. <updated>2012-06-29T03:02:00Z</updated>
  5501. <published>2012-06-29T03:02:00Z</published>
  5502. <link rel="alternate" type="text/html" href="" />
  5503. <content type="xhtml">
  5504. <div xmlns="">
  5505. <p>Today's great news story is that the US Supreme Court has upheld the
  5506. "individual mandate" portion of the Affordable Care Act. I won't bother
  5507. linking to any particular stories since it's everywhere by now. I also won't
  5508. comment here on whether or not the individual mandate (or indeed the act
  5509. itself) is a good idea; that would be a much longer post than I want to
  5510. write right now. Instead, I want to look at the Court's opinion from the
  5511. viewpoint I have posted about
  5512. <a href="">before</a>,
  5513. that the Court has turned what Chief Justice Marshall called the power "to
  5514. say what the law is" into something very different from what Marshall's
  5515. opinion in Marbury v. Madison was arguing for.</p>
  5517. <p>The Court's opinion is
  5518. <a href="">here</a>, and
  5519. is worth reading in full, if for no other reason than to see a good up
  5520. to date example of the kind of legal reasoning that the Court likes to
  5521. engage in. But the gist of it can be summarized briefly thus: the
  5522. individual mandate is not constitutional under the Commerce Clause or
  5523. the Necessary and Proper Clause (the primary argument made by the
  5524. Government in the case), but it <em>is</em> constitutional under the Taxing
  5525. Clause. In other words, the Federal government can't say that you are
  5526. required to buy health insurance, but it can force you to pay extra
  5527. taxes if you don't.</p>
  5529. <p>Of course the proponents of the individual mandate are calling this a
  5530. victory, but they should stop and think for a bit before getting too
  5531. overjoyed. The Supreme Court has just declared that the individual
  5532. mandate is a <em>tax</em>. That means the Affordable Care Act can now be termed
  5533. a tax increase, and you can bet that opponents are going to be doing
  5534. exactly that from now until election day. Moreover, by closing off the
  5535. Commerce Clause and Necessary and Proper Clause justifications, the Court
  5536. has basically said that <em>any</em> regulation of health care, if it is going
  5537. to be economically feasible (and the chief justification for the individual
  5538. mandate has always been that without it the act as a whole is not
  5539. economically feasible), is going to involve a tax increase. Not that I
  5540. disagree with this proposition; in fact I would be more than happy to
  5541. find people applying this kind of reasoning to <em>all</em> efforts by the
  5542. government to fix problems. But it's not the kind of reasoning that the
  5543. proponents of the act want people to engage in.</p>
  5545. <p>Even so, this ruling is undoubtedly bad news for those who had hoped
  5546. that the individual mandate would be ruled unconstitutional. In my
  5547. original post on the
  5548. <a href="">Marbury v. Madison</a>
  5549. decision, I argued that the Court has been allowing Federal power to
  5550. expand far beyond what the Framers intended for a long time, arguably
  5551. since the time of that very decision. So this ruling is no surprise on
  5552. that score. What is interesting, though, is the comparison between the
  5553. ruling on the individual mandate and the accompanying ruling on the
  5554. Medicaid portion of the act. In the latter ruling, the Court agreed
  5555. with the States that it is unconstitutional for Congress to withhold
  5556. funds for existing Medicaid benefits from States that decline to
  5557. support the expanded benefits authorized under the act. What's more,
  5558. the Court agreed with the States (and the opinion of the Court
  5559. disagreed sharply with the Court's own dissenters) that the expansion
  5560. of Medicaid was not "part of the existing program" (a claim based
  5561. largely on language in the original Medicaid act that allowed Congress
  5562. to "modify" the program), but was properly seen as a <em>new</em> Federal
  5563. program even though Congress had not labeled it as such.</p>
  5565. <p>So there is an interesting parallel between the two sections of the
  5566. Court's opinion: it ruled that the individual mandate is a tax, even
  5567. though it was not labeled as such, and it ruled that the expansion of
  5568. Medicaid is a new program, even though it was not labeled as such. True,
  5569. the two parallel rulings have opposite effects: the States won't be
  5570. forced to adopt expanded Medicaid, but we'll all have to either buy
  5571. health insurance or pay extra taxes. But in both cases, the end result
  5572. is that most of the act is upheld; expanded Medicaid and the individual
  5573. mandate are both still there, even if slightly muted in effect. And
  5574. the Court most certainly did <em>not</em> agree with the States that finding
  5575. two particular provisions of the act unconstitutional required nullifying
  5576. the entire act; it explicitly ruled that the two provisions are severable
  5577. from the rest of the act, which therefore remains in effect. (Since the
  5578. act contained explicit language about severability, this portion of the
  5579. Court's opinion is hardly surprising.)</p>
  5581. <p>The really interesting part is that Chief Justice Roberts wrote and
  5582. delivered the opinion of the Court. While there is a lot of
  5583. "strict constructionist" language in the opinion (and there are also
  5584. several comments to the effect that the Court is not expressing any
  5585. opinion on the wisdom of the act--as the concluding remarks of the
  5586. opinion put it, "that judgment is reserved to the people"), there is
  5587. nothing to hinder future expansion of Federal power in a practical
  5588. sense. Indeed, the Court's opinion practically gives a roadmap of how
  5589. to do so: just make it a tax. (True, there are plenty of comments on
  5590. the limits of such tactics, but they are not very restrictive limits,
  5591. practically speaking.) And Justice Ginsburg's opinion, which argues
  5592. that the individual mandate <em>should</em> have been upheld under the Commerce
  5593. Clause, gives plenty of scope for future Courts to find ways to ignore
  5594. the parts of today's opinion that are inconvenient for those who want
  5595. to keep expanding the government's power. (In fact, Ginsburg practically
  5596. admits this: "if history is any guide, today's constriction of the
  5597. Commerce Clause will not endure.")</p>
  5599. <p>Of course, the dissenting opinion by Justices Scalia, Kennedy, Thomas,
  5600. and Alito goes in the other direction, arguing that the entire act
  5601. should be struck down because the clauses that are unconstitutional
  5602. are central to its operation, and doing so with far stronger "strict
  5603. constructionist" language than the Court's opinion. In fact, this
  5604. opinion probably has the soundest arguments of any of those issued
  5605. today, considered logically. For example, a key argument for the
  5606. individual mandate being a necessary aspect of the act is the adverse
  5607. selection problem: without it, healthy people will simply not buy
  5608. health insurance, and rates will skyrocket. However, as the dissenting
  5609. opinion points out, that problem is not unique to health care (as
  5610. Justice Ginsburg's opinion claims); <em>any</em> industry that is regulated
  5611. by the government finds its market skewed by such regulation. The
  5612. dissent points out many other instances in which the majority opinion,
  5613. Justice Ginsburg's opinion, and the Government's arguments in the case
  5614. are, to say the least, questionable. But dissenting opinions,
  5615. particularly by the more conservative justices, have been like this
  5616. before (for example, consider Scalia's dissent in Planned Parenthood
  5617. v. Casey in 1992), and it hasn't had any effect yet.</p>
  5619. <p>(By the way, there appears to be considerable speculation that Roberts
  5620. switched his vote at the last minute, and that what now appears as the
  5621. dissenting opinion by the four Justices referred to above was originally
  5622. supposed to be part of the majority opinion. See, for example,
  5623. <a href="">this post at the Volokh Conspiracy</a>
  5624. and the links there. If that is true it makes the above observations
  5625. even more interesting.)</p>
  5627. <p>So nothing in today's events changes the general conclusions I reached
  5628. in my previous posts about the Supreme Court. I was actually somewhat
  5629. surprised by the ruling; I had expected (though not very strongly, as
  5630. it's always difficult to predict which way the Court will jump on an
  5631. issue this close) the individual mandate to be ruled unconstitutional.
  5632. That would have been a change from the pattern I noted in my previous
  5633. posts, but not much of one, I admit. And if Justice Ginsburg is right,
  5634. even the small change in pattern on the Commerce Clause (how many times
  5635. has the Court ruled that <em>anything</em> doesn't come under the scope of the
  5636. Commerce Clause?) will not last. (Before today I would have said it
  5637. would last at least as long as Roberts' Chief Justiceship, but now I'm
  5638. not so sure.) I wonder what James Madison would think.</p>
  5639. </div>
  5640. </content>
  5641. </entry>
  5643. <entry>
  5644. <title type="html">Climate Change Alarmists: Get Off The Soapbox</title>
  5645. <category term="/opinions" />
  5646. <id></id>
  5647. <updated>2012-04-25T03:19:00Z</updated>
  5648. <published>2012-04-25T03:19:00Z</published>
  5649. <link rel="alternate" type="text/html" href="" />
  5650. <content type="xhtml">
  5651. <div xmlns="">
  5652. <p>In my
  5653. <a href="">last post</a>
  5654. I mentioned that global warming would get its own post
  5655. sometime soon; it appears that now is the time.
  5656. I ran across a
  5657. <a href="">quick update</a>
  5658. from Steve McIntyre at Climate Audit (linked to from
  5659. <a href="">Watts Up With That</a>)
  5660. that mentions Michael Mann's new book:</p>
  5662. <blockquote>
  5663.  <p>I had also spent some time considering a response to Mann's book. It
  5664.  amazes me that a reputable scientific community would take this sort of
  5665.  diatribe seriously. Mann's world is populated by demons and bogey-men.
  5666.  People like Anthony Watts, Jeff Id, Lucia, Andrew Montford and myself
  5667.  are believed to be instruments of a massive fossil fuel disinformation
  5668.  campaign and our readers are said to be "ground troops" of
  5669.  disinformation. The book is an extended ad hominem attack, culminating
  5670.  in salivation in the trumped up plagiarism campaign against Wegman,
  5671.  arising out of copying of trivial "boilerplate" by students (not Wegman
  5672.  himself). Wegman's name appears nearly 200 times in the book (more, I
  5673.  think, than anyone else's).</p>
  5675.  <p>Virtually nothing in its discussion of our criticism can be taken at
  5676.  face value. Mann begins his account by re-cycling his original outright
  5677.  lie that we had asked him for an "excel spreadsheet". Mann's lies on
  5678.  this point had been a controversy back in November 2003. The incident
  5679.  was revived by the Penn State Investigation Committee, which had
  5680.  (anomalously on this point) asked Mann about an actual incident. Instead
  5681.  of "forgetting", as any prudent person would have done, Mann brazenly
  5682.  repeated his earlier lie to the Penn State Investigation Committee.
  5683.  Needless to say, the "Investigation" Committee didn't actually
  5684.  investigate the lie by crosschecking evidence, but accepted Mann's
  5685.  testimony as ending the matter. In the book, instead of leaving well
  5686.  enough alone, Mann once again re-iterated the lie.</p>
  5688.  <p>Or to pick another example, Mann noted the controversy about the
  5689.  contaminated Korttajarvi sediments (Tiljander), but conceded nothing.
  5690.  Mann said that there was no "upside down" in their "objective" methods
  5691.  and asserted that his results were "insensitive to whether or not these
  5692.  records were used", a statement contradicted in the SI to
  5693.  Mann et al 2009. In any sane world, Mann would have issued a retraction
  5694.  of the many claims of Mann et al 2008 that depended on the contaminated
  5695.  Korttajarvi sediments. But instead, more attacks on critics.</p>
  5696. </blockquote>
  5698. <p>Pretty strong language, which should not be a surprise to anyone who
  5699. has been following the ongoing contretemps between McIntyre and Mann.
  5700. But McIntyre is not the only one commenting on Mann's book; Harold
  5701. Ambler
  5702. <a href="">wonders</a>
  5703. how he can actually get his hands on all that funding that Mann claims
  5704. is being doled out by Big Oil to climate skeptics:</p>
  5706. <blockquote>
  5707.  <p>This is all a bit hard to take. I myself am a skeptical blogger and
  5708.  author, yet I am in no way funded by Big Oil. In fact, my
  5709.  three-and-a-half years of toiling on the subject of climate change has
  5710.  yielded approximately $4,000 worth of income. I'm not proud of this fact
  5711.  as a father, husband or man, but it does undercut the constant
  5712.  conspiracy theories about funding behind global-warming skepticism.
  5713.  Meanwhile, as I've noted elsewhere, mainstream climate scientists
  5714.  themselves have received grants totalling more than $1 billion from
  5715.  Exxon Mobil, Shell, BP and other large energy companies.</p>
  5716. </blockquote>
  5718. <p>And, of course, there's the continual stonewalling that's been going on
  5719. for years now at the RealClimate site, as noted by
  5720. <a href="">Anthony Watts</a>:</p>
  5722. <blockquote>
  5723.  <p>Email 2743, Sept 2009, Michael "Robust Debate" Mann: "So far, we've
  5724.  simply deleted all of the attempts by McIntyre and his minions to draw
  5725.  attention to this at RealClimate."</p>
  5726. </blockquote>
  5728. <p>And Mann himself, apart from his book, has
  5729. <a href="">blogged</a>
  5730. defending his position, with plenty of strong language on his own account:</p>
  5732. <blockquote>
  5733.  <p>As a climate scientist, I have seen my integrity perniciously attacked,
  5734.  politicians have demanded I be fired from my job, and I've been subject
  5735.  to congressional and criminal investigations. I've even had death
  5736.  threats made against me. And why? Because I study climate science and
  5737.  some people don't like what my colleagues and I have discovered. Their
  5738.  attacks on scientists are part of a destructive public-relations
  5739.  campaign being waged in a cynical effort to discredit climate science.</p>
  5740. </blockquote>
  5742. <p>My first inclination after collecting the above quotes was to put this
  5743. post in the "rants" section. After all, the battle lines are pretty well
  5744. drawn by this time, aren't they? But no. This wouldn't really be worth a
  5745. post just to rant, and anyway I'd be rather late to the party. And there
  5746. is, actually, an issue worth teasing out from the diatribes and discussing
  5747. on its own merits.</p>
  5749. <p>First, take a look at what seems to me to be the core of Mann's blog post:</p>
  5751. <blockquote>
  5752.  <p>By digging up and burning fossil fuels, humans are releasing much of
  5753.  the carbon that had been buried in the earth over the eons into the
  5754.  atmosphere in the form of carbon dioxide and other gases. Those gases
  5755.  are acting like a heat-trapping blanket around the planet.</p>
  5757.  <p>If we continue down this path, we will be leaving our children and
  5758.  grandchildren a different planet - one with more widespread drought and
  5759.  flooding, greater competition for diminishing water and food resources,
  5760.  and national-security challenges arising from that competition.</p>
  5762.  <p>As a father of a six-year-old daughter, I believe we have an ethical
  5763.  responsibility to make sure that she doesn't look back and ask why we
  5764.  left her generation a fundamentally degraded planet relative to the one
  5765.  we started with.</p>
  5767.  <p>There's a tendency for people to be so overwhelmed by the challenge and
  5768.  the threat of climate change that they go from concern to despair. They
  5769.  shouldn't. While some warming is already locked in, there's still time
  5770.  to turn the ship around. We can still limit our emissions in the decades
  5771.  to come in a way that prevents some of the most serious impacts of
  5772.  climate change from occurring.</p>
  5773. </blockquote>
  5775. <p>Forget all the criticisms and diatribes for a moment. Forget the fact
  5776. that the things Mann states as simple facts in the above are nothing of
  5777. the sort; they are <em>beliefs</em> that he holds, which may or may not be
  5778. justified. The question I want to start with is: does he <em>really</em> hold
  5779. these beliefs? Is he sincere? I think he is, and so are the other
  5780. climate scientists and politicians who preach alarmism about global
  5781. warming; and <em>that</em> is what we should really be worried about.</p>
  5783. <p>Without getting into the whole history of the "hockey stick" and other
  5784. debates that Mann has been involved in, let's just look at Mann's simple
  5785. statements above and apply some basic critical thinking. First: it is
  5786. true that the carbon in fossil fuels was stored there over a very long
  5787. period of time (millions of years during the Carboniferous period, some
  5788. 300 to 360 million years ago, if our current understanding is correct),
  5789. and we are now releasing it over a much shorter period of time. However:
  5790. there have been plenty of periods <em>in between</em> the Carboniferous period
  5791. and now when humans were <em>not</em> burning fossil fuels--so all that carbon
  5792. that's supposedly warming the planet now was <em>not</em> released, but stayed
  5793. buried--and yet the climate was <em>warmer</em> than it is now, sometimes
  5794. quite a bit warmer. What made it warm then, since it obviously wasn't
  5795. all that carbon that was buried?</p>
  5797. <p>Of course, I'm quite sure Mann would have an answer to this if we asked
  5798. him--though he might need some help from the other folks at RealClimate
  5799. to work it into a really good blog post. But the point remains: the
  5800. basic assumption underlying all of this hysteria about global warming is
  5801. that CO2 released by human burning of fossil fuels is the <em>primary driver</em>
  5802. of the climate. But it clearly wasn't in the past, so what's changed now?
  5803. Do the laws of atmospheric physics somehow adjust themselves because they
  5804. know that humans are now burning fossil fuels? Or could it be that there
  5805. are other things that drive the climate? After all, if Mann is correct,
  5806. we are releasing millions of years' worth of carbon over a few hundred
  5807. years, so if all that carbon really made so much of a difference, Earth's
  5808. climate should already be somewhere around where it was at the start of
  5809. the Carboniferous period, right? But it isn't.</p>
  5811. <p>But let's put that aside and go on to the next point. Suppose the Earth
  5812. <em>does</em> warm by another degree Celsius or so by 2100--so what? Mann
  5813. asserts that this will degrade the planet--but hold on a second. The
  5814. planet warmed by 0.6 degrees Celsius from 1900 to 2000, according to the
  5815. IPCC. Do you feel any impulse to look back and ask why those profligates
  5816. in the early 1900's left us such a degraded planet? True, things have
  5817. changed since then: but <em>we have adapted to the changes</em>. Not only that:
  5818. we have <em>innovated</em>. We have found new ways of doing things, and new
  5819. things to do, that the people in 1900 could not have imagined. And as a
  5820. result, the planet is much, much richer today. Yet somehow, all that is
  5821. supposed to stop, and we are just stuck? People won't find any more new
  5822. ways of adapting? That is just silly. I expect that Earth in 2100 will
  5823. be a lot richer than Earth today, and to the people of the future,
  5824. climate change will be a non-problem, not because it won't be happening,
  5825. but because adapting to it will be cheap, the same way that adapting to
  5826. changes in the weather is cheap in developed countries today.</p>
  5828. <p>(An aside: it could well turn out that not just adapting to climate
  5829. change, but <em>controlling</em> it, will be cheap by 2100. That would be even
  5830. nicer, just as being able to control the weather would be nicer than
  5831. having to adapt to it. I can wear a raincoat and carry an umbrella, but
  5832. if things could be arranged so the rain fell while I was sleeping and it
  5833. was always sunny out when I had to go somewhere, I certainly wouldn't
  5834. complain. But the main obstacle in the way of controlling the climate
  5835. is--climate science. As long as Mann and his ilk are running this field,
  5836. we will <em>never</em> really understand how the climate works, because they
  5837. are not trying to understand it; they are trying to force it to conform
  5838. to their predetermined conclusions. But that's another post.)</p>
  5840. <p>Mann talks about people being overwhelmed by the challenge; but it seems
  5841. to me that he and his crew of alarmists are the ones who are
  5842. overwhelmed, and are simply projecting their own feelings onto the rest
  5843. of us. They have a sense of planetary emergency because <em>they</em> can't
  5844. think of any ways to adapt--the only response they can come up with is
  5845. alarm: stop emitting CO2 RIGHT NOW! Well, let me reassure you, Mr. Mann:
  5846. the rest of us have <em>plenty</em> of ways to adapt. It may well turn out that
  5847. we burn a lot less fossil fuel in the future than we do today--but for
  5848. reasons that have little or nothing to do with climate change. Gasoline
  5849. is well over $4 a gallon in the US as I write, and hybrid vehicles are
  5850. selling like hotcakes. Some of those people are probably buying hybrids
  5851. because they're concerned about the climate, but I expect a lot more are
  5852. buying them simply because they want to save money. Or because they're
  5853. concerned about their dollars going to oil-rich countries in the Middle
  5854. East. But regardless of the reason, people do respond to <em>reasonable</em>
  5855. incentives to change their behavior. What they don't respond well to is
  5856. being told that their only choices are to emasculate the economy or
  5857. destroy the planet.</p>
  5859. <p>So I'm not responding to Mann's book by writing the long, long post I
  5860. could write about all the details of what is wrong with Mann's so-called
  5861. research, why the hockey stick is bunk, why the climate models are
  5862. worthless, and so on. (I may write that post anyway sometime, just for
  5863. fun--or perhaps to go more deeply into the underlying issue of how we
  5864. <em>should</em> be doing science--but this isn't it.) None of that really
  5865. matters to the bottom line, which actually ties in nicely with my
  5866. <a href="">last post</a>:</p>
  5868. <blockquote>
  5869.  <p>Humanity has always faced problems, and we always will. The only choice
  5870.  we have is whether we face them with hope or with fear. But there is not
  5871.  only an individual choice, for each of us to make for ourselves; there is
  5872.  also a social choice, whether or not to let the fears of the fearful
  5873.  constrain the hopes of the hopeful.</p>
  5874. </blockquote>
  5876. <p>When it comes right down to it, Michael Mann isn't being investigated
  5877. because he did bad science. Scientists are human, and can make mistakes;
  5878. we all realize that. He's being investigated because he did bad science
  5879. and then used it to justify declaring a planetary emergency. When you do
  5880. that, people take notice, and if it later turns out--after you have
  5881. <a href="">tried your best to obstruct the investigation</a>--that
  5882. you didn't do the science right, people get annoyed. Particularly if,
  5883. even supposing the problem you are worried about <em>is</em> real, there are
  5884. other ways of dealing with it besides pushing the emergency button.
  5885. The real problem with climate change alarmism is that the alarmists
  5886. just can't get this; they just can't get that the whole alarmism thing
  5887. is <em>their</em> personal thing, and most other people just don't share it.
  5888. It's not that we don't want to "save the planet"; of course, everybody
  5889. wants to save the planet. It's just that we don't agree that your
  5890. declaration of a planetary emergency is the way to save the planet.
  5891. It may even be counterproductive, since it involves committing a <em>lot</em>
  5892. of resources that could be put to better use elsewhere.</p>
  5894. <p>So here's my advice to climate change alarmists: get off the soapbox.
  5895. Yes, we know you're concerned, and we appreciate your concern. By all
  5896. means, buy a hybrid car, support alternative energy research (and by
  5897. the way, it would be nice if you would include nuclear power under
  5898. "alternative energy", but that's another post), look for other ways to
  5899. help reduce our use of fossil fuels. There are other good reasons to
  5900. do that anyway. But we are not going to stop everything else and cripple
  5901. the world's economy. You've had your say, and there are plenty of other
  5902. pressing issues to attend to. Deal with it.</p>
  5903. </div>
  5904. </content>
  5905. </entry>
  5907. <entry>
  5908. <title type="html">Discovery Retires</title>
  5909. <category term="/opinions" />
  5910. <id></id>
  5911. <updated>2012-04-20T03:07:00Z</updated>
  5912. <published>2012-04-20T03:07:00Z</published>
  5913. <link rel="alternate" type="text/html" href="" />
  5914. <content type="xhtml">
  5915. <div xmlns="">
  5916. <p>Along with a lot of other people, I watched <em>Discovery</em> fly over Washington,
  5917. DC on its way to Dulles Airport. A good sequence of pictures is
  5918. <a href="">here</a>.
  5919. I lamented
  5920. <a href="">a while back</a>
  5921. that today's NASA seems to have fallen far from the NASA of the Apollo
  5922. missions, but obviously there's not much to be done about that except to
  5923. move on (see below for more on that), and anyway, that's not <em>Discovery</em>'s
  5924. fault. The Shuttle deserves a good retirement, and will get one; I plan
  5925. to go see it in its new home.</p>
  5927. <p>Then I saw news of a new venture called
  5928. <a href="">Planetary Resources</a>,
  5929. backed by James Cameron and the founders of Google, which, if speculations
  5930. are correct, plans to
  5931. <a href="">mine asteroids</a>.
  5932. So as I said in that post a while back, we don't need NASA to go into
  5933. space now; private ventures, at least in the US, will do it on their
  5934. own dime. It will be interesting to see if NASA's
  5935. <a href="">current plans</a>
  5936. get there before a private venture does.</p>
  5938. <p>Many people are probably wondering why a private venture would even get
  5939. into this business. I am sad to note that the magazine closely associated
  5940. with my alma mater, <em>Technology Review</em>, appears to be in this category;
  5941. their
  5942. <a href="">response</a>
  5943. to the press release announcing the venture can only be described as
  5944. snarky:</p>
  5946. <blockquote>
  5947.  <p>According to the company's press release (below):</p>
  5949.  <p>"[...] the company will overlay two critical sectors - space exploration
  5950.  and natural resources - to add trillions of dollars to the global GDP.
  5951.  This innovative start-up will create a new industry and a new definition
  5952.  of 'natural resources'."</p>
  5954.  <p>That sounds like asteroid mining. Because what else is there in space
  5955.  that we need here on earth? Certainly not a livable climate or a replacement
  5956.  for our dwindling supplies of oil.</p>
  5957. </blockquote>
  5959. <p>Regular readers of TR (which I still am, but may not be for much longer if
  5960. they keep going the way they're going) will recognize the "livable climate"
  5961. bit as a reference to global warming; I won't comment on that here since it
  5962. deserves a whole separate post (and will probably get one sometime soon).
  5963. The "dwindling supplies of oil" bit is just as bad; we don't need to go
  5964. into space to find
  5965. <a href="">substitutes</a>
  5966. for oil. (And that's not even considering other alternatives like nuclear
  5967. power--TR does know that MIT has a whole Nuclear Engineering department
  5968. working on that, right?) And even if the venture is intended to be nothing
  5969. more than asteroid mining, at least to start with, there are a <em>lot</em> of
  5970. resources to be mined from asteroids; that "trillions of dollars" part is
  5971. not hyperbole (in fact it is probably an underestimate of the total value
  5972. that will eventually be realized). Does that not count enough in TR's
  5973. calculus of value to merit more than a passing comment before the snark
  5974. begins?</p>
  5976. <p>But the real problem with this attitude is the narrowness of vision, the
  5977. underlying belief that the best way to face our problems is to withdraw
  5978. inside our shell, to scramble as best we can for our share of a limited
  5979. pie, rather than looking for new ways to make more pie. Isaac Asimov wrote
  5980. a story called
  5981. <a href="">The Martian Way</a>
  5982. which showed this kind of contrast. Human settlers on Mars have a problem:
  5983. Earth is reducing shipments of water to Mars, which are crucial for the
  5984. settlement because it needs much more water than it can tap from Mars'
  5985. polar cap. The temptation is there to withdraw, to accept the limitation
  5986. and reduce the settlers' quality of life by restricting their water usage;
  5987. ultimately, it is quite possible that this would mean the settlement would
  5988. have to be abandoned altogether, and its people would have to return to
  5989. Earth to an impoverished existence. But the Martian people choose another
  5990. option: go and get water from the rings of Saturn, which turn out to be
  5991. composed of huge mile-wide chunks of mostly pure water ice (this was the
  5992. actual scientific belief then and still is today). They bring back enough
  5993. to be able to not only meet their own needs but to sell water back to
  5994. Earth in order to help with the water shortage there that forced the
  5995. shipments to be reduced.</p>
  5997. <p>Humanity has always faced problems, and we always will. The only choice
  5998. we have is whether we face them with hope or with fear. But there is not
  5999. only an individual choice, for each of us to make for ourselves; there is
  6000. also a social choice, whether or not to let the fears of the fearful
  6001. constrain the hopes of the hopeful. When we sent astronauts to the Moon,
  6002. we chose not to let that happen; there were plenty of naysayers who said
  6003. it could not be done, and plenty more who said that maybe it could be
  6004. done, but that it was too risky to try. We made the same choice, to a
  6005. lesser extent, when <em>Discovery</em> and the other Shuttles flew. But neither
  6006. of those ventures was supposed to be the end; both were just first steps
  6007. and were meant to be followed by others. And it looks like they will be.
  6008. I'm glad to see that happen. It appears that TR is not, which is fine;
  6009. but at least they should have the decency to keep it to themselves.</p>
  6011. <p>So happy retirement, <em>Discovery</em>, and I hope you'll have some company in
  6012. time, when the museum exhibit opens showing the first retired spacecraft
  6013. that made the run to the asteroids and back.</p>
  6014. </div>
  6015. </content>
  6016. </entry>
  6018. <entry>
  6019. <title type="html">Delimiters Suck</title>
  6020. <category term="/rants" />
  6021. <id></id>
  6022. <updated>2012-03-09T04:28:00Z</updated>
  6023. <published>2012-03-09T04:28:00Z</published>
  6024. <link rel="alternate" type="text/html" href="" />
  6025. <content type="xhtml">
  6026. <div xmlns="">
  6027. <p>A while back I explained
  6028. <a href="">why I use Python, not Lisp</a>.
  6029. However, after reading
  6030. <a href="">this review of Go</a>,
  6031. I realized that I left out something important, something
  6032. that sets Python apart from pretty much <em>every</em> other language
  6033. out there, and certainly from every "C-oid" language, which is
  6034. all that the author of the review seems able to find himself
  6035. wishing for.</p>
  6037. <p>Here it is, in a nutshell: <em>delimiters suck</em>.</p>
  6039. <p>Back in the old days, when CPU cycles were scarce and parsers had
  6040. to be streamlined, it made a kind of sense to make the programmer
  6041. do the lion's share of the work of defining the structure of the
  6042. source code. Curly braces, parentheses, semicolons, and so forth
  6043. helped simplify the process of parsing and compiling code, and if
  6044. that made the difference between a short wait for your code to
  6045. compile and run so you could test it, and having to go out and grab
  6046. coffee (or even lunch) while your code compiled, that was a tradeoff
  6047. worth making.</p>
  6049. <p>But this is 2012, for goodness' sake. Compilation is cheap. Code
  6050. gets compiled on the fly all the time now; after all, web pages are
  6051. code that needs to be compiled every time the page loads. And that
  6052. means that <em>any</em> programming language that makes more work for the
  6053. programmer in the name of making less work for the compiler is simply
  6054. braindead in today's world. And yet that's exactly what programming
  6055. languages with delimiters do. Why should I have to tell the parser
  6056. that, oh, here's a new block; oh, here's the condition for an if
  6057. statement; oh, here's the end of a statement? Why can't it figure
  6058. all that out for itself?</p>
  6060. <p>The answer, of course, is that it can; but our lazy programming
  6061. languages don't <em>require</em> it to. Except one. And that's why Python is
  6062. my favorite language: no delimiters. No curly braces, no parentheses,
  6063. no semicolons. <em>None</em>. (Yes, strictly speaking, there is one delimiter:
  6064. the colon that opens a new indent block. But typing that feels natural,
  6065. since the whole point is that I'm starting a new block, and typing the
  6066. colon <em>feels</em> like I'm just starting the new block, especially since
  6067. I <em>don't</em> need to type any corresponding <em>closing</em> delimiter at the end
  6068. of the block. Still, if Python were to upgrade its parser to eliminate
  6069. the need for the colon, I wouldn't complain.) Python indicates code
  6070. structure the sane way, with indentation. In other words, it uses
  6071. something that every coder does anyway, simply because it's necessary
  6072. to make the code readable. Indentation requires no extra typing; I do
  6073. what I'm best at, writing actual code instead of delimiters, and the
  6074. computer does what it's best at, applying complex but precisely
  6075. specified formal rules much faster and more accurately than a human can.</p>
  6077. <p>And truth be told, even those who claim to prefer other languages know
  6078. in their hearts that I'm right. For example, the reviewer applauds the
  6079. fact that Go doesn't require semicolons--but then says:</p>
  6081. <blockquote>
  6082.  <p>Well, actually there are semicolons, but they are discouraged. It
  6083.  works like JavaScript, there is a simple rule that makes the parser
  6084.  insert a semicolon at certain line ends.</p>
  6085. </blockquote>
  6087. <p>In other words, the Go designers know that the parser is perfectly
  6088. capable of spotting the end of a logical statement without requiring
  6089. a delimiter. Then why isn't the delimiter <em>eliminated</em>, instead of
  6090. just being "discouraged"?</p>
  6092. <p>But wait; it gets better. The next comment we get is this:</p>
  6094. <blockquote>
  6095.  <p>Next in category "pure heresy": Go defines a canonical indentation
  6096.  and the One True Bracing Style.</p>
  6097. </blockquote>
  6099. <p>In other words, the Go designers also know that indentation is
  6100. important; yet they still cling to the delusion that somehow the
  6101. indentation needs help from curly braces. But if you're going to
  6102. define a One True Bracing Style, why can't the parser simply <em>deduce</em>
  6103. where the braces would go, the same way it can deduce where semicolons
  6104. would go? Well, the reviewer says this about the canonical indentation
  6105. style:</p>
  6107. <blockquote>
  6108.  <p>Like Python, only I think python has a tad too little visual cues.
  6109.  Indentation alone isn't always sufficiently clear, so we get to keep
  6110.  our beloved braces.</p>
  6111. </blockquote>
  6113. <p>Oh, so the braces are to help the <em>programmer</em> know where the ends
  6114. of blocks are? Strange how so many people are able to program in
  6115. Python perfectly well without this crutch. Of course, if you don't
  6116. <em>have</em> a canonical indentation style, you can't depend on it as a
  6117. cue, so you may have problems with indentation being "sufficiently
  6118. clear". But the reviewer just told us that Go <em>does</em> have a canonical
  6119. indentation style! I think he hasn't fully grasped the implications.</p>
  6121. <p>(An aside here: the Go designers are making a huge mistake with the
  6122. canonical indentation, by using <em>tabs</em> instead of spaces. Of course,
  6123. all Pythonistas know that indenting with tabs is just a straight road
  6124. to perdition. See
  6125. <a href="">here</a>
  6126. and
  6127. <a href="">here</a>
  6128. for the details.)</p>
  6130. <p>There's another funny bit in the next item in the review. We read
  6131. this...</p>
  6133. <blockquote>
  6134.  <p>Speaking of braces, there are no brace-free forms of if and loops.</p>
  6135. </blockquote>
  6137. <p>...followed almost immediately by this:</p>
  6139. <blockquote>
  6140.  <p>One important technical reason for the previous point is the fact
  6141.  that those control statements no longer have parens.</p>
  6142. </blockquote>
  6144. <p>In other words, take away parentheses around conditionals (good),
  6145. but then use that as an excuse to <em>keep</em> curly braces (braindead).
  6146. Give with one hand and take away with the other. But wait; there's
  6147. a reason:</p>
  6149. <blockquote>
  6150.  <p>Only Perl6 tries to parse paren-less if without curlies, and we all
  6151.  know how complicated Perl parsers used to be (and obviously, still are).</p>
  6152. </blockquote>
  6154. <p>I guess the reviewer forgot that <em>Python</em> has no trouble parsing
  6155. paren-less if without curlies. And without forcing you to use Perl and
  6156. its parser, either.</p>
  6158. <h1>Static Typing Sucks Too</h1>
  6160. <p>Since we're on the subject of Python vs. "C-oid" languages, I'll go
  6161. ahead and rant about variable typing as well. The reviewer notes that
  6162. Go is smart enough to use duck typing, as Python does, instead of
  6163. requiring explicit interface declarations:</p>
  6165. <blockquote>
  6166.  <p>Go uses interfaces exclusively. Unlike Java, however, you don't
  6167.  declare that a given type conforms to some interface. If it does, it
  6168.  automatically is usable as that interface.</p>
  6169. </blockquote>
  6171. <p>Great! Go objects are duck-typable. But that raises the obvious next
  6172. question: what about <em>references</em> to objects, i.e., variables? Well,
  6173. that's better than C, in that variable declarations don't require
  6174. explicit type specifiers:</p>
  6176. <blockquote>
  6177.  <p>If you leave out the type, the type is instead taken from the assignment.
  6178.  You may even leave out the declaration altogether by using the new
  6179.  declare-and-initialize operator.</p>
  6180. </blockquote>
  6182. <p>Okay, good. But once you've done that once, you're stuck; the new
  6183. variable declaration syntax</p>
  6185. <blockquote>
  6186.  <p>does not introduce dynamic typing, it does not allow changing a
  6187.  declared variable's type, it does not remove the need to declare variables,
  6188.  it does not allow you to declare variables twice. It's really the same as
  6189.  before, semantically, but much lighter on syntax. Feels like an ad-hoc
  6190.  scripting language but still gives you the benefits of static typing.</p>
  6191. </blockquote>
  6193. <p>Um, <em>what</em> benefits of static typing? <em>Objects</em> are already typed; once
  6194. you create an object, its type is fixed forever. That's true in every
  6195. language, including Python; I'm not sure I can see how you could implement
  6196. an object system at all without it. That, plus duck typing, ensures that
  6197. if you try to do anything with an object that it doesn't support, the
  6198. language will tell you. If you want anything more draconian than that,
  6199. you don't want a language like Go, or Python, or C for that matter. You
  6200. want something like Visual Basic, because you're going to let people use
  6201. it who can't be trusted with fully capable tools.</p>
  6203. <p>But given that you're cool with duck typing, whyinhell would you want to
  6204. restrict variable bindings? When you reassign a variable, you aren't
  6205. changing the type of its object; you are binding it to a <em>new</em> object,
  6206. which has been freshly created, and of course when you write the code
  6207. that creates it, you have already decided what type you want it to be,
  6208. just like when you bound the variable to its original value. Sure, the
  6209. vast majority of the time, the new object will have the same type as the
  6210. old; but what about those cases where you don't <em>want</em> it to? What if,
  6211. for example, you want to assign a new object that is duck-type compatible
  6212. with the old one, but has no common base type with the old one, so the
  6213. type system can't tell they're compatible? Again, if you want your type
  6214. system to prevent you from doing this, you don't want a fully capable
  6215. language in the first place.</p>
  6217. <p>I could go on about other stuff, but I think I've said enough to make it
  6218. clear, once again, why Python is my favorite language. Frankly, the only
  6219. advantage I can see with Go over Python (and not just based on this
  6220. review; I've been following Go's progress for several years now) is speed;
  6221. but Moore's Law and projects like
  6222. <a href="">PyPy</a>
  6223. are making that less and less of an issue every day. It's great that
  6224. people keep on trying new things with programming languages, but at least
  6225. this time, I'm sticking with Python.</p>
  6227. <p>(Update: there is a discussion of this post on
  6228. <a href="">Hacker News</a>.)</p>
  6229. </div>
  6230. </content>
  6231. </entry>
  6233. <entry>
  6234. <title type="html">One Rant Deserves Another</title>
  6235. <category term="/rants" />
  6236. <id></id>
  6237. <updated>2012-03-04T01:50:00Z</updated>
  6238. <published>2012-03-04T01:50:00Z</published>
  6239. <link rel="alternate" type="text/html" href="" />
  6240. <content type="xhtml">
  6241. <div xmlns="">
  6242. <p>Cary Sherman, the CEO of the RIAA, is upset. He says those mean and nasty
  6243. Internet companies shut down SOPA and PIPA by spreading misinformation
  6244. and claiming it was fact. Well, after reading his recent
  6245. <a href="">op-ed in the New York Times</a>,
  6246. I will certainly concede that Mr. Sherman ought to know about that sort
  6247. of thing, since he is evidently an expert at it. Just for fun, I
  6248. thought I would post some examples.</p>
  6250. <blockquote>
  6251.  <p>Since when is it censorship to shut down an operation that an American
  6252.  court, upon a thorough review of evidence, has determined to be illegal?</p>
  6253. </blockquote>
  6255. <p>Oh, is that all Mr. Sherman wants to do? He can do that now, under existing
  6256. law, and his organization certainly hasn't been shy about it. In fact, he
  6257. hasn't even been shy about shutting down operations <em>without</em> going through
  6258. all the hassle of taking them to court or getting a review of the evidence,
  6259. but simply on his say-so. One wonders why he needs SOPA and PIPA in the
  6260. first place. Whatever happened to "innocent until proven guilty"?</p>
  6262. <blockquote>
  6263.  <p>As it happens, the television networks that actively supported SOPA and
  6264.  PIPA didn't take advantage of their broadcast credibility to press their
  6265.  case. That's partly because "old media" draws a line between "news" and
  6266.  "editorial." Apparently, Wikipedia and Google don't recognize the
  6267.  ethical boundary between the neutral reporting of information and the
  6268.  presentation of editorial opinion as fact.</p>
  6269. </blockquote>
  6271. <p>What? Google has editorials plastered all over its search page? I must
  6272. have missed it. Wikipedia has "SOPA is bad" banners at the top of every
  6273. article? I guess I need to get my eyes checked because I didn't see them.
  6274. Or maybe Google and Wikipedia said negative things about SOPA/PIPA on
  6275. their <em>blogs</em> or their <em>editorial pages</em>--you know, the places which are
  6276. clearly marked as expressing their <em>opinions</em>, not facts.</p>
  6278. <p>Of course, Mr. Sherman would certainly like to have <em>his</em> opinion
  6279. accepted as fact. In fact, it's hard to tell whether or not he even
  6280. knows the difference. He writes an "opinion" piece and gets it published
  6281. on the op-ed page, where you are supposed to give your opinions, and then
  6282. states his opinions as facts and hopes nobody notices. Hmm.</p>
  6284. <p>Perhaps the problem is that Mr. Sherman has some vocabulary issues. For
  6285. example:</p>
  6287. <blockquote>
  6288.  <p>Policy makers had recognized a constitutional (and economic) imperative
  6289.  to protect American property from theft, to shield consumers from
  6290.  counterfeit products and fraud, and to combat foreign criminals who
  6291.  exploit technology to steal American ingenuity and jobs.</p>
  6292. </blockquote>
  6294. <p>By "American property" he means "the money RIAA companies make by
  6295. exploiting the work of artists", not "the actual work done by the
  6296. artists, which is what people actually want to listen to".</p>
  6298. <blockquote>
  6299.  <p>They knew that music sales in the United States are less than half of
  6300.  what they were in 1999, when the file-sharing site Napster emerged,</p>
  6301. </blockquote>
  6303. <p>By "music sales" he means "sales of CDs", not "sales of music". In other
  6304. words, he means "sales of stuff I want to sell because I'm too lazy to
  6305. actually give customers what they want".</p>
  6307. <blockquote>
  6308.  <p>and that direct employment in the industry had fallen by more than half
  6309.  since then, to less than 10,000.</p>
  6310. </blockquote>
  6312. <p>By "the industry" he means "companies which are members of the RIAA". He
  6313. does <em>not</em> mean "anybody who actually makes music".</p>
  6315. <blockquote>
  6316.  <p>They studied the problem in all its dimensions, through multiple
  6317.  hearings.</p>
  6318. </blockquote>
  6320. <p>None of which included the people who actually make the Internet work.
  6321. In fact, as I noted
  6322. <a href="">in a previous post</a>,
  6323. several technical experts were supposed to testify before Congress, the
  6324. first opportunity any such experts had had to do so, on January 18th;
  6325. but there's no indication that the hearing ever actually happened.</p>
  6327. <p>But perhaps the real problem is something else. Mr. Sherman lets slip
  6328. an interesting comment towards the end of his article:</p>
  6330. <blockquote>
  6331.  <p>The conventional wisdom is that the defeat of these bills shows the
  6332.  power of the digital commons. Sure, anybody could click on a link or
  6333.  tweet in outrage - but how many knew what they were supporting or
  6334.  opposing?</p>
  6335. </blockquote>
  6337. <p>Good question. How many legislators who initially said they supported
  6338. SOPA/PIPA knew what they were supporting? How many who switched to
  6339. opposing it did so because they actually <em>read</em> the bills, so they <em>did</em>
  6340. know what they were opposing?</p>
  6342. <p>The root of the problem is that Mr. Sherman and the rest of the "old
  6343. media" simply don't understand what's happened. They don't understand
  6344. that the reason they're having all these problems is that the Internet
  6345. has killed their business model. They think it must be some evil plot
  6346. by "hackers" and "pirates", because surely people wouldn't just stop
  6347. buying CDs because, say, they wanted more convenience and knew there
  6348. were ways to get it? Naw, that couldn't be it.</p>
  6350. <blockquote>
  6351.  <p>No doubt, some genuinely wanted to protect Americans against theft but
  6352.  were sincerely concerned about how the language in the bill might be
  6353.  interpreted. But others may simply believe that online music, books and
  6354.  movies should be free.</p>
  6355. </blockquote>
  6357. <p>Or we may believe that if we are going to pay money for music, books,
  6358. and movies, we should get something worth paying that money for, not
  6359. something that's been emasculated and micromanaged to the point where
  6360. it's more trouble than it's worth. And that our money should go to the
  6361. people that actually create, not the people that exploit them. And that
  6362. we should not be treated like potential thieves when all we want to do
  6363. is listen to music, read books, or watch movies.</p>
  6364. </div>
  6365. </content>
  6366. </entry>
  6368. <entry>
  6369. <title type="html">More From The Internet Front</title>
  6370. <category term="/opinions" />
  6371. <id></id>
  6372. <updated>2012-02-25T05:56:00Z</updated>
  6373. <published>2012-02-25T05:56:00Z</published>
  6374. <link rel="alternate" type="text/html" href="" />
  6375. <content type="xhtml">
  6376. <div xmlns="">
  6377. <p>Eric Raymond has published an
  6378. <a href="">Open Letter to Chris Dodd</a>
  6379. in response to Dodd's
  6380. <a href="">recent speech</a>.
  6381. Any comments from me would be superfluous (and if you've read my previous
  6382. posts on this subject you'll know where I'm coming from anyway); just read
  6383. Eric's post. It's worth it.</p>
  6384. </div>
  6385. </content>
  6386. </entry>
  6388. <entry>
  6389. <title type="html">Internet Blackout Day</title>
  6390. <category term="/opinions" />
  6391. <id></id>
  6392. <updated>2012-01-20T00:56:00Z</updated>
  6393. <published>2012-01-20T00:56:00Z</published>
  6394. <link rel="alternate" type="text/html" href="" />
  6395. <content type="xhtml">
  6396. <div xmlns="">
  6397. <p>Many sites on the Internet (including this blog) were "blacked out"
  6398. yesterday as part of a protest against SOPA/PIPA. Opposition to these
  6399. bills has been mounting for some time, and a few days ago it appeared
  6400. that SOPA, at least, had been
  6401. <a href="">shelved by the House</a>,
  6402. but it turned out that it had only been
  6403. <a href="">delayed until February</a>.
  6404. (Even if does eventually get "shelved", the cynical read on that, which
  6405. would be mine, will be that Congress is simply taking it off the radar,
  6406. knowing how short the public's attention span is, and will try to slip
  6407. it in later as an amendment to some other bill that is expected to
  6408. pass without much scrutiny. It certainly wouldn't be the first time.)</p>
  6410. <p>CNN ran
  6411. <a href="">an article</a>
  6412. about the blackout which summarized the position of
  6413. SOPA's supporters:</p>
  6415. <blockquote>
  6416.  <p>SOPA's supporters -- including CNN parent company Time Warner and groups
  6417.  such as the MPAA -- say that online piracy leads to U.S. job losses
  6418.  because it deprives content creators of income.</p>
  6420.  <p>The bill's supporters dismiss accusations of censorship, saying the
  6421.  legislation is meant to revamp a broken system that doesn't adequately
  6422.  prevent criminal behavior.</p>
  6423. </blockquote>
  6425. <p>The MPAA, as quoted in
  6426. <a href="">Ars Technica</a>,
  6427. was a bit more vituperative about the blackout:</p>
  6429. <blockquote>
  6430.  <p>Only days after the White House and chief sponsors of the legislation
  6431.  responded to the major concern expressed by opponents and then called
  6432.  for all parties to work cooperatively together, some technology business
  6433.  interests are resorting to stunts that punish their users or turn them
  6434.  into their corporate pawns, rather than coming to the table to find
  6435.  solutions to a problem that all now seem to agree is very real and
  6436.  damaging.</p>
  6438.  <p>It is an irresponsible response and a disservice to people who rely on
  6439.  them for information and use their services. It is also an abuse of
  6440.  power given the freedoms these companies enjoy in the marketplace today.
  6441.  It's a dangerous and troubling development when the platforms that serve
  6442.  as gateways to information intentionally skew the facts to incite their
  6443.  users in order to further their corporate interests.</p>
  6444. </blockquote>
  6446. <p>After reading this, I was tempted to add another post to the "rants"
  6447. section of this blog. But I'll resist the temptation, because there
  6448. actually is a serious issue here that deserves some non-ranting
  6449. discussion. (It's worth noting, though, that Google's service was <em>not</em>
  6450. unavailable during the blackout, nor were most others--though Wikipedia
  6451. was. So there actually wasn't any "disservice to people who rely on
  6452. them." But I'll pass over that.)</p>
  6454. <p>The MPAA speaks of "the freedoms that these companies enjoy in the
  6455. marketplace today." What freedoms, exactly, are they referring to? Google
  6456. gets the lion's share of search traffic because it's the best search
  6457. engine on the web. So they certainly enjoy the freedom to provide an
  6458. excellent service and have users choose to use it. But they don't have
  6459. the freedom to force people to use their service, much less to make
  6460. people believe their version of the facts. People are protesting because
  6461. they rightly feel outrage at attempts to emasculate the Internet, not
  6462. because Google or anyone else is "inciting" them. I would protest just
  6463. as loudly, as I'm sure would many others who protested yesterday, if
  6464. Google went to the government to try to get laws passed to favor their
  6465. business model over others.</p>
  6467. <p>The simple fact is that the business model on which the corporations that
  6468. make up the MPAA and RIAA were built is dead. The Internet killed it.
  6469. It's not a question of "piracy"; it's a simple question of efficiency.
  6470. Why should customers pay some company to make millions of copies of a
  6471. plastic disk when the same information can be transmitted essentially
  6472. for free over the Internet? It's not as though there are no new business
  6473. models to replace the old ones; Netflix and iTunes are proof of that.
  6474. In fact, the MPAA and the RIAA are up against that very freedom of the
  6475. marketplace that they talk about so blithely: the freedom of we, the
  6476. customers, to choose what we will and will not pay for.</p>
  6478. <p>But what about all those jobs that are being threatened, and all those
  6479. content creators who are being deprived of income? This was the part
  6480. that really tempted me to write a rant, because in fact the MPAA and
  6481. the RIAA do <em>not</em> create content. Artists, and writers, and musicians,
  6482. and actors, and directors, and so forth, create content. The
  6483. corporations that make up the MPAA and the RIAA <em>distribute</em> content.
  6484. That's all they do. The Internet has not stopped content creators; on
  6485. the contrary, content creators are <em>empowered</em> by the Internet. I would
  6486. not be publishing this blog if I had to make photocopies of every post
  6487. and send them out by mail.</p>
  6489. <p>Of course, I don't expect to make money from this blog. What about people
  6490. whose livelihood is creating content? The MPAA and RIAA want you to
  6491. believe that these people are worse off because of "piracy". But what
  6492. do the content creators themselves say? You'll note that the MPAA and
  6493. RIAA most carefully don't quote them. That's because they would have to
  6494. quote things like
  6495. <a href="">this</a>,
  6496. which would start people thinking about the way these corporations
  6497. actually treat artists, as shown for example
  6498. <a href="">here</a>.
  6499. The truth is that the MPAA and RIAA do far more to deprive content
  6500. creators of income than any amount of "piracy" could ever do.</p>
  6502. <p>(It's worth noting in this connection that the Hollywood "guilds",
  6503. including the Screen Actors Guild,
  6504. <a href="">sent a letter supporting SOPA</a>
  6505. to several members of Congress yesterday, to coincide with the
  6506. Internet blackout. More on that below.)</p>
  6508. <p>The opposition to these bills does appear to have had some effect
  6509. beyond provoking the predictable responses discussed above. For
  6510. example, the MPAA appears to have
  6511. <a href="">backed off on DNS filtering</a>
  6512. (this may have been the "major concern" referred to in the MPAA quote
  6513. from the Ars Technica article early in this post), although the
  6514. article notes that this may not last. However, they also make this
  6515. interesting statement:</p>
  6517. <blockquote>
  6518.  <p>"The future of our industry relies on the Internet," [the MPAA's
  6519.  tech chief] said, noting that movie studios were increasingly selling
  6520.  their products to consumers via the Internet.</p>
  6521. </blockquote>
  6523. <p>I'm not sure exactly what this is supposed to mean. If it just means
  6524. that I can buy DVDs from Amazon instead of at the store, well, yes, I
  6525. suppose it's true. But if there's a movie studio out there that is
  6526. running a site like Netflix that streams video direct to people without
  6527. making them jump through hoops, it's a very well-kept secret. Or
  6528. perhaps the quote is referring to Netflix itself; but even here the
  6529. studios appear to want to restrict the channel, as
  6530. <a href="">CNN Money</a>
  6531. noted in an article some time ago:</p>
  6533. <blockquote>
  6534.  <p>Netflix subscribers got a taste of the studios' new hardball approach
  6535.  last month, when hundreds of Sony (SNE) movies -- including high-profile
  6536.  titles like "The Social Network" and "Salt" -- abruptly vanished from
  6537.  Netflix's "watch now" catalog.</p>
  6539.  <p>In a
  6540.  <a href="">blog post</a>,
  6541.  Netflix pinned the blame on a "temporary contract issue" between Sony and
  6542.  Starz, a pay cable network that licenses Sony's movie catalog. Back in
  6543.  2008, Netflix struck a four-year deal with Starz that gave it streaming
  6544.  access to Starz' offerings.</p>
  6546.  <p>But Starz' deal with Sony included a cap on the number of subscribers
  6547.  who can watch Sony movies online, a source told the LA Times. Once
  6548.  Netflix' audience exceeded the cap, the contract was null. Starz'
  6549.  catalog of Disney movies available for online streaming is on the verge
  6550.  of triggering a similar contractual cap, the newspaper reported. </p>
  6551. </blockquote>
  6553. <p>So it really looks like the MPAA and RIAA simply don't understand what
  6554. "selling via the Internet" actually means. More evidence of this is
  6555. found in the MPAA's own
  6556. <a href="">blog post</a>
  6557. in response to a statement by the Obama Administration on anti-piracy
  6558. legislation:</p>
  6560. <blockquote>
  6561.  <p>We also share the Administration's desire to encourage innovation.
  6562.  The American businesses that are victimized on a daily basis by global
  6563.  Internet thieves are among the most innovative industries in this nation
  6564.  and we welcome the Administration's support of these American businesses.</p>
  6565. </blockquote>
  6567. <p>So the "most innovative industries in this nation", according to the
  6568. MPAA, are industries that want to ship movies on plastic disks instead
  6569. of as bits over the Internet, and want to restrict the ways that
  6570. legitimate buyers of their products can use them (by, for example,
  6571. region-encoding DVDs and restricting viewing of movies online), while
  6572. companies like Google that have changed the way the entire world
  6573. searches for information are just trying to take away American jobs
  6574. by enabling "piracy":</p>
  6576. <blockquote>
  6577.  <p>Every day, American jobs are threatened by thieves from foreign-based
  6578.  rogue websites.</p>
  6579. </blockquote>
  6581. <p>Which jobs? Later on, the post says "the 2.2 million Americans whose jobs
  6582. depend on the film and television industries". But how many of those
  6583. people are actually involved in <em>creating</em> content, as opposed to
  6584. distributing it? And how many of them actually have a decent share in
  6585. the profits? Of course the MPAA isn't sharing that information. Not only
  6586. that, but how many of those jobs are actually threatened by "piracy", as
  6587. opposed to being threatened by the inability of the very corporations
  6588. who make up the MPAA to join the rest of us in the 21st century? Netflix
  6589. and iTunes aren't stopping movies and music from being made.</p>
  6591. <p>This is why I think that organizations like the Screen Actors Guild are
  6592. mistaken in supporting legislation like SOPA. They do so because they
  6593. see their jobs being threatened if the system put in place by the MPAA
  6594. and RIAA for making and distributing films and music is threatened. I
  6595. understand their concern, but I think they're making a huge mistake by
  6596. hitching their fates to the fates of organizations as hidebound as the
  6597. MPAA and RIAA. I would be <em>more</em> likely to pay to see movies if I knew
  6598. that my money was going to the people who actually make them instead of
  6599. Sony and the other media corporations (and the few high-profile actors
  6600. who can command huge fees--not that I mind the fees, but they are
  6601. <a href="">not representative</a>
  6602. of what most of the people who make movies or TV shows earn). As I
  6603. said above, the actual content creators are not being well served by
  6604. the existing system, and they ought to seriously consider ditching it,
  6605. regardless of what the MPAA and RIAA think.</p>
  6607. <p>And for all of us as Internet users, I don't expect these bills to go
  6608. away. The corporations whose business models are dead still have a lot
  6609. of money, and they will continue to spend it to try to put the Internet
  6610. genie back in the bottle. We should not let them.</p>
  6612. <h1>Postscript</h1>
  6614. <p>I said I wouldn't make this post into a rant, but
  6615. <a href="">the latest from Hollywood</a>
  6616. is just too good to pass up. Apparently the "medial moguls" are not
  6617. pleased that (in what they see as a drastic turnaround from the
  6618. previous Administration statement that I referred to above) President
  6619. Obama has taken a stand against SOPA:</p>
  6621. <blockquote>
  6622.  <p>The moguls are reminding Obama et al that, in the words of one studio
  6623.  chief, "God knows how much money we've given to Obama and the Democrats
  6624.  and yet they're not supporting our interests. There's been no greater
  6625.  supporters of him than we've been from the first day and the first
  6626.  fundraisers continuing until he was elected. We all were pleased. And,
  6627.  at its heart institutionally, Hollywood supports the Democrats. Now we
  6628.  need the administration to support us."</p>
  6629. </blockquote>
  6631. <p>In other words, as a number of commenters have noted, Hollywood is
  6632. accusing President Obama of failing the classic test of an honest
  6633. politician: he's been bought, but he won't <em>stay</em> bought.</p>
  6635. <p>But there actually is a more serious nugget here:</p>
  6637. <blockquote>
  6638.  <p>"The issue at hand -- piracy -- is a legitimate concern. But Google
  6639.  and those Internet guys have been swiftboating the entertainment
  6640.  industry by saying we're trying to shut down the Internet just because
  6641.  we don't want them to advertise pirated movies."</p>
  6642. </blockquote>
  6644. <p>I'll be charitable and grant that the "studio chief" quoted here sincerely
  6645. believes that he is not trying to shut down the Internet by trying to get
  6646. SOPA and similar legislation passed. He's not a technical expert. But his
  6647. sincere beliefs don't count; what counts is the actual effect that such
  6648. legislation would have if passed. He appears to think that "Google and
  6649. those Internet guys" could quite easily stop advertising pirated movies
  6650. if they wanted to. Perhaps he thinks that Larry Page and Sergey Brin sit
  6651. in a big room and supervise a bunch of people who assiduously review every
  6652. ad by hand. Chris Dodd displays similar ignorance
  6653. <a href="">here</a>:</p>
  6655. <blockquote>
  6656.  <p>"When the Chinese told Google that they had to block sites or they
  6657.  couldn't do [business] in their country, they managed to figure out
  6658.  how to block sites."</p>
  6659. </blockquote>
  6661. <p>No, they agreed to shut down the pathway that had allowed people in
  6662. China to access Google without going through the Chinese government's
  6663. firewall, since they knew the Chinese government was going to shut it
  6664. down anyway. They most certainly did <em>not</em> put in place any kind of
  6665. system that could block individual sites.</p>
  6667. <p>Actually, of course, it would be impossible for Google, or any other
  6668. Internet site of any size that allows users to post content, to police
  6669. everything that they are hosting and still provide the immensely
  6670. valuable services they provide. There is simply too much content and not
  6671. enough time for humans to review even a tiny fraction of it; the
  6672. processes have to be automated if they are going to work at all. And how
  6673. can we expect computers to reliably tell "pirated" content from the rest,
  6674. when humans often have to go through many levels of the judicial system to
  6675. get a decision on the question?</p>
  6677. <p>Not to mention that, once you give people a way to shut down websites by
  6678. claiming they are "rogue" sites, this ability will be abused, and the
  6679. collateral damage from such abuse will be far worse than any possible
  6680. damage from "piracy". I've mentioned this
  6681. <a href="">once before</a>,
  6682. but there are plenty of other examples, such as
  6683. <a href="">this</a>
  6684. and
  6685. <a href="">this</a>.
  6686. The first article talks about Universal Music's claims that popular "hip
  6687. hop" sites and online magazines covering that music genre are "rogue"
  6688. sites; so much for the "media moguls" being concerned about the actual
  6689. "content creators", who, of course, <em>want</em> their music to be heard and
  6690. talked about on these sites. The second talks about Monster Cable's
  6691. claims that Costco and Sears, among others, are "rogue" sites.</p>
  6693. <p>So this really is a binary choice. Either we have valuable Internet
  6694. services like the ones we have, or we have a locked-down system where
  6695. everything else is sacrificed to the goal of stopping "piracy". And
  6696. even then the goal won't be achieved. Mr. Dodd and the MPAA would like
  6697. the US to be more like China in the way it controls the internet. But
  6698. the
  6699. <a href="">International Intellectual Property Alliance</a>,
  6700. which counts among its members--wait for it--the MPAA and RIAA,
  6701. is continually
  6702. <a href="">agitating</a>
  6703. about how much piracy there is in China. Of course, this may be as
  6704. much a matter of the Chinese government not caring all that much
  6705. about piracy as anything else. But that only underscores the point:
  6706. Americans <em>do</em> care. As I noted in
  6707. <a href="">my last post</a>,
  6708. we don't <em>want</em> to steal. But we also don't want to pay for outmoded
  6709. products, and we don't want to put up with inconvenience and being
  6710. treated like potential pirates when we just want to watch a movie or
  6711. listen to a song. Does that justify stealing? No. But the alternative
  6712. is not for us to start buying what the MPAA and RIAA would like to
  6713. sell us; the alternative is for us to spend our money somewhere else
  6714. entirely. Trying to buy legislation is going to make that <em>more</em> likely,
  6715. if anything, not less. Maybe the "media moguls" should stop trying to
  6716. emasculate the Internet, and start working on giving their customers
  6717. what they actually want.</p>
  6719. <h1>Post-Postscript</h1>
  6721. <p>The phenomenon of corporations seeking to buy legislation to prop up
  6722. dead business models is not limited to the arts. A number of companies,
  6723. notably Elsevier, have for years had a sweetheart deal with the US
  6724. government allowing them strict publishing rights for scientific journals.
  6725. Now the Internet has killed that business model too, but the companies
  6726. don't want it to die, as
  6727. <a href="">PZ Myers reports</a>:</p>
  6729. <blockquote>
  6730.  <p>Along with SOPA and PIPA, our government is contemplating another
  6731.  acronym with deplorable consequences for the free dissemination of
  6732.  information: RWA, the
  6733.  <a href="">Research Works Act</a>.
  6734.  This is a bill to, it says, "ensure the continued publication and
  6735.  integrity of peer-reviewed research works by the private sector", where
  6736.  the important phrase is "private sector" -- its purpose is to guarantee
  6737.  that for-profit corporations retain control over the publication of
  6738.  scientific information...</p>
  6740.  <p>This is a blatant attempt to invalidate the NIH's requirement that
  6741.  taxpayer-funded research be made publicly available. The internet was
  6742.  initially developed to allow researchers to easily share information...and
  6743.  that's precisely the function this bill is intended to cripple.</p>
  6744. </blockquote>
  6746. <p>PZ is a biologist, which is why he refers specifically to the NIH, but
  6747. the same problem exists in other scientific fields. However, the physics
  6748. and math communities got tired of this years ago and started
  6749. <a href="">arXiv</a>,
  6750. a site where preprints of scientific papers are made freely available.
  6751. As you can see from the site's home page, other scientific fields have
  6752. joined this effort as well. Papers on the arXiv can still be (and often
  6753. are) published in journals, but until they are actually submitted, the
  6754. scientists who write them have the publication rights, and they routinely
  6755. publish preprints; in fact, they have to in order to enable the very
  6756. "peer review" process that the journals claim to be facilitating. In
  6757. the days before the Internet, preprints were circulated by mail, but
  6758. of course the Internet makes it all much, much easier, just as it was
  6759. intended to do (as PZ notes).</p>
  6761. <p>In other words, the very justification for the US government giving the
  6762. sweetheart deals to journals in the first place, to "ensure the continued
  6763. publication and integrity of peer-reviewed research works", is no longer
  6764. valid. Scientists no longer <em>need</em> these companies to help them share
  6765. information, because of the Internet. The companies' response: try to
  6766. buy legislation. Of course the journals' business model is in even worse
  6767. shape than those of the film and record companies, because the latter
  6768. can at least offer some additional value to customers in the form of
  6769. particular actors or brands. The journals are pure middlemen; the <em>only</em>
  6770. value they added to the process was distribution. (They claim that they
  6771. also added value to the peer review process, by organizing and vetting
  6772. editors and reviewers, but even before the arXiv site was stood up, the
  6773. process had long been shifting in the direction of more and more review
  6774. of preprints and less and less review of actual journal submissions. Now
  6775. of course, the first question anyone, even a journal reviewer, asks
  6776. about a paper is "is a copy on arxiv?" There is even a
  6777. <a href="">proposal</a>
  6778. to formalize this process by allowing signed reviews of papers to be
  6779. linked to the papers themselves.)</p>
  6781. <p>All of this just reinforces the fact that <em>any</em> attempt on the part of
  6782. government to interfere with things is fraught with risk. The
  6783. interventions are usually reasonable at the time: in their day, the
  6784. journals really did facilitate a lot of sharing of scientific
  6785. information. But the interventions always end up long outliving their
  6786. usefulness, and in the process they may do harm that more than outweighs
  6787. the good they did originally. It's up to us, the people, to try to keep
  6788. that from happening.
  6789. <a href="">Eternal vigilance is the price of liberty</a>.</p>
  6790. </div>
  6791. </content>
  6792. </entry>
  6794. <entry>
  6795. <title type="html">The Latest From The SOPA Front</title>
  6796. <category term="/rants" />
  6797. <id></id>
  6798. <updated>2012-01-10T03:58:00Z</updated>
  6799. <published>2012-01-10T03:58:00Z</published>
  6800. <link rel="alternate" type="text/html" href="" />
  6801. <content type="xhtml">
  6802. <div xmlns="">
  6803. <p>This is just a quick update to
  6804. <a href="">my</a>
  6805. <a href="">previous</a>
  6806. <a href="">posts</a>
  6807. <a href="">on</a>
  6808. <a href="">SOPA</a>
  6809. to collect a few more links of interest.</p>
  6811. <p>First, a "mainstream" media channel (CNN) is now at least
  6812. <a href="">covering the issue</a>.
  6813. No surprises in the article, but at least it means the issue is getting
  6814. some attention.</p>
  6816. <p>Next, The Register has posted about a
  6817. <a href="">study</a>
  6818. that finds that, first, the vast majority of people prefer to obtain
  6819. content legally (no surprise for anyone who has heard of Netflix or
  6820. iTunes, but it seems like media companies still haven't gotten the
  6821. memo), and second:</p>
  6823. <blockquote>
  6824.  <p>When it comes to the penalties for piracy the American public is a
  6825.  lot more forgiving than the courts. Three quarters of those surveyed
  6826.  felt that fines of less than $100 per song were acceptable and only 16
  6827.  per cent felt that cutting off internet access was justified to stop
  6828.  piracy. Only a quarter who approved of disconnection felt that more
  6829.  than a one month ban was warranted.</p>
  6830. </blockquote>
  6832. <p>Just in case anyone was still wondering whether SOPA and similar
  6833. legislation actually represents what the people want, here's your
  6834. sign: it doesn't.</p>
  6836. <p>Finally, it appears that one of the founders of
  6837. <a href="">Reddit</a>,
  6838. the CEO of
  6839. <a href="">Rackspace</a>,
  6840. and
  6841. <a href="">Dan Kaminsky</a>,
  6842. a world-class expert on Internet security and DNS, will
  6843. <a href="">testify before Congress</a>
  6844. on January 18th. One of the main thrusts of their testimony will be that
  6845. SOPA and the Protect IP Act will in fact be harmful to US national
  6846. security. (The fact that SOPA will
  6847. <a href="">break DNSSEC</a>
  6848. is one aspect of this, but not the only one.) Hopefully that will help
  6849. to keep these bills from passing.</p>
  6850. </div>
  6851. </content>
  6852. </entry>
  6854. <entry>
  6855. <title type="html">Another Brief Nerd Interlude</title>
  6856. <category term="/general" />
  6857. <id></id>
  6858. <updated>2011-12-16T04:55:00Z</updated>
  6859. <published>2011-12-16T04:55:00Z</published>
  6860. <link rel="alternate" type="text/html" href="" />
  6861. <content type="xhtml">
  6862. <div xmlns="">
  6863. <p>Years ago,
  6864. <a href="">Doug McIlroy</a>, the inventor
  6865. of the
  6866. <a href="">Unix pipe</a>,
  6867. published a
  6868. <a href="">paper</a>
  6869. on techniques for computing the terms of power series. The paper talks
  6870. about a number of key concepts in programming, such as "lazy" evaluation,
  6871. that were not well supported by most programming languages at the time,
  6872. which is why McIlroy spent a good portion of the paper describing an
  6873. implementation of his techniques in a new language designed by
  6874. <a href="">Rob Pike</a>.</p>
  6876. <p>I came across this paper recently and realized that Python's
  6877. <a href="">generators</a>
  6878. would be a perfect fit for representing power series. They support all
  6879. the key techniques McIlroy described, particularly "lazy" evaluation (a
  6880. generator doesn't compute any specific term of its series until it is
  6881. asked for it in sequence). You can see the Python implementation I came
  6882. up with on github
  6883. <a href="">here</a>.
  6884. A particularly neat feature is that you can recursively include a Python
  6885. generator in itself; this allows the recursive nature of many power series
  6886. to be directly represented in the code. For example, here's the exponential
  6887. series:</p>
  6889. <div class="codehilite"><pre><span class="k">def</span> <span class="nf">_exp</span><span class="p">():</span>
  6890.    <span class="k">for</span> <span class="n">term</span> <span class="ow">in</span> <span class="n">integral</span><span class="p">(</span><span class="n">EXP</span><span class="p">,</span> <span class="n">Fraction</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="mi">1</span><span class="p">)):</span>
  6891.        <span class="k">yield</span> <span class="n">term</span>
  6892. <span class="n">EXP</span> <span class="o">=</span> <span class="n">PowerSeries</span><span class="p">(</span><span class="n">_exp</span><span class="p">)</span>
  6893. </pre></div>
  6896. <p>No monkey business with factorials; just a generator that recursively
  6897. integrates itself. This same trick also works for implementing operations
  6898. on power series; for example, any series can be exponentiated by a method
  6899. similar to the above. The reciprocal and inverse operations on series use
  6900. similar tricks, which basically make the code look just like the
  6901. mathematical descriptions of those operations in McIlroy's paper.</p>
  6903. <p>Once I got the Python implementation working smoothly, I began checking
  6904. online to see what other recent implementations of these techniques
  6905. existed, and found that McIlroy posted an
  6906. <a href="">implementation in Haskell</a>
  6907. on the web in 2007. All of the key operations are one-liners. This is
  6908. possible because Haskell has built-in support for expressing these
  6909. operations declaratively, instead of having to define functions and use
  6910. for loops and so on. So in a sense, my Python implementation is a case of
  6911. <a href="">Greenspun's Tenth Rule</a>.
  6912. But it's still fun.</p>
  6913. </div>
  6914. </content>
  6915. </entry>
  6917. <entry>
  6918. <title type="html">Don't Tread On Our Internet: The Sequel</title>
  6919. <category term="/rants" />
  6920. <id></id>
  6921. <updated>2011-11-30T03:14:00Z</updated>
  6922. <published>2011-11-30T03:14:00Z</published>
  6923. <link rel="alternate" type="text/html" href="" />
  6924. <content type="xhtml">
  6925. <div xmlns="">
  6926. <p>I
  6927. <a href="">thought</a>
  6928. <a href="">I</a>
  6929. <a href="">was</a>
  6930. <a href="">done</a>
  6931. with this topic for now, but I can't help adding one more quick post,
  6932. because it now appears that it isn't just media companies who want to put
  6933. a stranglehold on the Internet.
  6934. <a href="">Chanel</a>
  6935. is getting in on the act.
  6936. Yes, the perfume maker. Based on what appears
  6937. to be extremely meager evidence, a Nevada federal judge has ordered that
  6938. "hundreds" of domain names can be seized by Chanel and transferred to
  6939. a US registrar (GoDaddy). The judge also ordered that "all Internet
  6940. search engines" and "all social media websites" must "de-index" the
  6941. seized domains.</p>
  6943. <p>In case you're wondering what "meager" evidence means, it appears that,
  6944. of the most recent batch of 228 domains that were seized, <em>three</em> were
  6945. actually verified to be shipping counterfeit merchandise, by placing an
  6946. order and seeing what was delivered. (Even this "verification" is
  6947. somewhat dubious, since it was done by Chanel itself and not by a neutral
  6948. third party. Also, if you're wondering about that "batch", Chanel has
  6949. been bringing these claims to court in groups, and there does not seem
  6950. to be any endpoint to this process; they'll just keep on doing it as
  6951. long as they feel like it.) The rest were seized based on "a Chanel
  6952. anti-counterfeiting specialist browsing the Web", according to the
  6953. article linked to above.</p>
  6955. <p>For extra fun, the
  6956. <a href="">court order</a>
  6957. that authorizes seizure of the latest 228 domains calls itself a
  6958. "Temporary Restraining Order". Anyone who has ever tried to switch
  6959. even an ordinary, non-seized domain from one registrar to another knows
  6960. how Byzantine the process is. I can only imagine what the companies
  6961. whose sites were seized, should they be found to not actually be selling
  6962. counterfeit merchandise, will have to go through to get control of their
  6963. domains back. So the order might as well say "Permanent" since that's
  6964. what it will effectively end up being. Eventually, Chanel might even
  6965. sell the domains it has seized; who says the domain seizure business
  6966. isn't lucrative?</p>
  6968. <p>One should also note that the court order, and the complaint that gave
  6969. rise to it, are <em>ex parte</em>, which is legalese for "the defendants didn't
  6970. get a chance to rebut anything". Also, the order is dated November 14,
  6971. 2011, and it sets a hearing date of November 29, 2011, with any responses
  6972. to the complaint required to be filed with the court by November 23, 2011.
  6973. Is it just me, or is that a <em>really short</em> response time? (Particularly
  6974. as many of the seized domains probably were not even based in the US,
  6975. which makes a Federal court's jurisdiction rather problematic. Indeed,
  6976. some non-US registrars apparently have not complied with the court's
  6977. order to transfer domain registrations to GoDaddy.)</p>
  6979. <p>I should make one thing clear: if Chanel wants to defend its trademarks
  6980. (that's what "counterfeit" amounts to in this case, infringement of
  6981. trademarks), it is entitled to use legal means to do so. I personally
  6982. think that anyone who is buying the counterfeit stuff is not going to be
  6983. a potential customer for the real stuff anyway (for one thing, the real
  6984. stuff's price is not for the squeamish, and people who are willing to
  6985. spend that much on perfume don't <em>want</em> the counterfeit stuff), so I
  6986. strongly doubt any of these "counterfeit" websites are losing Chanel any
  6987. sales. But that's their call; if they want to spend time and money on
  6988. what to me is a fruitless pursuit, they're welcome to.</p>
  6990. <p>But the appropriate legal means for that fruitless pursuit are <em>not</em>
  6991. seeking a "temporary" restraining order with an extremely short fuse as
  6992. a <em>first action</em>. Even the US Congress' previous attempt to emasculate
  6993. the Internet (and other things), the
  6994. <a href="">DMCA</a>,
  6995. doesn't allow that. (Yes, I know the DMCA refers to copyright, not
  6996. trademark; but the principle is the same.) In fact, one wonders why the
  6997. judge's first question to Chanel was not "Have you contacted any of these
  6998. websites and demanded that they cease and desist?" As the article I
  6999. linked to above notes, if the US government and US courts are going to
  7000. behave like this, the Protect IP Act/SOPA battle may end up being rather
  7001. superfluous. I understand that companies try these shortcuts all the
  7002. time, but that doesn't mean the government and courts have to cooperate.</p>
  7003. </div>
  7004. </content>
  7005. </entry>
  7007. <entry>
  7008. <title type="html">Yet Another Reason NOT To Tread On Our Internet</title>
  7009. <category term="/rants" />
  7010. <id></id>
  7011. <updated>2011-11-24T01:56:00Z</updated>
  7012. <published>2011-11-24T01:56:00Z</published>
  7013. <link rel="alternate" type="text/html" href="" />
  7014. <content type="xhtml">
  7015. <div xmlns="">
  7016. <p>This is just a quick update to
  7017. <a href="">yesterday's post</a>.
  7018. According to
  7019. <a href="">Ars Technica</a>,</p>
  7021. <blockquote>
  7022.  <p>Last Thursday, the European Parliament adopted a resolution ahead of a
  7023.  forthcoming summit between Europe and the United States. It included a
  7024.  section on "the need to protect the integrity of the global Internet and
  7025.  freedom of communication by refraining from unilateral measures to revoke
  7026.  IP addresses or domain names."</p>
  7028.  <p>That provision was added at the urging of the civil liberties organization
  7029.  European Digital Rights (EDRi). In a presentation to the Parliament's Civil
  7030.  Liberties Committee, EDRi's Joe McNamee noted that "the United States has, up
  7031.  until recently, never sought to exploit its theoretical jurisdiction over the
  7032.  companies and infrastructure that are at the core of the Internet."</p>
  7033. </blockquote>
  7035. <p>The Internet was created in the US, and as the article goes on to note, key
  7036. pieces of the Internet's infrastructure are based in the US, such as
  7037. <a href="">ICANN</a>,
  7038. the company that coordinates domain name assignments, many key DNS root
  7039. servers, and the registries for .com, .org, and other popular TLDs. Up to
  7040. now, nobody has really had a problem with this, because the US has been
  7041. careful not to abuse its privileged position. If our lawmakers want to
  7042. change that, and squander our position by abusing it, they're going the
  7043. right way about it. It's ironic that we now have the European Union, whose
  7044. <a href="">Constitution</a>
  7045. weighs in at more than 350 pages (more than double that if the "Protocols
  7046. and Annexes" and "Declarations" are included) and has to be delivered as
  7047. PDFs, trying to tell the United States, whose
  7048. <a href="">Constitution</a>
  7049. can fit in a single reasonably sized HTML page (or two such if the
  7050. Amendments are included--what extravagance!), how something as simple as
  7051. protecting freedom of communication and civil liberties is supposed to
  7052. work. But so it is.</p>
  7054. <p>I suppose one could argue that, in the long run, it would be better for
  7055. the Internet's infrastructure to come under the control of an international
  7056. body that was not controlled by any single country's government. In theory
  7057. that would be a good argument. The problem with it is that all our evidence
  7058. about such bodies shows that they do not work. The United Nations was
  7059. supposed to end war and ensure human rights for everyone. As Dr. Phil would
  7060. say, how's that workin' out for ya? As much as I complain about the US
  7061. government, I still think that, up to now, its stewardship of the Internet
  7062. has served the Internet better than any possible alternative, and that
  7063. ending that stewardship would be a turn for the worse. I really hope it
  7064. doesn't come to that.</p>
  7065. </div>
  7066. </content>
  7067. </entry>
  7069. <entry>
  7070. <title type="html">No, Really, DON'T Tread On Our Internet</title>
  7071. <category term="/rants" />
  7072. <id></id>
  7073. <updated>2011-11-23T04:59:00Z</updated>
  7074. <published>2011-11-23T04:59:00Z</published>
  7075. <link rel="alternate" type="text/html" href="" />
  7076. <content type="xhtml">
  7077. <div xmlns="">
  7078. <p>I've
  7079. <a href="">posted</a>
  7080. <a href="">twice</a>
  7081. now about the
  7082. <a href="">Protect IP Act</a>,
  7083. or
  7084. <a href="">SOPA</a>
  7085. (the former is the Senate version, the latter is the House version),
  7086. which is the latest attempt on the part of big media companies to put a
  7087. stranglehold on the Internet.
  7088. As you can see, since this is the third time
  7089. around on this topic, I'm not going to mince words. I've mentioned some of
  7090. the damage this bill will cause in previous posts, but it's worth taking
  7091. a look at the Wikipedia article on SOPA linked to above and seeing all the
  7092. different issues raised under "Ramifications". (The Wiki article also has
  7093. lots of good reference links.) If you haven't already done so, and you are
  7094. a US citizen, I strongly urge you to contact your elected representatives
  7095. and demand that they make sure this bill doesn't pass. Two good sites to
  7096. help you do that are the
  7097. <a href="">Electronic Frontier Foundation</a>
  7098. and
  7099. <a href="">Demand Progress</a>.
  7100. You can also go to the
  7101. <a href="">Stop Censorship</a>
  7102. site to register your opposition to the bill and to put your name on a list
  7103. of citizens that Senator Ron Wyden intends to read from the Senate floor if
  7104. he is forced to filibuster the Senate bill.</p>
  7106. <p>Having got that out of the way, I can now vent in a more leisurely fashion.
  7107. (This post is filed under "rants", but the previous paragraph is not just
  7108. venting, as my rants usually are. The issue is a serious one, so I wanted
  7109. to get the serious part out of the way first; but that's only a paragraph
  7110. and the rest of this post is, well, longer, so into the "rants" category
  7111. it goes.) Today I came across a
  7112. <a href="">blog post</a>
  7113. comparing SOPA to China's "Great Firewall", which is the Chinese government's
  7114. massive infrastructure dedicated to controlling what the Chinese people can
  7115. and can't see on the Internet. The post is worth a read, if for no other
  7116. reason than that it uses the same word I did above, "stranglehold", to
  7117. describe the aim of this bill. :-) (It also has a number of good reference
  7118. links.)</p>
  7120. <p>What gets me, though, is that as you'll see if you read the post, some US
  7121. lawmakers actually think that the fact that legislation like this would make
  7122. the US more like China in its control of the Internet is a feature, not a
  7123. bug. If you're a person of
  7124. <a href="">Heinlein's class two</a>,
  7125. like me, this sort of thinking just seems so far out in left field that it's
  7126. hard to understand how it can survive and even thrive. Don't these people
  7127. understand that the Internet is not something you can control? Don't they
  7128. realize that the Internet is individual empowerment, individual freedom,
  7129. individual choice, in just about the purest form that those things have
  7130. ever existed? The United Nations realizes it; as the blog post I linked to
  7131. above notes, the UN has declared free and uncensored Internet access to be a
  7132. <a href="">basic human right</a>.
  7133. Shouldn't that be enough to stop this kind of legislation from even being
  7134. considered? Hey, folks, um, trying to put into law something that the UN
  7135. considers a violation of human rights is probably a bad idea, okay? It's
  7136. not as though there aren't plenty of other pressing matters to attend to,
  7137. like, say, trying not to let the country's credit rating slip again.</p>
  7139. <p>But of course the proponents of this legislation do realize all the above.
  7140. <em>That's why they're trying to get it passed.</em> The analogy with the Chinese
  7141. government is far closer than it might seem. China is the classic example
  7142. of a country run by people of
  7143. <a href="">Heinlein's class one</a>;
  7144. the government wants to control <em>everything</em>. The United States of America
  7145. was supposed to be the exact opposite: the government was supposed to
  7146. control as little as possible. But it's really, really tough to sustain
  7147. that vision in the face of harsh reality. The US government in 1790 didn't
  7148. have much of a choice, because the country was so large and the technology
  7149. of the time so limited. The pattern ever since has been for the amount of
  7150. control the Federal government exerts to increase, to the point where today
  7151. we have laws and regulations covering all manner of things that would have
  7152. been unimaginable as subjects of Federal interest to an American of 1790
  7153. (or even one of 1890, for that matter). But at least, for most of that time,
  7154. each individual law or regulation appeared to be a good idea in itself;
  7155. what caused trouble was the cumulative effect of all of them, combined, of
  7156. course, with the law of unintended consequences.</p>
  7158. <p>Now, though, we have something different: we have a law that, I would venture
  7159. to say, looks like a <em>bad</em> idea to most Americans, and yet it won't go away.
  7160. It's not the first such law; take a look at many of the laws and regulations
  7161. that cover the financial industry, and you will see the same pattern (and,
  7162. not coincidentally, a lot of reasons for the economic meltdown we have been
  7163. experiencing). It's certainly not a new thing to have a bunch of large
  7164. corporations with outdated business models trying to buy legislation that
  7165. will prop up those business models, regardless of the damage it might cause
  7166. elsewhere, using the cover story that it will "protect society" from something
  7167. or other. We heard the same story from the financial industry all the way up
  7168. to the big crash, and now we're hearing from them that we need more regulations
  7169. to "protect" us from another one. (Remember the classic definition of insanity?
  7170. Doing the same thing over and over again and expecting different results?) And
  7171. we're hearing the same story in the fight over the US budget and whether or not
  7172. the Bush tax cuts should be allowed to expire; we're told that keeping the tax
  7173. cuts will "protect" small businesses and help innovation, when in fact a number
  7174. of commentaries, for example these articles in
  7175. <a href="">USA Today</a>
  7176. and
  7177. <a href="">Business Week</a>,
  7178. have argued that letting the tax cuts expire would not hurt small businesses
  7179. at all, and might even help them, since the current rules actually define
  7180. "small business" in a way that excludes a lot of the businesses that are
  7181. actually (a) small, and (b) innovating, while including a lot of entities
  7182. that are really more tax shelters for the wealthy than anything else.</p>
  7184. <p>So why am I picking this particular issue, a free and uncensored Internet, as
  7185. the one we really, really need to take a stand on? Because with a free and
  7186. uncensored Internet, it's a lot <em>easier</em> to fight all those other battles.
  7187. Information is power, and the big media companies who are trying to put a
  7188. stranglehold on the Internet know it. So do lawmakers who would love to give
  7189. the US government the power to shut down sites like Wikileaks. But we in the
  7190. USA are supposed to understand that we, the people, have the power. <em>We</em> get
  7191. to decide how information flows. The Internet is our medium for making those
  7192. decisions. If we want to say something, we post it. If we like something
  7193. someone else says, we link to it. If we <em>don't</em> like what someone else says,
  7194. we <em>refute</em> it. We don't censor it. We fight bad information with better
  7195. information, <em>not</em> with a Great Firewall. Ultimately, if we don't like what's
  7196. on a given website, we exercise our own freedom of choice by surfing somewhere
  7197. else. But we need a free and uncensored Internet to do all this. Don't let it
  7198. be taken away.</p>
  7199. </div>
  7200. </content>
  7201. </entry>
  7203. <entry>
  7204. <title type="html">Dennis Ritchie, RIP</title>
  7205. <category term="/opinions" />
  7206. <id></id>
  7207. <updated>2011-10-15T01:50:00Z</updated>
  7208. <published>2011-10-15T01:50:00Z</published>
  7209. <link rel="alternate" type="text/html" href="" />
  7210. <content type="xhtml">
  7211. <div xmlns="">
  7212. <p>Amidst all the news about Steve Jobs' passing, you may not have heard that
  7213. Dennis Ritchie, creator of the C programming language and one of the
  7214. original designers of Unix, also passed away this past weekend.
  7215. The news first broke on the Internet through
  7216. <a href="">this post from Rob Pike on Google Plus</a>.
  7217. Pike later followed up with
  7218. <a href="">another Google Plus post</a>
  7219. giving a longer tribute to Ritchie and his work. Tributes to Ritchie have
  7220. also appeared in the
  7221. <a href="">New York Times</a>,
  7222. on
  7223. <a href="">Wired, by way of CNN Tech</a>
  7224. (including the classic photo of Ritchie and Ken Thompson, the founders of Unix,
  7225. working on a PDP-11), on
  7226. <a href="">BoingBoing</a>,
  7227. in
  7228. <a href="">The Register</a>,
  7229. on
  7230. <a href="">Tim Bray's blog</a>,
  7231. and on
  7232. <a href="">Herb Sutter's blog</a>.
  7233. Finally, this
  7234. <a href="">follow-up Google Plus post by Rob Pike</a>
  7235. is worth a quick read, since it quotes from an email Pike received from
  7236. Ritchie encouraging him to pursue a programming project that turned out
  7237. to be important in the history of Unix.</p>
  7239. <p>I don't have much to add to what you'll read at the above links, but I
  7240. do want to comment on the headline of the Wired/CNN story: "Dennis
  7241. Ritchie: The shoulders Steve Jobs stood on." Jobs, and pretty much
  7242. everybody else who uses a computer. Ritchie created the C language, as
  7243. the story notes, "because he and Ken Thompson needed a better way to
  7244. build UNIX." But it turned out that the feature of C that enabled that,
  7245. the fact that it was "portable" between different types of computer
  7246. hardware, turned out to be a better way to write almost all programs,
  7247. not just Unix. C is now the foundation of pretty much every piece of
  7248. software in existence; if the software is not written in C directly,
  7249. it's built on top of something that is. (My personal favorite language,
  7250. <a href="">Python</a>,
  7251. for example, is implemented in C.) The story quotes Brian Kernighan,
  7252. another key figure in the development of C, as saying, "There's that
  7253. line from Newton about standing on the shoulders of giants...We're all
  7254. standing on Dennis' shoulders." I'm glad Wired and CNN recognized this
  7255. and gave Ritchie his due.</p>
  7257. <h1>Postscript</h1>
  7259. <p>I shouldn't end without mentioning
  7260. <a href="">Ritchie's page at Bell Labs</a>,
  7261. which has links to many of his writings. If you're a programmer, or even
  7262. if you're not, they're worth reading.</p>
  7263. </div>
  7264. </content>
  7265. </entry>
  7267. <entry>
  7268. <title type="html">Some Items About Steve Jobs</title>
  7269. <category term="/opinions" />
  7270. <id></id>
  7271. <updated>2011-10-09T02:39:00Z</updated>
  7272. <published>2011-10-09T02:39:00Z</published>
  7273. <link rel="alternate" type="text/html" href="" />
  7274. <content type="xhtml">
  7275. <div xmlns="">
  7276. <p>Unusually for me, this post will be almost entirely links to and quotes
  7277. from articles by others. But I should explain briefly why I'm linking to
  7278. them and quoting them. It's not to set the stage for my own comments about
  7279. Mac OS X, or about iPods and iPads and so forth. I made comments about
  7280. OS X in an
  7281. <a href="">earlier post</a>,
  7282. and there's no need to rehash them here. Nor do I have any personal
  7283. anecdotes to share. My reason for linking to these articles, and quoting
  7284. briefly from them, is, quite simply, to draw attention to what they say.</p>
  7286. <p>First, Eric Raymond's post
  7287. <a href="">On Steve Jobs's Passing</a>,
  7288. in which, towards the end, he says:</p>
  7290. <blockquote>
  7291.  <p>Commerce is powerful, but culture is even more persistent. The lure of
  7292.  high profits from secrecy rent can slow down the long-term trend towards
  7293.  open source and user-controlled computing, but not really stop it.
  7294.  Jobs's success at hypnotizing millions of people into a perverse love
  7295.  for the walled garden is more dangerous to freedom in the long term than
  7296.  Bill Gates's efficient but brutal and unattractive corporatism. People
  7297.  feared and respected Microsoft, but they love and worship Apple - and
  7298.  that is precisely the problem, precisely the reason Jobs may in the end
  7299.  have done more harm than good.</p>
  7300. </blockquote>
  7302. <p>Next, quotes from two articles that Raymond links to. One is an op-ed in
  7303. the New York Times entitled
  7304. <a href="">Steve Jobs, Enemy of Nostalgia</a>,
  7305. by Mike Daisey:</p>
  7307. <blockquote>
  7308.  <p>I have traveled to southern China and interviewed workers employed in
  7309.  the production of electronics. I spoke with a man whose right hand was
  7310.  permanently curled into a claw from being smashed in a metal press at
  7311.  Foxconn, where he worked assembling Apple laptops and iPads. I showed
  7312.  him my iPad, and he gasped because he'd never seen one turned on. He
  7313.  stroked the screen and marveled at the icons sliding back and forth,
  7314.  the Apple attention to detail in every pixel. He told my translator,
  7315.  "It's a kind of magic."</p>
  7317.  <p>Mr. Jobs's magic has its costs. We can admire the design perfection and
  7318.  business acumen while acknowledging the truth: with Apple's immense
  7319.  resources at his command he could have revolutionized the industry to
  7320.  make devices more humanely and more openly, and chose not to. If we view
  7321.  him unsparingly, without nostalgia, we would see a great man whose genius
  7322.  in design, showmanship and stewardship of the tech world will not be seen
  7323.  again in our lifetime. We would also see a man who in the end failed to
  7324.  "think different," in the deepest way, about the human needs of both his
  7325.  users and his workers. </p>
  7326. </blockquote>
  7328. <p>The other quote is from
  7329. <a href="">SkepticBlog</a>:</p>
  7331. <blockquote>
  7332.  <p>Most pancreatic cancers are aggressive and always terminal, but Steve
  7333.  was lucky (if you can call it that) and had a rare form called an islet
  7334.  cell neuroendocrine tumor, which is actually quite treatable with
  7335.  excellent survival rates - if caught soon enough. The median survival
  7336.  is about a decade, but it depends on how soon it's removed surgically.
  7337.  Steve caught his very early, and should have expected to survive much
  7338.  longer than a decade. Unfortunately Steve relied on a diet instead of
  7339.  early surgery. There is no evidence that diet has any effect on islet
  7340.  cell carcinoma. As he dieted for nine months, the tumor progressed, and
  7341.  took him from the high end to the low end of the survival rate.</p>
  7342. </blockquote>
  7344. <p>Finally, to put that last quote in perspective, here's one from another
  7345. article, on Science Blogs, about
  7346. <a href="">Steve Jobs, neuroendocrine tumors, and alternative medicine</a>:</p>
  7348. <blockquote>
  7349.  <p>Did Jobs significantly decrease his chance of surviving his cancer by
  7350.  waiting nine months to undergo surgery? It seems like a no-brainer,
  7351.  but it turns out that that's actually a very tough question to answer.
  7352.  Certainly, it's nowhere near as certain as Dunning [the author of the
  7353.  SkepticBlog article] tries to make it seem when he writes things like:</p>
  7355.  <p>"Eventually it became clear to all involved that his alternative therapy
  7356.  wasn't working, and from then on, by all accounts, Steve aggressively
  7357.  threw money at the best that medical science could offer. But it was too
  7358.  late. He had a Whipple procedure. He had a liver transplant. And then he
  7359.  died, all too young."</p>
  7361.  <p>After over seven years of science-based treatments that prolonged his life.</p>
  7363.  <p>One has to be very, very careful about making this sort of argument. For one
  7364.  thing, it could not have been apparent that it was "too late" back in 2004,
  7365.  when it became clear that Jobs' dietary manipulations weren't working. For
  7366.  another thing, we don't know how large the tumor was, whether it progressed
  7367.  or simply failed to shrink over those nine months, and by how much it
  7368.  increased in size, if increase in size it did...</p>
  7370.  <p>In retrospect, we can now tell that Jobs clearly had a tumor that was
  7371.  unusually aggressive for an insulinoma. Such tumors are usually pretty
  7372.  indolent and progress only slowly. Indeed, I've seen patients and known a
  7373.  friend of a friend who survived many years with metastatic neuroendocrine
  7374.  tumors with reasonable quality of life. Jobs was unfortunate in that he
  7375.  appears to have had an unusually aggressive form of the disease that probably
  7376.  would have killed him no matter what. That's not to say that we shouldn't
  7377.  take into account his delay in treatment and wonder if it contributed to his
  7378.  ultimate demise. It very well might have, the key word being "might." We
  7379.  don't know that it did, which is one reason why we have to be very, very
  7380.  careful not to overstate the case and attribute his death as being definitely
  7381.  due to the delay in therapy due to his wanting to "go alternative." It's
  7382.  also important to remember that, as much of a brilliant visionary Jobs was,
  7383.  even brilliant visionaries can make bad decisions when it comes to health.</p>
  7384. </blockquote>
  7386. <p>All of the articles are worth reading in full; none of them are very long.
  7387. All of them recognize what everyone knows, that Steve Jobs was indeed a
  7388. brilliant visionary; as President Obama said in his
  7389. <a href="">statement eulogizing Jobs</a>,</p>
  7391. <blockquote>
  7392.  <p>there may be no greater tribute to Steve's success than the fact that much
  7393.  of the world learned of his passing on a device he invented.</p>
  7394. </blockquote>
  7396. <p>The quotes I've given above focus on other aspects of Jobs's life and work,
  7397. not because they're more important, necessarily, but because it's always
  7398. worth trying to put things in perspective. And that's all for now.</p>
  7399. </div>
  7400. </content>
  7401. </entry>
  7403. <entry>
  7404. <title type="html">Why I'm Not Crazy About The Cloud</title>
  7405. <category term="/opinions" />
  7406. <id></id>
  7407. <updated>2011-09-26T22:02:00Z</updated>
  7408. <published>2011-09-26T22:02:00Z</published>
  7409. <link rel="alternate" type="text/html" href="" />
  7410. <content type="xhtml">
  7411. <div xmlns="">
  7412. <p>I've seen a number of online articles and blog posts recently with the
  7413. common theme of being uncomfortable with Facebook. For instance,
  7414. <a href="">this</a>
  7415. at Slate, or
  7416. <a href="">this</a>
  7417. from a programmer, or
  7418. <a href="">this</a>
  7419. from a Facebook developer.</p>
  7421. <p>Of course the privacy implications of putting your data in the cloud,
  7422. with Facebook or anywhere else, are (or should be) obvious; as
  7423. <a href="">this article by Rick Moen</a>
  7424. says,</p>
  7426. <blockquote>
  7427.  <p>Experience suggests that "Possession is nine points of the law", and the
  7428.  best way to prevent
  7429.  <a href="">abuse of your personal data by strangers</a>
  7430.  is not to give it to them.</p>
  7431. </blockquote>
  7433. <p>I personally do not do Facebook, and keep only minimal data in the online
  7434. accounts I do have, for exactly this reason. But apparently most people
  7435. are not like me, and are quite willing to post lots of personal information
  7436. where it is visible, if not to the entire Internet, at least to anyone who
  7437. uses the same social media services they do (which is a lot of the entire
  7438. Internet). That's a decision everyone is entitled to make.</p>
  7440. <p>However, the common theme of the above articles goes beyond that, and it
  7441. was brought into focus for me by
  7442. <a href="">this article on The Atlantic's website</a>
  7443. (which I got to via <a href="">Hacker News</a>
  7444. ).</p>
  7446. <p>The key observation is this:</p>
  7448. <blockquote>
  7449.  <p>We've always been dependent on software providers to create the digital
  7450.  spaces we inhabit, but when your email and documents and music are in
  7451.  the cloud, you're giving up the lock on the door and allowing changes to
  7452.  be made on the schedule of the parent. He or she may clean up or buy you
  7453.  a new desk. He or she may take away the car or decide you can't do
  7454.  something you think you should be able to.</p>
  7455. </blockquote>
  7457. <p>This is not a new problem. The question of whether to own or rent (or
  7458. more precisely to live rent-free on someone's property, which is what the
  7459. "cloud" amounts to, as the Atlantic article notes later on) is a very old
  7460. one, and the tradeoff has always been the same: ownership can be a burden,
  7461. yes, but without it you lose control. Facebook and Google and all these
  7462. other wonderful "digital spaces" <em>do not belong to you</em>. And complaining
  7463. that the changes are affecting your "user experience" is kind of missing
  7464. the point; your user experience makes a difference to them only to the
  7465. extent that it affects their ability to collect data that they can sell.
  7466. It's not as though you are a paying customer.</p>
  7468. <p>But I seem to be an outlier; most people don't seem to see things in the
  7469. stark way I just put them. Perhaps what makes me (and Rick Moen, and other
  7470. programmers who have posted about this kind of thing) different is that I
  7471. don't need or want "software providers" to create any "digital spaces" for
  7472. me. Sure, there are some services (like Google) that I can't reasonably
  7473. create for myself, so I use them (but I try to give Google
  7474. <a href="">as little information as I can</a>
  7475. when doing so). And I do participate in online
  7476. <a href=";postcount=631">forums</a>
  7477. in areas I'm interested in, and where I think I have something to
  7478. contribute. But those are limited uses for limited purposes. As far as
  7479. my email and documents and music go, I can see paying a hosting provider
  7480. to maintain one's own personal "cloud" online, both as a backup and for
  7481. convenience. (I don't even do this for a lot of my data, but as I said,
  7482. I'm an outlier; not everybody has a RAID array on a file server, not to
  7483. mention multiple other machines to store copies of data, in their house.
  7484. Although these days, external USB hard drives are cheap enough that anyone
  7485. can have their own personal backup storage.) But the idea of putting <em>all</em>
  7486. my data on the servers of someone, even Google, to whom I am <em>not</em> a paying
  7487. customer--<em>that</em> is where I draw the line. No amount of convenience is
  7488. worth that.</p>
  7490. <p>And it is basically about convenience, as the Atlantic article makes
  7491. clear:</p>
  7493. <blockquote>
  7494.  <p>This isn't a bug in the way that cloud services work. It is a feature.
  7495.  What we lose in freedom we gain in convenience. Maybe the tradeoff is
  7496.  worth it. Or maybe it's something that just happened to us, which we'll
  7497.  regret when we realize the privacy, security, and autonomy we've given
  7498.  up to sync our documents and correspondence across computers.</p>
  7499. </blockquote>
  7501. <p>In case the reader is still in any doubt, I'm betting on the second
  7502. option. I could be wrong; it could be that "cloud" services will evolve
  7503. in the direction of less centralization and more privacy. But I think
  7504. that will only happen if their business model is changed. And <em>that</em>
  7505. will only happen if we, as users, start to be willing to pay for some
  7506. things that we are used to getting for free, in order to ensure that
  7507. they are set up in <em>our</em> interests. We'll see.</p>
  7509. <p><strong>Update</strong>: Not long after posting the above I came across
  7510. <a href="">this post</a>
  7511. taking an even bleaker position than mine:</p>
  7513. <blockquote>
  7514.  <p>The promise of the open web looks increasingly uncertain. The
  7515.  technology will continue to exist and improve. It looks like you'll
  7516.  be able to run your own web server on your own domain for the
  7517.  foreseeable future. But all the things that matter will be controlled
  7518.  and owned by a very small number of Big Web companies.</p>
  7519. </blockquote>
  7521. <p>However, it's clear from the rest of the post that this kind of thing is
  7522. much more of an issue for businesses and organizations than for individual
  7523. people. I am certainly not opposed to as many people as possible reading
  7524. this blog, but I don't <em>need</em> people to read it the way a lot of
  7525. businesses need people to visit their sites. And for a business, the
  7526. tradeoff between keeping control of data and making information visible
  7527. is very different. I don't use Facebook in my personal life, but as a
  7528. business owner (see the
  7529. <a href="">PDApps</a>
  7530. link under Wizard Projects on this page), we certainly have a social
  7531. media presence; we <em>want</em> people to see what we're up to as a business.
  7532. No small business has ever been able to control the rules by which that
  7533. game is played; companies that a decade or two ago were trying to get
  7534. press "hits" (see
  7535. <a href="">Paul Graham's essay on PR firms</a>
  7536. for a good account of how this worked in the mid-1990's for a web startup)
  7537. are now doing search engine optimization and trying to build Facebook and
  7538. Twitter followings. I simply don't see this sort of thing as a harbinger of
  7539. doom for the web, or anything else; it's just business as usual.</p>
  7541. <p>But the post I quoted from just now also has links to projects that are
  7542. aimed at making it easier to have the convenience of data in the "cloud",
  7543. without having to sacrifice privacy and control. For example,
  7544. <a href="">Unhosted</a>
  7545. is building software to allow you to have your own "cloud", which you can
  7546. host anywhere but which only you (or those you give user accounts to) can
  7547. access, because everything stored in the "cloud" is encrypted.
  7548. <a href="">Diaspora</a>
  7549. is taking a somewhat different approach, building tools to let people set
  7550. up their own social media networks where they control their own data and
  7551. how it is shared.
  7552. <a href="">Thimbl</a>
  7553. is making a micro-blogging platform that is free and open source. And
  7554. even the W3C, the web standards body, is working on specs for a
  7555. <a href="">Federated Social Web</a>.
  7556. So even on the personal side, the picture is not all bleak. There will
  7557. always be ways to opt out of the Big Web companies' offerings if you
  7558. want to. Maybe most people won't want to; that's their choice. But it's
  7559. important that there <em>is</em> a choice.</p>
  7560. </div>
  7561. </content>
  7562. </entry>
  7564. <entry>
  7565. <title type="html">George Smiley is Back</title>
  7566. <category term="/general" />
  7567. <id></id>
  7568. <updated>2011-09-15T01:29:00Z</updated>
  7569. <published>2011-09-15T01:29:00Z</published>
  7570. <link rel="alternate" type="text/html" href="" />
  7571. <content type="xhtml">
  7572. <div xmlns="">
  7573. <p>This is a culture interlude, not a nerd interlude, although of
  7574. course the <em>kind</em> of culture I'm about to focus on might well be
  7575. called nerdish. A new production of Le Carre's classic spy novel,
  7576. <em>Tinker, Tailor, Soldier, Spy</em> is out. A short video of the cast
  7577. at the premiere in Britain is
  7578. <a href="">here</a>.</p>
  7580. <p>Very interesting casting: Gary Oldman as George Smiley; it will
  7581. be interesting to see how his take on the role differs from the
  7582. book and Alec Guiness' classic portrayal, since this role is
  7583. kind of against type for him. Not that I'm against that; good
  7584. actors deserve chances to do different roles. John Hurt is
  7585. "Control", Smiley's former boss who died too soon to finish the
  7586. hunt for the Russian mole in The Circus (Le Carre's name for
  7587. British Intelligence), leaving Smiley to complete the job. That
  7588. should be interesting; the role has possibilities that weren't
  7589. really explored in the 1970's BBC adaptation.</p>
  7591. <p>I won't say any more since I want to avoid the temptation to
  7592. speculate without having seen the film, or to include spoilers
  7593. based on my knowledge of the story from the books. (Yes, I've
  7594. read every one that I mention below, and have re-read most of
  7595. them a number of times. I warned you about what kind of interlude
  7596. this was.) I have just one final comment. At the end of the video
  7597. clip, Le Carre was asked about possible sequels, and he took what
  7598. I thought was just the right line: yes, it would be great, but it
  7599. has to be done with "fear, not confidence." I would love to see
  7600. this crew do <em>Smiley's People</em>, assuming this film does well
  7601. (which I think and hope it will). Having them do every Le Carre
  7602. novel in which Smiley appears (there are five in which he plays
  7603. the leading role, one, <em>The Spy Who Came In From The Cold</em>, in
  7604. which he is a major character but not the lead, and at least
  7605. two others I can think of in which he appears) is probably too
  7606. much to ask, but one can hope. :-)</p>
  7607. </div>
  7608. </content>
  7609. </entry>
  7611. <entry>
  7612. <title type="html">Why I Run Linux</title>
  7613. <category term="/rants" />
  7614. <id></id>
  7615. <updated>2011-09-11T03:59:00Z</updated>
  7616. <published>2011-09-11T03:59:00Z</published>
  7617. <link rel="alternate" type="text/html" href="" />
  7618. <content type="xhtml">
  7619. <div xmlns="">
  7620. <p>Having spent enough time using all three of the major OS's to have a
  7621. decent understanding of their flaws, it's easy to explain why I use
  7622. Linux whenever I have a choice: its flaws are much easier to manage.
  7623. Part of that is the Unix tradition: everything is visible, and you can
  7624. always take full manual control of something if you really need to. On
  7625. Windows, there's a lot of stuff happening behind the scenes that you
  7626. can't really see, and can't really control. On my Linux box, I know
  7627. exactly what every single running process is there for, what data it
  7628. is able to clobber, and what functionality will go away if I kill it.
  7629. On Windows there are all these "system services" lying around that I
  7630. can't make go away and can't know what they could potentially mess up.
  7631. And then there's the Windows Registry...</p>
  7633. <p>But Mac OS X also has Unix under the hood, so the Unix tradition isn't
  7634. all there is to it. The problem I have with Mac OS X is that there are
  7635. too many layers of spackle over the Unix core, and those layers are not
  7636. transparent the way Unix is supposed to be. And although OS X lets you
  7637. do things in the traditional Unix way, without going through all the
  7638. layers of spackle, it strongly discourages you from doing it very much.
  7639. And yet, some basic GUI tools simply aren't there, such as a task
  7640. scheduler (even Windows has a GUI task scheduler) to automate things
  7641. like backups. Yes, I know OS X has Time Machine, but that's another
  7642. example of the same problem. If I want to use some other mechanism for
  7643. backups, such as rsync, or pushing to a version control repository, I
  7644. can't use Time Machine and the other GUI tools; I have to drop back to
  7645. the Unix command line, and OS X doesn't even like you to use crontab
  7646. the way it was designed. Basically, OS X wants you to do everything in
  7647. the One True Apple Way, which is fine if that works for you, but when
  7648. it doesn't, things get very difficult.</p>
  7650. <p>Linux has flaws too, of course, but as I said above, they are much
  7651. easier to manage. For example,
  7652. <a href="">KDE 4 sucks</a>,
  7653. but I don't have to use KDE 4; I can use KDE 3, or any of a number of
  7654. other desktops. Or I can be more minimalist and just pick a good
  7655. window manager, of which there are quite a few, and build my own GUI
  7656. desktop tailored to my needs. Or I could be even more hard-core
  7657. minimalist and run bare-bones X Windows. (I'm pretty sure that there
  7658. are very few, if any, nerds left who are so hard-core that they refuse
  7659. to run a GUI at all; even if you're just running emacs, a terminal,
  7660. and a text-only browser, it's still nice to have multiple windows on
  7661. screen at the same time.) As another example, Ubuntu's standard GUI
  7662. for network configuration is braindead enough not to have a way to
  7663. specify DNS servers when you're using DHCP, even though the DHCP
  7664. config files have an option designed precisely for this purpose. I
  7665. need this option because, for reasons explained by Rick Moen in
  7666. <a href="">this article</a>,
  7667. I run my own recursive DNS server on my home network instead of just
  7668. defaulting to the ones run by my ISP (who shall remain nameless here
  7669. because, at least in this particular case, they're no worse than any
  7670. of the others). If Windows lacked this option in its GUI, I'd be
  7671. helpless (ok, maybe not totally helpless, since I suppose there's
  7672. some registry key somewhere that might help with this, but it might
  7673. take a lot of pointless work to find it). If OS X did, I'd be, if not
  7674. exactly helpless, certainly wearied after what I would have to do to
  7675. make sure the config change persisted across reboots. On Linux, I
  7676. make a quick change to /etc/dhcp3/dhclient.conf and I'm done.</p>
  7678. <p>Another thing the various Linux distributions have done very well
  7679. is providing a cryptographically verified distribution chain for
  7680. software, both binary and source. On Windows, yes, you can get
  7681. software updates from the Windows Update site, but that only
  7682. covers software that Microsoft distributes, and many of the
  7683. so-called "updates" aren't really for your benefit anyway; they
  7684. are to enable Microsoft to control more of what you do with your
  7685. computer. On OS X, again, you can get updates for Apple's software
  7686. from Apple, but not for third party software. For third party
  7687. software on both Windows and Mac, you are pretty much stuck with
  7688. surfing to some website, downloading an .msi or .dmg file that
  7689. may or may not have any means of verifying that it's actually what
  7690. you think it is, and installing the thing by hand in the hope that
  7691. it isn't a virus or spyware. And, of course, in neither case is the
  7692. source code open to you. With Linux, you use the package management
  7693. tool that comes with your distribution, which can verify that every
  7694. piece of software it downloads and installs for you is digitally
  7695. signed and comes from a trustworthy source.</p>
  7697. <p>I realize that this doesn't sound much like making flaws easier
  7698. to manage, but bear with me for a moment. When people find out I
  7699. run Linux, they often ask how I keep my system free from viruses.
  7700. The definitive answer to this question is by Rick Moen,
  7701. <a href="">here</a>,
  7702. and basically boils down to this: viruses and malware are not
  7703. problems in themselves; they are symptoms of deeper problems with
  7704. either the OS or the user or both. No OS is perfect; there are
  7705. always flaws in any software system that can in principle be
  7706. exploited. But one can still design an OS to not only make it as
  7707. free of flaws as possible, but also to minimize the impact of
  7708. flaws when they are found. And included in "flaws" from this point
  7709. of view are ways that a user can make mistakes and do damage to
  7710. the system. Linux, and Unix variants generally, are designed
  7711. explicitly to contain such damage, and so the kinds of anti-virus,
  7712. anti-spyware, anti-whatever tools that are ubiquitious on Windows
  7713. are simply not necessary. Mac OS X, being built on Unix, gains
  7714. this advantage to an extent, but the key thing it lacks, compared
  7715. to Linux, is precisely the cryptographically secure distribution
  7716. chain, which removes an obvious way for users to make mistakes
  7717. and do damage. (Windows, by contrast, has had what security
  7718. features it has bolted on as afterthoughts, and it shows.)</p>
  7720. <p>Of course, as computer users go, I am a statistical outlier. I am
  7721. a programmer, which means that I have to have a degree of control
  7722. over my computer that most users do not, because I have to be
  7723. able to configure it to build and test software, not just use it.
  7724. It also means that I actually <em>like</em> doing things like editing
  7725. /etc/dhcp3/dhclient.conf by hand to make sure DNS is set up just
  7726. the way I want it. Users with different requirements for their
  7727. computers will make different tradeoffs between control and
  7728. convenience, and that's fine. This post is about why <em>I</em> run Linux,
  7729. not why everyone should.</p>
  7731. <p>But to close, I do want to say one thing about Linux that <em>should</em>
  7732. be important to everyone. Control over your own computer is not just
  7733. for programmers. I'm not saying that every user should have to edit
  7734. /etc/dhcp3/dhclient.conf; but every user should <em>own</em> their computer,
  7735. and the software that runs on it. If you run Windows, you may not
  7736. realize it, but you don't own your software; you are "licensed" to
  7737. use it by Microsoft, and if you read the fine print in all those
  7738. license agreements that you probably clicked through without thinking,
  7739. you will find that you have signed up to allow Microsoft some pretty
  7740. draconian control over not just the specific software they gave you,
  7741. but your whole computer. And if the RIAA and MPAA get their way, it
  7742. could be law in the US that you <em>have</em> to allow such external control
  7743. over your computer, in the name of preventing "piracy". Apple is
  7744. somewhat better, since you can run the core of Mac OS X without
  7745. signing up for their DRM, but as soon as you use iTunes, you're in
  7746. about the same position as a Windows user. Only Linux has remained a
  7747. fully open OS where <em>you</em> own your computer, and you don't have to
  7748. "license" anything from anyone, or cede any of your rights to anyone,
  7749. to use it.</p>
  7750. </div>
  7751. </content>
  7752. </entry>
  7754. <entry>
  7755. <title type="html">Linux Kernel Site (Not) Cracked</title>
  7756. <category term="/opinions" />
  7757. <id></id>
  7758. <updated>2011-09-02T03:03:00Z</updated>
  7759. <published>2011-09-02T03:03:00Z</published>
  7760. <link rel="alternate" type="text/html" href="" />
  7761. <content type="xhtml">
  7762. <div xmlns="">
  7763. <p>Unless you're a Linux nerd like me, you probably didn't hear that the
  7764. <a href=""></a>
  7765. site, the "home" of the Linux kernel and the "official" place to get a
  7766. copy of its source code, was recently cracked. As far as I can tell
  7767. from the
  7768. <a href="">Internet oracle</a>, this
  7769. hasn't made the news outside of the Linux developer and distribution
  7770. community. If you're a conspiracy theorist, you might be thinking that
  7771. this not making the news is some kind of nefarious scheme to hide
  7772. flaws in the security of Linux. When a bank's server gets cracked,
  7773. everybody finds out in a New York minute. Why should Linux's kernel
  7774. source be any different?</p>
  7776. <p>The answer is in
  7777. <a href="">this article</a>
  7778. on Basically, the cracking of the server was a
  7779. non-event, in security terms, because even though the cracker gained
  7780. root access to the server, he couldn't change any of the kernel source
  7781. code stored there without immediately raising alarms, because that code
  7782. is cryptographically signed in a way that cannot be forged. So the damage
  7783. was limited to having to take the servers offline once the cracking was
  7784. detected, to reinstall their operating systems and restore the stored
  7785. code repositories from backups (which were themselves checked against
  7786. the cryptographic signatures to make sure they were correct).</p>
  7788. <p>As a matter of fact, I wish this event <em>would</em> get wide news coverage,
  7789. but not because it shows any problem with Linux. Quite the opposite:
  7790. it is a perfect example of how professional software development and
  7791. distribution, particularly of something as critical as an operating
  7792. system, is <em>supposed</em> to be done. First, extremely strong security
  7793. precautions were taken against the possibility of a server being
  7794. cracked, even though such events are very rare on well-managed servers.
  7795. Nobody sat back and said, well, we do such a good job of securing our
  7796. servers against intrusion, we don't have to worry about what would be
  7797. compromised if somebody <em>did</em> get in. Second, when the compromise did
  7798. happen, the kernel development community was completely open about it.
  7799. Even though their server was cracked, they quite literally had nothing
  7800. to hide; everything was out in plain sight anyway, and the security
  7801. features that protected the source code have been public knowledge for
  7802. years. All the site maintainers had to do was point to them.</p>
  7804. <p>Oh, by the way: the Linux kernel, <em>and</em> the version control system
  7805. that protects its source code,
  7806. <a href="">git</a>,
  7807. are both free. And yet their developers set a standard of
  7808. professionalism that I strongly suspect is unmatched by
  7809. <a href="">proprietary</a>
  7810. <a href="">systems</a>
  7811. that users pay good money for. Of course, we don't know if the source
  7812. code for those has ever been compromised, because it isn't out in the
  7813. open where anyone can check it. Perhaps that secrecy makes it safe.
  7814. We don't know. With Linux, we <em>know</em>. Which is one reason for what I
  7815. showed in my brief nerd interlude on
  7816. <a href="">Linux's 20th birthday</a>.</p>
  7817. </div>
  7818. </content>
  7819. </entry>
  7821. <entry>
  7822. <title type="html">Another Look At Marbury v. Madison</title>
  7823. <category term="/opinions" />
  7824. <id></id>
  7825. <updated>2011-08-31T21:59:00Z</updated>
  7826. <published>2011-08-31T21:59:00Z</published>
  7827. <link rel="alternate" type="text/html" href="" />
  7828. <content type="xhtml">
  7829. <div xmlns="">
  7830. <p>I just came across an
  7831. <a href="">article</a>
  7832. that shows I'm not the only one who thinks that US Constitutional law
  7833. has gone a little overboard in its interpretation of the
  7834. <a href="">Marbury v. Madison</a>
  7835. Supreme Court decision.
  7836. The article's position is summarized in this
  7837. paragraph, towards the end of the introduction:</p>
  7839. <blockquote>
  7840.  <p>It is the fundamental betrayal of Marbury's premises and Marbury's logic
  7841.  that accounts for nearly all of what is wrong with "constitutional law"
  7842.  today. The twin peaks of constitutional law today are judicial supremacy
  7843.  and interpretive license. Marbury refutes both propositions. Correctly
  7844.  read, Marbury stands for constitutional supremacy rather than judicial
  7845.  supremacy. And constitutional supremacy implies strict textualism as a
  7846.  controlling method of constitutional interpretation, not free-wheeling
  7847.  judicial discretion.</p>
  7848. </blockquote>
  7850. <p>The bit about "strict textualism" is explained further later on in the
  7851. article:</p>
  7853. <blockquote>
  7854.  <p>Marbury's conception of written constitutionalism implies a particular
  7855.  methodology of constitutional interpretation: originalist textualism -
  7856.  that is, the binding authority of the written constitutional text,
  7857.  considered as a whole and taken in context, as its words and phrases
  7858.  would have been understood by reasonably well-informed speakers or
  7859.  readers of the English language at the time. </p>
  7860. </blockquote>
  7862. <p>As I noted in my previous post, this does raise questions when
  7863. considering issues that weren't even contemplated by the Framers, such
  7864. as whether wiretapping constitutes a "search and seizure" under the
  7865. Fourth Amendment, as in
  7866. <a href="">Olmstead v. United States</a>.
  7867. But such questions always end up being questions of classification,
  7868. not principle: whether wiretapping <em>counts as</em> a search and seizure,
  7869. rather than whether evidence obtained by a search and seizure without
  7870. a warrant is admissible (no one disputed that it was not). As I noted
  7871. in my previous post, many of the Court's classifications (such as what
  7872. they think counts as interstate commerce) are highly questionable, and
  7873. the logic suggested by the above quote can be used to see why. Would
  7874. an average American in 1790 have agreed that
  7875. <a href="">growing food on one's own property for one's own use counts as "interstate commerce"</a>?
  7876. I'm guessing not. So on the whole I think the methodology advocated
  7877. in the article is sound.</p>
  7879. <p>The full article is worth a read, and I won't belabor the details of its
  7880. arguments here (though I will note that the article focuses, as I did,
  7881. on Chief Justice's Marshall's statement that the Supreme Court's job is
  7882. to "say what the law is", and how that statement has been taken way out
  7883. of context). However, I can't resist one more quote:</p>
  7885. <blockquote>
  7886.  <p>Marbury truly fits Mark Twain's definition of a "classic": a work that
  7887.  everybody praises but nobody actually reads. Marbury is invoked today
  7888.  for the myth it has become, not for its actual reasoning and logic.</p>
  7889. </blockquote>
  7891. <p>Which is a good brief summation of what I was getting at in my previous
  7892. post. (Please note that this is <em>not</em> a blanket endorsement of everything
  7893. on the Free Republic website, which is where the article is posted. The
  7894. article itself is from the Northwestern University School of Law.)</p>
  7895. </div>
  7896. </content>
  7897. </entry>
  7899. <entry>
  7900. <title type="html">Don't Tread On Our Internet, Part II</title>
  7901. <category term="/opinions" />
  7902. <id></id>
  7903. <updated>2011-08-28T16:55:00Z</updated>
  7904. <published>2011-08-28T16:55:00Z</published>
  7905. <link rel="alternate" type="text/html" href="" />
  7906. <content type="xhtml">
  7907. <div xmlns="">
  7908. <p>In a
  7909. <a href="">previous post</a>
  7910. I mentioned the
  7911. <a href="">Protect IP Act</a>
  7912. as an example of government making things worse instead of better when it
  7913. tries to censor the Internet. Today I came across an article talking about
  7914. another very bad effect that the Protect IP Act would have if it were
  7915. passed: it would
  7916. <a href="">break DNSSEC</a>,
  7917. which is a key security mechanism that lets your computer validate DNS
  7918. records, so that, for example, when you type your bank's URL into your
  7919. browser, you know that you're talking to your bank's server, instead of
  7920. some rogue site that has been set up to impersonate it.
  7921. Of course, as
  7922. the article also notes, this will not actually reduce online copyright
  7923. infringement, since people who really want to infringe can simply
  7924. bypass any blocking technology that is put in place (for example, if the
  7925. US were to mandate DNS filtering, people could just use DNS servers that
  7926. are outside the US). So once again, the law would impose substantial
  7927. burdens on legitimate uses of the Internet, without making a dent in
  7928. illegitimate ones. As I noted when I posted about
  7929. <a href="">my favorite Heinlein quote</a>
  7930. (I warned you I'd be referring to it again),
  7931. whenever you try to fix things by fiat, by controlling people, it
  7932. always ends up being a net loss.</p>
  7934. <p>(By the way, I haven't even gone into the fact that two key organizations
  7935. that are trying to get the government to pass the Protect IP Act, the
  7936. RIAA and MPAA, are doing it to prop up their outdated business models,
  7937. not out of any genuine concern for the people that actually <em>create</em>
  7938. "intellectual property". If they were really concerned about the actual
  7939. artists that create the IP they are selling, they wouldn't go to such great
  7940. lengths to
  7941. <a href="">rip them off</a>.
  7942. But that's a whole other post.)</p>
  7943. </div>
  7944. </content>
  7945. </entry>
  7947. <entry>
  7948. <title type="html">Happy Birthday, Linux!</title>
  7949. <category term="/opinions" />
  7950. <id></id>
  7951. <updated>2011-08-26T16:37:00Z</updated>
  7952. <published>2011-08-26T16:37:00Z</published>
  7953. <link rel="alternate" type="text/html" href="" />
  7954. <content type="xhtml">
  7955. <div xmlns="">
  7956. <p>Twenty years ago (yesterday, to be exact, but cut me some slack here),
  7957. Linus Torvalds posted a message to the Usenet newsgroup comp.os.minix,
  7958. announcing that he was working on a free operating system and wanted
  7959. to know what features people were interested in. The original message
  7960. is on Google Groups
  7961. <a href="">here</a>.
  7962. So it's time for another
  7963. <a href="">brief nerd interlude</a>:</p>
  7965. <div class="codehilite"><pre><span class="gp">peter@localhost:~$</span> uname
  7966. <span class="go">Linux</span>
  7967. </pre></div>
  7970. <p>At some point I'll do a longer post on why the above is true, but for
  7971. now I think I'll just let it stand by itself. Thanks, Linus, for starting
  7972. it all 20 years ago, and thanks to all the developers and distributions
  7973. who have kept it going.</p>
  7974. </div>
  7975. </content>
  7976. </entry>
  7978. <entry>
  7979. <title type="html">My Favorite Heinlein Quote</title>
  7980. <category term="/opinions" />
  7981. <id></id>
  7982. <updated>2011-08-14T22:17:00Z</updated>
  7983. <published>2011-08-14T22:17:00Z</published>
  7984. <link rel="alternate" type="text/html" href="" />
  7985. <content type="xhtml">
  7986. <div xmlns="">
  7987. <p>I've already referred to my favorite Heinlein quote
  7988. <a href="">once</a>,
  7989. and I'm sure I'll be doing it again, so I figured I might as well lay
  7990. it out in full and unpack in detail why it's my favorite quote. Here
  7991. it is, from <em>Time Enough For Love</em>, as noted on
  7992. <a href="">wikiquote</a>:</p>
  7994. <blockquote>
  7995.  <p>The human race divides politically into those who want people to be
  7996.  controlled and those who have no such desire. The former are idealists
  7997.  acting from highest motives for the greatest good of the greatest
  7998.  number. The latter are surly curmudgeons, suspicious and lacking in
  7999.  altruism. But they are more comfortable neighbors than the other sort.</p>
  8000. </blockquote>
  8002. <p>I'll start the unpacking with the following hypothesis: human progress,
  8003. which I will define in a moment, only comes from people of the second
  8004. class.</p>
  8006. <p>Heinlein himself, of course, was of the second class. All that business
  8007. about the people of the first class being "idealists" and those of the
  8008. second being surly curmudgeons was irony, particularly sharp irony
  8009. since it describes well how people of both classes see themselves, but
  8010. while people of the first class do tend to see those of the second as
  8011. surly curmudgeons, people of the second, like Heinlein and like me, do
  8012. <em>not</em> see those of the first as actually doing good. We see them (or
  8013. more precisely those of them who want to actually <em>do</em> the controlling
  8014. of others, as opposed to just wanting others to be controlled; I'll
  8015. expand on that later in this post) as infernal busybodies, whose general
  8016. effect on human history is best summed up by
  8017. <a href="">this quote</a>
  8018. from Rick Moen:</p>
  8020. <blockquote>
  8021.  <p>Experience suggests that, if we were able to kill off the
  8022.  well-intentioned at birth, as a preventative measure, the leftover
  8023.  evil-doers would be small potatoes, in comparison.</p>
  8024. </blockquote>
  8026. <p>But I'm getting ahead of myself.</p>
  8028. <p>How do I define "human progress"? Simple: it is anything that expands
  8029. the choices people have for how to live their lives, without infringing
  8030. on other people's choices. The reason I like this definition is that
  8031. if you propose it to people of the first class, you will have them
  8032. nodding their heads in agreement; yes, indeed, that's what progress
  8033. means. And yet, when we take this definition of progress to its logical
  8034. conclusion, we'll find that it confirms my hypothesis above: actual
  8035. human progress always comes from people of the second class. And so,
  8036. if you are of the first class, and you are in favor of human progress,
  8037. the best thing you can do to help is to switch to the second class.</p>
  8039. <p>Consider this simple question: do you, yourself, want to <em>be</em>
  8040. controlled? No, of course not. Nobody does. Everyone wants complete
  8041. freedom for himself. Most of us (grudgingly) accept that we can't have
  8042. complete freedom, because we have to share the planet with others. But
  8043. we want as much as we can get. We want it because we want, as the
  8044. definition of true human progress above says, to have as many choices
  8045. as possible for how to live our lives. We don't want to be restricted
  8046. by someone else's idea of what is "good" for us, or what is "proper",
  8047. or what is "right". We want, as responsible adults, the freedom to
  8048. decide those things for ourselves.</p>
  8050. <p>If you are of Heinlein's second class of people, this is no problem. You
  8051. don't want to be controlled, and you don't want to control others. The
  8052. situation is symmetric. You also, as Heinlein noted, will be a more
  8053. comfortable neighbor, because you aren't always sticking your nose into
  8054. other people's business.</p>
  8056. <p>But if you are of the first class, you have a cognitive dissonance
  8057. problem. You don't want to be controlled, but you do want others to
  8058. be controlled, even though the others themselves don't want to be
  8059. controlled. This situation is not symmetric, but our evolved ape
  8060. brains are very adept at inventing rationalizations for the asymmetry.
  8061. We say that people need to be controlled in the name of "security",
  8062. for example. Or we say that they need to be controlled because
  8063. otherwise they will corrupt our children. And so on.</p>
  8065. <p>But, you might think, wouldn't people's natural desire not to <em>be</em>
  8066. controlled stop these rationalizations from working? Well, that's where
  8067. somebody of class one, somewhere way back in human history, invented a
  8068. neat trick. A person of class one wants other people to be controlled,
  8069. but this does not mean every person of class one wants to do the
  8070. controlling. Like everything else, it's easier to get someone or
  8071. something else to do the hard work if at all possible. So when a person
  8072. of class one comes along who <em>does</em> want to do the controlling, they
  8073. can simply attach a little rider to all those rationalizations I
  8074. mentioned above: yes, people need to be controlled to protect our
  8075. security, or to protect the children, or whatever, and <em>I will take
  8076. care of all that if you just give me the power to do the controlling</em>.
  8077. And it works! People of class one, it turns out, will happily submit
  8078. to being controlled if they think that they are thereby obtaining
  8079. security from the crazy antics of all those other loons who don't
  8080. want to be controlled.</p>
  8082. <p>And note that if you are the lucky person of class one who gets to do
  8083. the controlling, you get the best of both worlds: you get to claim to
  8084. be a person of class one, so you don't set off the alarms that people
  8085. of class two set off in the minds of all the other people of class one,
  8086. but you also get all the freedom that people of class two want, <em>plus</em>
  8087. power. Nice work if you can get it. We'll come back to this in a
  8088. little bit.</p>
  8090. <p>I have stated the above in roundabout terms, but of course what we
  8091. have here, in plain English, is a dominance-submission hierarchy.
  8092. It is far older than the human species. It has been one of
  8093. evolution's handy solutions to organizing social animals since its
  8094. invention. Cognitive scientists are finding that we have "modules" for
  8095. this behavior pattern pretty much hard-wired into our brains. From
  8096. that standpoint, it is not a mystery why the trick works. What is a
  8097. mystery, at least from this viewpoint, is why it doesn't work on
  8098. everyone: why there are any people at all of the second class.</p>
  8100. <p>Of course old-fashioned liberal philosophy has the answer to this one:
  8101. human beings are not <em>just</em> animals. Somehow, we managed to cross a line
  8102. and become <em>persons</em>. And once you're a full-fledged person, the whole
  8103. system of dominance and submission looks the way socialism looked to
  8104. E. O. Wilson:
  8105. <a href="">great idea, wrong species</a>.
  8106. Human beings are not insects. You don't have to believe that humans were
  8107. made in the image of God to believe this; indeed, you don't even have to
  8108. believe that there <em>is</em> a God. But you <em>do</em> have to believe that human
  8109. beings are persons, and that persons, while they are evolved animals,
  8110. are not <em>just</em> animals. We have some extra quality that makes us unique,
  8111. different from other animals. This extra quality doesn't have to be
  8112. anything mystical or mysterious; the simple fact that we are capable of
  8113. even thinking about and discussing the question of "human progress" at
  8114. all will do. Insects don't have elaborate debates on how best to run
  8115. the hive.</p>
  8117. <p>In other words, as soon as you have the ability to even think about
  8118. a concept like the "good" life, as soon as you can even consider one
  8119. potential future vs. another in the abstract and label one as "better",
  8120. over a whole lived life instead of just limited to the next meal, you
  8121. have crossed a line, and you have only two choices. One is to exit the
  8122. whole scheme that evolution evolved for organizing social animals; it
  8123. quite simply no longer applies to you, because it evolved in animals
  8124. that didn't have the conceptual ability that you have. This makes you
  8125. a person of the second class: you don't want to be controlled, and you
  8126. don't want to control anyone else either. As I noted above, this
  8127. situation is symmetric, and it makes for a minimum of hassle.</p>
  8129. <p>The other choice is to yield to the temptation of the Ring, as
  8130. <a href="">Tolkien</a>
  8131. would have put it. You find yourself in possession of a great secret:
  8132. you now know that it is possible to <em>not</em> be controlled, that you can
  8133. have a "good" life completely free of the dominance-submission hierarchy
  8134. that evolution saddles social animals with. But why should you share
  8135. this secret with anyone else? Why not <em>use</em> it instead, to put yourself
  8136. at the top of the hierarchy?</p>
  8138. <p>Of course, to do that consciously, with full malice aforethought, as it
  8139. were, you would have to be <em>evil</em>, wouldn't you? But there's a handy
  8140. dodge for this too. Taking power, putting yourself at the top of the
  8141. hierarchy, just for yourself would be evil, but if you do it <em>for a
  8142. good cause</em>, that is quite another matter. After all, there is so much
  8143. that is <em>wrong</em> with the world, and yet nobody seems to be <em>fixing</em> all
  8144. these wrong things; they just seem to go on and on. But you, of
  8145. course, have a great idea for really fixing something, really <em>making
  8146. a difference</em>. All you need is the power to do so. And we've already
  8147. seen how you can get that: just work the trick we saw above on all the
  8148. other people of class one, and get them to help you impose your system
  8149. of control on the people of class two.</p>
  8151. <p>But doesn't this count as human progress? Okay, so it was done using
  8152. the Ring; but even so, didn't something get fixed that needed fixing?
  8153. Maybe, for a time anyway. But the fixes, even if they are fixes at
  8154. first, don't last. What does last is the system of power that the
  8155. well-intentioned people of class one construct to implement their
  8156. fixes. And the first thing such a system does is to make sure that
  8157. nobody else has a chance at the top. And so, as Richard Feynman said
  8158. in his eloquent essay,
  8159. <a href="">The Value Of Science</a>,</p>
  8161. <blockquote>
  8162.  <p>we suppress all discussion, all criticism, saying, "This is it, boys,
  8163.  man is saved!" and thus doom man for a long time to the chains of
  8164.  authority, confined to the limits of our present imagination.</p>
  8165. </blockquote>
  8167. <p>In other words, as soon as you use power to "fix" something by fiat,
  8168. by controlling people, you always end up taking away more choices
  8169. than you open up. It always turns out to be a net loss, not a net gain.
  8170. And so no real human progress ever comes from people of class one.</p>
  8172. <p>It's often difficult to see this because the perceived gains from
  8173. controlling people are visible and immediate, while the losses from
  8174. restricting other choices are often invisible and take time to be
  8175. felt. For example, in the
  8176. <a href="">previous post</a>
  8177. where I referred to my favorite Heinlein quote, I talked about efforts
  8178. to censor the Internet in the name of some "good cause", such as
  8179. preventing computer viruses or child porn. Advocates of such schemes
  8180. always talk about the obvious, visible benefits, but they never talk
  8181. about the hidden costs: all the good things that could be done on the
  8182. Internet that haven't been thought of yet, but which would be harder,
  8183. or even impossible, on a censored Internet. We can't know today how
  8184. valuable those things might be, any more than someone in, say, 1995
  8185. could know how valuable Google would be today. But that's exactly why
  8186. we can't afford to, as Feynman said, confine future innovators to the
  8187. limits of our present imagination.</p>
  8189. <p>The conflict I am describing is an age-old one in philosophy:
  8190. which is more important, virtue or liberty? W. H. Auden described it
  8191. as the difference between the European and the American viewpoints;
  8192. an article in Life magazine in November 1947, which is
  8193. <a href=";pg=PA38&amp;lpg=PA38#v=onepage&amp;q&amp;f=false">readable online</a>
  8194. via the wonder of Google Books, quotes him thus:</p>
  8196. <blockquote>
  8197.  <p>Europe's faith, as a recent essay by W. H. Auden puts it, involves
  8198.  the proposition that virtue is prior to liberty: you must do and think
  8199.  right, by free choice if possible, but in any case you <em>must</em> do and
  8200.  think right. The American proposition is that liberty is prior to
  8201.  virtue: it is better to choose wrong than to have right chosen for you,
  8202.  since freedom of choice "is the human prerequisite without which virtue
  8203.  and vice have no meaning."</p>
  8204. </blockquote>
  8206. <p>(I believe the Auden essay in question is in <em>The Dyer's Hand</em>, but I
  8207. have not found it online.)</p>
  8209. <p>It should be obvious that Heinlein and I and everybody of class two
  8210. are on the American side in this. "Freedom of choice" is simply
  8211. another unique quality of humans that makes us persons, unlike other
  8212. animals. (I could argue that it's actually the same quality that I
  8213. was talking about above, but that will have to wait for another post.)
  8214. The only thing I would add is that the European proposition assumes,
  8215. without proof, that it is even <em>possible</em> to ensure virtue by making
  8216. it prior to liberty; and if history teaches anything, it teaches that
  8217. that is a pipe dream. Even the most draconian systems of control in
  8218. history, such as the medieval Christian Church, came nowhere near
  8219. ensuring that people would "do and think right". If anything, people
  8220. have probably been more virtuous, on average, in societies that
  8221. have given them the liberty to decide for themselves what "virtue" is.
  8222. Which is, of course, precisely the American point.</p>
  8224. <p>Does any of this guarantee that real progress will come from people
  8225. of class two? Of course not; there are no guarantees. But at least
  8226. we of class two recognize the plain fact that human beings <em>cannot</em> be
  8227. controlled. We are simply too complex. Still less can reality itself
  8228. be controlled. That may seem strange to say in this scientific age,
  8229. but any real scientist will be the first to tell you how vast is our
  8230. ignorance. The supposed "security" that the people of class one at the
  8231. top of the hierarchy are promising to the other people of class one
  8232. underneath them is an illusion; it can't actually be had at any
  8233. price. The best we can do is to let everyone use their own judgment
  8234. about how to make the choices they have. This certainly isn't perfect,
  8235. but at least it makes progress possible, which is more than can be
  8236. said of the alternative.</p>
  8237. </div>
  8238. </content>
  8239. </entry>
  8241. <entry>
  8242. <title type="html">Two Cultures Redux: But Wait, There's More</title>
  8243. <category term="/rants" />
  8244. <id></id>
  8245. <updated>2011-08-09T02:40:00Z</updated>
  8246. <published>2011-08-09T02:40:00Z</published>
  8247. <link rel="alternate" type="text/html" href="" />
  8248. <content type="xhtml">
  8249. <div xmlns="">
  8250. <p>The New York Times, which is certainly a bastion of the liberal arts
  8251. types if anywhere is, has been running a debate about law school that
  8252. is similar to the one about college in general that I discussed in
  8253. my post a couple of weeks ago on the
  8254. <a href="">two cultures</a>.
  8255. The question under debate is: "Should the standard three years of law
  8256. school, followed by the bar exam, be the only path to a legal career?"</p>
  8258. <p>I won't bother canvassing most of the responses, which are predictable.
  8259. What got my attention were a couple of responses that got into that same
  8260. "two cultures" territory that prompted my last rant.</p>
  8262. <p>First up is a law professor who wants us to know that law school is
  8263. <a href="">not a trade school</a>:</p>
  8265. <blockquote>
  8266.  <p>At the risk of sounding "liberal artsy," law school should emphasize
  8267.  educated citizenship. It prepares people to become leaders in our
  8268.  society, which makes it imperative that they be rigorously trained as
  8269.  thinkers. They will become stewards of policies that affect our everyday
  8270.  lives: in our schools, our jobs and our families. All of this
  8271.  responsibility, in diverse fields, comes from legal education.</p>
  8272. </blockquote>
  8274. <p>I hardly know where to start, but perhaps that phrase "liberal artsy"
  8275. will do. So a degree in, say, physics, or chemistry, or engineering
  8276. doesn't qualify one to be a "steward" of important policies? Not even
  8277. a degree in one of those other liberal arts, like literature? "<em>All</em>
  8278. of this responsibility" can only be had if you first get a law degree?
  8279. Wow.</p>
  8281. <p>The standard liberal artsy argument, of course, is that one doesn't
  8282. have to actually <em>understand</em> the details of a technical field, like
  8283. physics or chemistry, in order to be a "steward of policies" that are
  8284. dependent on such a field. One can always learn enough to know which
  8285. experts in the field to trust. I would hope that my last rant, and the
  8286. observations of C. P. Snow that I quoted there, would be sufficient
  8287. refutation of that, but in case you don't think it is, consider this
  8288. from
  8289. <a href="">Paul Graham</a>:</p>
  8291. <blockquote>
  8292.  <p>Try this thought experiment. A dictator takes over the US and sends
  8293.  all the professors to re-education camps. The physicists are told they
  8294.  have to learn how to write academic articles about French literature,
  8295.  and the French literature professors are told they have to learn how
  8296.  to write original physics papers. If they fail, they'll be shot. Which
  8297.  group is more worried?</p>
  8299.  <p>We have some evidence here: the famous parody that physicist Alan Sokal
  8300.  got published in Social Text. How long did it take him to master the art
  8301.  of writing deep-sounding nonsense well enough to fool the editors? A
  8302.  couple weeks?</p>
  8304.  <p>What do you suppose would be the odds of a literary theorist getting a
  8305.  parody of a physics paper published in a physics journal?</p>
  8306. </blockquote>
  8308. <p>My experience leads me to a similar conclusion. I would much rather have
  8309. a scientifically educated person making policy and having to learn the
  8310. political stuff as they go, then have a liberal arts educated person
  8311. making policy and having to learn the science as they go.</p>
  8313. <p>But even that conclusion assumes that those are the only choices. Why
  8314. do they have to be? And no, I'm not talking about those
  8315. "interdisciplinary" degrees. I'm talking about the misconception that
  8316. there should be any rule for choosing "leaders in our society", other
  8317. than doing stuff that works. No one group, no one academic discipline,
  8318. has a lock on "educated citizenship". This is America; we are <em>all</em>
  8319. supposed to be educated citizens. And there are no degrees in that
  8320. discipline; we all learn it on the job.</p>
  8322. <p>Even focusing on "degrees" in the first place ignores all the other
  8323. areas of experience that are highly relevant to citizenship, such as
  8324. serving in the military, or even something simple like being a good
  8325. neighbor. If we look at the track records of "stewards of policies",
  8326. it certainly doesn't look to me like law degrees, or indeed any degrees,
  8327. stack up very well against those other types of experiences. (The hard
  8328. sciences know this; people outside the standard "academic" circles can
  8329. get work published if it's good. Einstein, of course, was a Patent
  8330. Office clerk when he published his five classic scientific papers, in
  8331. 1905, in the most prestigious physics journal in the world at that
  8332. time, Annalen der Physik.) Have our Presidents, or Senators, or
  8333. Representatives, with law degrees been better, on balance, than
  8334. those with other backgrounds? Congress is certainly chock full of
  8335. lawyers now, and look at it.</p>
  8337. <p>But wait, there's more. Another law professor says that a law degree
  8338. is
  8339. <a href="">priceless</a>
  8340. because of the experience it gives you:</p>
  8342. <blockquote>
  8343.  <p>My own decision to attend law school was based in part upon my
  8344.  perception, still shared by those who rush our doors, that a legal
  8345.  education provides an unparalleled opportunity to understand the
  8346.  intersection of private and public power, to explore the rationale for
  8347.  the organization of human society and to participate more knowledgeably
  8348.  and effectively in every aspect of human endeavor.</p>
  8349. </blockquote>
  8351. <p>I guess I can't quarrel with the "intersection of private and public
  8352. power" part; lawyers certainly get to see that first-hand. But unlike
  8353. the law professor, I regard that as a bug, not a feature. As Chesterton
  8354. <a href="">said</a>:</p>
  8356. <blockquote>
  8357.  <p>[T]he horrible thing about all legal officials, even the best, about
  8358.  all judges, magistrates, barristers, detectives, and policeman, is not
  8359.  that they are wicked (some of them are good), not that they are stupid
  8360.  (several of them are quite intelligent), it is simply that they have
  8361.  got used to it. Strictly they do not see the prisoner in the dock; all
  8362.  they see is the usual man in the usual place. They do not see the awful
  8363.  court of judgment; they only see their own workshop.</p>
  8364. </blockquote>
  8366. <p>It seems to me that the best way to "participate more knowledgeably
  8367. and effectively" in society is to actually <em>participate</em> in it, not to
  8368. sit above it and tinker with the rules. And I'm all for exploring the
  8369. "rationale for the organization of human society", but again, that's
  8370. part of educated citizenship, and no one group or discipline has a
  8371. special private line to the right answers. We can only judge by the
  8372. actual track record.</p>
  8374. <p>In fact, the law professor says so himself:</p>
  8376. <blockquote>
  8377.  <p>When the history of legal education is written, the important
  8378.  question...will be, "Did our legal education system deliver equal
  8379.  justice under law?"</p>
  8380. </blockquote>
  8382. <p>The answer in our society today, I submit, is quite often "no". Too
  8383. bad the professor thought the question was just rhetorical.</p>
  8384. </div>
  8385. </content>
  8386. </entry>
  8388. <entry>
  8389. <title type="html">A Brief Nerd Interlude</title>
  8390. <category term="/general" />
  8391. <id></id>
  8392. <updated>2011-08-06T02:19:00Z</updated>
  8393. <published>2011-08-06T02:19:00Z</published>
  8394. <link rel="alternate" type="text/html" href="" />
  8395. <content type="xhtml">
  8396. <div xmlns="">
  8397. <p>For non-nerd readers, I promise I won't do this very often, but once in
  8398. a while I just have to get these sorts of things out of my system. Does
  8399. anyone else find the following (from a transcript of a short Unix shell
  8400. session) a little weird?</p>
  8402. <div class="codehilite"><pre><span class="gp">peter@localhost:~$</span> <span class="nb">true</span>
  8403. <span class="gp">peter@localhost:~$ echo $</span>?
  8404. <span class="go">0</span>
  8405. <span class="gp">peter@localhost:~$</span> <span class="nb">false</span>
  8406. <span class="gp">peter@localhost:~$ echo $</span>?
  8407. <span class="go">1</span>
  8408. </pre></div>
  8411. <p>The test of whether you're a nerd reader or not, of course, is first,
  8412. whether the above makes sense to you, and second, if it does, do you
  8413. immediately see why I find it weird?</p>
  8415. <p>(And for the really nerdy nerds who are brimming full of explanations
  8416. of why it's not really weird at all, it makes perfect sense for things
  8417. to be that way, yes, I know why it's that way. I write Unix shell programs
  8418. too. I just said it was a little weird, not that it was bad engineering.
  8419. It isn't, all things considered.)</p>
  8420. </div>
  8421. </content>
  8422. </entry>
  8424. <entry>
  8425. <title type="html">Wow, Sometimes Things Actually Work</title>
  8426. <category term="/rants" />
  8427. <id></id>
  8428. <updated>2011-07-29T21:44:00Z</updated>
  8429. <published>2011-07-29T21:44:00Z</published>
  8430. <link rel="alternate" type="text/html" href="" />
  8431. <content type="xhtml">
  8432. <div xmlns="">
  8433. <p>In the interest of keeping the record honest following my
  8434. <a href="">last post</a>,
  8435. it's only fair to report that I have now been pleasantly surprised.
  8436. Not relishing the prospect of a phone call (including most probably a
  8437. lengthy time spent on hold), I decided to try email first. Believe it
  8438. or not, my email was actually acted on within a day, and I have now
  8439. received confirmation that the claim is being handled properly. Whoever
  8440. read my email and did the right thing, I doubt you're reading this, but
  8441. thanks. You've saved my wife and me (and your company as well) a
  8442. significant amount of hassle. It's nice to be reminded that things can
  8443. actually work.</p>
  8444. </div>
  8445. </content>
  8446. </entry>
  8448. <entry>
  8449. <title type="html">Beam Me Up Scotty, There's No Intelligent Life Here</title>
  8450. <category term="/rants" />
  8451. <id></id>
  8452. <updated>2011-07-28T03:10:00Z</updated>
  8453. <published>2011-07-28T03:10:00Z</published>
  8454. <link rel="alternate" type="text/html" href="" />
  8455. <content type="xhtml">
  8456. <div xmlns="">
  8457. <p>Today I had one of those experiences that make you wonder how anything
  8458. ever gets accomplished in our society.
  8459. A couple of days ago I faxed in a
  8460. claim form to our health care spending account for some prescription
  8461. copays. After I faxed the form, I realized that I hadn't signed it, so
  8462. I signed it and faxed it again.</p>
  8464. <p>Today I got an email from the agency that processes the forms. It
  8465. referenced two claim forms received. The first claim form was marked as
  8466. "pending" with the following note:</p>
  8468. <blockquote>
  8470. </blockquote>
  8472. <p>The second claim form was marked as "denied", with the following note:</p>
  8474. <blockquote>
  8476. </blockquote>