Saturday, August 30, 2008

Russia, Georgia, and Backwards Priorities

I have no special expertise on the Georgian conflict, nor much sympathy for either side. I am, however, struck by the fact that most of the international condemnation is aimed at the most defensible, not the least defensible, part of what Russia has done.

Russian recognition of two breakaway provinces, both of which seem to have been effectively independent for over a decade, may or may not be a good idea, but I find it hard to see any reason to be outraged over it. The current governments there may be better or worse than the government of Georgia, but they pretty clearly have more support from the local population--and in any case, governments don't decide what other governments to recognize primarily on the basis of whether they approve of them. Yet it is the Russian recognition of South Ossetia and Abkhazia that western governments have been expressing outrage over.

What is outrageous, in terms of international law and norms as I understand them, is the fact that the Russian military continues to hold territory well inside Georgia and well outside South Ossetia. That is both a violation of public Russian promises and an invasion of undisputed territory. But nobody--at least, none of the nations that are condemning Russia at the moment--seems to be paying very much attention to it.

Friday, August 29, 2008

Who is Against Evolution?

"And the religious right has been the chief force against teaching evolution."
(Quoted from Barbara Forrest, a Southeastern Lousiana University philosophy professor and prominent critic of creationist science.)

It's a widespread view, but true in only a narrow sense. People who say they are against teaching the theory of evolution are very likely to be Christian fundamentalists. But people who are against taking seriously the implications of evolution, strongly enough to want to attack those who disagree, including those who teach those implications, are quite likely to be on the left.

Consider the most striking case, the question of whether there are differences between men and women with regard to the distribution of intellectual abilities or behavioral patterns. That no such differences exist, or if that if they exist they are insignificant, is a matter of faith for many on the left. The faith is so strongly held that when the president of Harvard, himself a prominent academic, merely raised the possibility that one reason why there were fewer women than men in certain fields might be such differences, he was ferociously attacked and eventually driven to resign.

Yet the claim that such differences must be insignificant is one that nobody who took the implications of evolution seriously could maintain. We are, after all, the product of selection for reproductive success. Males and females play quite different roles in reproduction. It would be a striking coincidence if the distribution of abilities and behavioral patterns that was optimal for one sex turned out to also be optimal for the other, rather like two entirely different math problems just happening to have the same answer.

The denial of male/female differences is the most striking example of left wing hostility to the implications of Darwinian evolution, but not the only one. The reasons to expect differences among racial groups as conventionally defined are weaker, since males of all races play the same role in reproduction, as do females of all races. But we know that members of such groups differ in the distribution of observable physical characteristics--that, after all, is the main way we recognize them. That is pretty strong evidence that their ancestors adapted to at least somewhat different environments.

There is no a priori reason to suppose that the optimal physical characteristics were different in those different environments but the optimal mental characteristics were the same. And yet, when differing outcomes by racial groups are observed, it is assumed without discussion that they must be entirely due to differential treatment by race. That might turn out to be true, but there is no good reason to expect it. Here again, anyone who argues the opposite is likely to find himself the target of ferocious attacks, mainly from people on the left.

Next consider the whole nature/nurture debate, in which the left has, for half a century or more, mostly taken a strong pro-nurture position. It is hard to see how humans could have evolved intelligence if intelligence is not heritable.

Finally, consider the question of how maleable human nature is or isn't. It is not logically impossible that we evolved as general purpose computers, with all details determined by the program, not the hardware. But it does not seem likely, given the obvious advantages of hardwiring in whatever rules worked in the environment where we evolved. Nor does it seem plausible given that most of our evolutionary history predates human rationality, making it likely that humans retain quite a lot of pre-human traits.

To be fair, one of the most important supporters of the evolutionary view in this particular context is also a prominent left wing intellectual: Noam Chomsky. His professional reputation, after all, was largely due to convincing arguments for the proposition that human linguistic abilities were to a large extent hardwired. I am sure he is not the only example; it is possible to defend conclusions, left, right, libertarian, or other in a variety of different ways. Not everyone on the left rejects the conclusions of evolution, any more than all Christians do.

Nonetheless, I think there is a pretty clear pattern. Almost everyone on the left believes that he believes in evolution. Yet I find it hard to think of any proposition popular on the left that is deduced from that belief. And, as I think I have shown, important dogmas of much of the left are inconsistent with it.

[An interesting discussion of evidence and theory on male/female differences is webbed here.]

Business Opportunity for a Linux Geek

In a recent post, I discussed my attempts to select among subnotebook models. It occurs to me that my problems could be someone else's profit.

The Acer Aspire One comes either with Linpus Lite and an 8GB flash disk, with Windows XP and a 120GB hard drive, or with Windows XP, 160GB, and a six cell battery. What I want is the 120GB model dual booting XP and Linux, probably Linpus, since Acer has presumably optimized their version for the machine. I expect there are other people who there who would like the 120GB or 160GB model with either Linpus or some other Linux.

It looks, from poking around the web, as though someone familiar with Linux could,with a little effort, convert the XP versions to either Linux or dual boot. Once the initial work of figuring out how to do it and preparing a suitable CD or flash disk had been done, doing it again should be very little work. I would happily pay $50 plus shipping cost for the service, and I expect that a significant number of other people would as well.

If I'm right, it's a business opportunity.

Obama and Community Organizing

Having said some positive things about Obama in the past, I thought it worth noting one negative point--not so much about Obama as about an argument offered for Obama. His supporters interpret his decision to go into community organizing instead of joining a law firm as evidence of his good moral character. In fact, it is nothing of the sort. For a young man aiming at a career in politics, especially a black man in an urban setting, community organizing is an investment, a way of building up contacts and other resources that will be useful at the next stage of that career.

To be fair, I should also say that considering Obama's behavior as that of an aspiring politician weakens the argument being made by some on the right about his past association with people on the left, in particular William Ayers, an unrepentant ex-Weatherman. The Hyde Park neighborhood of Chicago, which is where I grew up, is the sort of place where an ex-Weatherman can easily enough be a prominent figure. You don't start a career in politics by going out of your way to refuse to shake the hands of people who lots of your constituents respect, or even refusing to sit with them on a board of directors or share a platform with them--not, at least, unless you are aiming at a deliberate political gesture. And making a point of how hostile you are to left wing radicals, while it might be useful for a Democrat running for President, would not be prudent for a Democrat seeking political support in Hyde Park.

And while on the subject of the election, I should probably say that, while I have no idea how good a President Sarah Palin would make, she sounds, from a brief look at old interviews, like a nice lady.

Thursday, August 28, 2008

Subnotebooks [Warning: Geekish Content]

Some time back I got an Asus eee 900, a two pound laptop, for traveling. It's a nice machine and worked well for the trip I got it for, but it has one serious fault—the keyboard. The right shift key is small and next to the up arrow, with the result that when I try to use the right shift while word processing I am quite likely to suddenly find myself typing two lines above where I should be. A further problem is that, because the left shift and the shift lock are small, it's easy to hit the latter when I want to hit the former.

Neither of these turns out to be a problem for my daughter, who uses a fast hunt and peck style of typing rather than touch typing. I have accordingly sent my eee off to college with her—its size and weight make it an ideal machine for taking notes in class—and am considering a replacement.

The improved version of my machine, the 901, has much longer battery life but essentially the same keyboard. The larger versions—the eee 1000 and its direct competitor, the MSI WIND, which have 10" screens instead of 8.9"—have satisfactory keyboards but are noticeably bigger and heavier hence less portable. I already have a MacBook, and although these machines are smaller and lighter I'm not sure they are enough smaller and lighter to be worth giving up its advantages or paying the cost of a second machine.

It looks, however, as though I have found a solution. The Acer Aspire One is a little wider than my eee 900 but about the same weight. It has an entirely satisfactory keyboard—I think as good as the ten inch machines. And it is less expensive than any of the others, currently $329 in its least expensive configuration.

That, however, does not end my purchasing dilemma (perceptive readers may suspect that I enjoy purchasing dilemmas). The least expensive version of the Aspire uses a version of Linux and a flash disk. Two other versions, one of which doesn't yet seem to be available, use Windows XP and a physical hard disk (120 or 160 GB vs 8 GB on the flash disk). Linux is interesting, and the version that comes with the Aspire is said to boot in about 15 seconds, which is impressive and convenient. But XP means that I can run World of Warcraft (slowly) without fiddling around with WINE or some other kludge for running Windows programs under Linux, something I tried unsuccessfully on my eee.

The obvious solution is to get one of the XP versions and set it up to dual boot with XP and the Aspire's version of Linux. Doing that should be possible; I'm not sure how hard it is. The very best solution would be for Acer to offer the machine in that configuration. They already have an XP license, Linux is free, so it should cost them practically nothing and I would happily pay for the convenience. I could use Linux most of the time, for the fast boot, lower demand on system resources, and general fun of playing with it, but XP when I wanted to run a Windows program.

All of which leads me to a puzzle and a question. I can find webbed figures on how fast the Aspire boots in its Linux/flash disk configuration (very). I can find complaints about how slowly it boots if you install XP on that configuration—I gather the flash disk comes formatted in a form that XP has a hard time writing to. But I cannot find any figures on how fast the hard drive version boots either in XP, which it comes with, or in Linux, which presumably can be installed on it.

My current plan is to wait until both the flash disk and the hard disk versions are actually available in stores I can get to—the former already is, which is how I know how good the keyboard is. Then I can do the experiment myself and produce two of the three numbers I want. Finding out how fast Linux boots on the hard disk version, however, will require the cooperation of someone who has the hard disk model and has installed Linux, preferably the Aspire version, which apparently boots faster than other versions on their machine.

I don't suppose any of my readers ... .
[Later Addition]

I found a store with the 120 GB XP version. Boot time is about 75 seconds. I also timed a Linux/flash disk machine. Boot time is about 15-20 seconds.

One commenter points out that I can always suspend the XP, which I gather corresponds to sleeping a Macintosh. I don't know how much power a suspended machine uses; my impression is that if I leave my Mac laptop asleep for several days, it will be low on power when it wakes up. Also, of course, airlines require you to shut down a machine for takeoff and landing, and I assume sleep doesn't qualify.

Using the XP machine reminded me that I don't much like Windows. On the other hand, the Linux machine comes with an interface designed for people not very used to computers. It's possible to get at some additional features of the OS via a command in the terminal window, but I haven't yet found a way of converting the interface to a standard windowed one, similar to the Mac or Windows interface. That may be one reason why some users replace the Linux variant that comes on the machine with an alternative version.

I'll probably end up with the Linux/Flash disk version, but I'm still intrigued by the possibility of a hard disk and dual boot.

Thursday, August 21, 2008

Is H.L. Mencken Alive and Well at the NYT?

Mencken's famous bathtub hoax was an invented history of the bathtub, designed to appeal to what readers wanted to believe about the ignorance and irrationality of people in the past. He published it as a demonstration of human credulity, reported with glee on how many people repeated it as gospel despite its obvious inconsistency with easily established historical facts, published multiple retractions pointing out how obviously false it was—and, by his account, never managed to kill the story.

My previous post pointed to a modern equivalent. The NYT (and, I think, the AP, but not Reuters, which got it right) misread a GAO report in a way that drastically altered its meaning, converting it from a plausible but boring result (a substantial majority of corporations reported no taxable income in at least one year out of an eight year period) to a wildly implausible result that nicely fitted what a lot of people wanted to believe (two-thirds of corporations reported no taxable income over that eight year period). They simultaneously made another mistake almost as bad, calculating what the corporations "should" have owed on the basis of their revenue, not their profit. The Times discovered the latter mistake and corrected it; so far as I can tell, they have not yet noticed the former mistake.

Googling around, I found an enormous numbers of online references to the story. So far, I have not found a single one, other than my piece on this blog, that spotted the mistake. I've posted the actual facts on a fair number of them, but it's like a teacup in a tempest. I have no doubt that, years from now, millions of people will still remember the scandalous, and wholly imaginary, fact from the Times article.

Wednesday, August 20, 2008


According to a New York Times Story:

"Two out of every three United States corporations paid no federal income taxes from 1998 through 2005, according to a report released Tuesday by the Government Accountability Office, the investigative arm of Congress."

According to a Reuters story on the same report:

"The Government Accountability Office said 72 percent of all foreign corporations and about 57 percent of U.S. companies doing business in the United States paid no federal income taxes for at least one year between 1998 and 2005."

There is a huge difference between the two claims. It isn't in the least surprising that a majority of corporations had at least one year during the period with no taxable income. It would be quite surprising if a majority had no taxable income at all throughout the period, which is what the Times is claiming. I would give high odds that it's the Times, not Reuters, that is wildly misstating the facts.

[Later Addition]

One of the comments to this post gives the URL of the GAO study in question. It looks as though both Reuters and the Times misread the report, but in different ways. The Reuters story is clearly based on:

"As figure 2 shows, about 72 percent of FCDCs and 55 percent of USCCs reported no tax liability for at least 1 year during the 8 years. About 57 percent of FCDCs and 42 percent of USCCs reported no tax liability in multiple years—2 or more years—and about 34 percent of FCDCs and 24 percent of USCCs reported no tax liability for at least half the study period—4 or more years."

While the text doesn't say so, the figure's label makes it clear that it is showing large corporations not, as Reuters implied, all corporations. I made the same mistake in an earlier version of this addition, and was corrected by a poster in the comments thread.

I could find nothing in the report that corresponded precisely to the claim in the Times story. One possibility is that they were using the same figures as Reuters, combining the numbers for the two kinds of corporations, and misreading one year in eight as eight years in eight. Alternatively, they might have been misreading the first graph in the report which showed, not the percentage of firms that reported no tax liability in any of the years 1998-2005 but, for each of those years, the percentage that reported no tax liability in that year.

Perhaps the author intended to write, not "Two out of every three United States corporations paid no federal income taxes from 1998 through 2005" but "In each of the years 1998 through 2005, two out of every three United States corporations paid no federal income taxes," which is at least roughly true, sounds very similar, but is in fact very different.

Tuesday, August 19, 2008

Is College Segregation Declining? A Hopeful Sign

In recent decades, many colleges have instituted their own form of residential segregation, creating dorms with titles such as "Third World House" or "African Heritage House" on the theory that they provide a friendly environment for members of minority groups and those sympathetic to them. On the whole, it strikes me as an unfortunate development, promoting the idea that racial categories are of central importance, that an Afro-American student of physics must have more in common with an African-American ed student than with a fellow physics major whose ancestors came from China or India.

I recently came across a small scrap of evidence suggesting that this pattern may be in decline, not with those running dorms but with those living in them. A student about to start her freshman year at an elite liberal arts college was sent the name of the dorm she had been assigned to; a little online research found it listed as the college's African heritage house. Since the student was neither Afro-American nor especially interested in Afro-American culture, she called the college authorities to suggest that there had been some mistake. They explained to her that one wing of the dorm in question was being used not as part of the Afro-American theme house but as overflow housing .

The capacity of that dorm is about half the number of Afro-American students enrolled at the college. It follows that fewer than half of those students want to live in a theme house dedicated to their racial background--considerably fewer if we assume that some residents are not Afro-American and that a wing represents a substantial fraction of the dorm. And the fact that webbed descriptions of the dorm do not mention that it is only in part a theme house suggests that the number who want to live in an African heritage house may be declining.

Which, if true, is good news.

Sunday, August 17, 2008

Russo-Finnish Trade: A Research Proposal

I have been having an interesting Usenet conversation on the subject of economic relations between Finland and Russia in the postwar period. The facts almost everyone agrees on are:

Finland started by paying Russia a large sum in "reparations."

Thereafter trade was conducted as barter, with an accounting system to keep track of which country owed how much to which. Russians provided Finns with raw materials, especially oil, at a price said to be somewhat below the current world price. Finns provided Russians with a variety of manufactured items--it isn't clear how the prices were determined. No money was actually supposed to change hands. At the end of the period, when the Soviet Union collapsed, it turned out that the Soviets owed quite a lot to the Finns; Russia has now paid off the debt.

The obvious conjecture, given the reparations, the relative strength of the two countries, and their previous history, is that the trade was a form of disguised tribute, that the Finns were paying off the Russians not to attack them. Several people in the conversation, however, who claimed to be familiar with the facts--Finns and an American long resident in Finland--claimed that it was the other way around. The Russians provided the Finns oil at below market prices and accepted low quality manufactured goods from the Finns at prices higher than they could have sold them for elsewhere.

In the course of the exchange, three different explanations were offered for this pattern of trade:

1. The Russians were bribing the Finns to provide them political support. Finland was the one democratic and capitalist country that was in some sense an ally of the Soviet Union; the relation could be used by the Soviets as evidence of their interest in peaceful coexistence.

2. The Russians were constrained by their own ideology in ways that either made them relatively indifferent to the real terms of the exchange or made it hard for them to trade with other capitalist countries and so forced them to accept the terms they could get from the Finns.

3. The Finns were smarter than the Russians, or at least understood trade better, and so tricked the Russians into trading on terms favorable to the Finns.

Of these explanations, the first strikes me as plausible, the others considerably less so.

The project I am suggesting is to figure out, on net, whether the exchange was profitable for Finland at the expense of Russia, profitable for Russia at the expense of Finland, profitable for both, or a loss for both. The Finns seem sure that it was in the economic interest of Finland, perhaps in the political interest of Russia, but they did not offer a lot of evidence. The trade and the terms were arranged government to government, providing lots of opportunities to fudge the figures. In particular, it was possible for the Soviets to accumulate a substantial interest free debt, as they eventually did. It is claimed that that happened only late in the relationship. It is also claimed that the people concerned were surprised when they discovered how much the Soviets, on net, owed--which suggests that nobody had been keeping very careful track of it.

The hardest part of project, supposing one could get data on what each side delivered to the other when, would be pricing the goods--figuring out what they would have sold for elsewhere.

[Anyone interested in the Usenet exchange can find most of it at with this search.]

Two Paths for the Libertarian Party

“Libertarian,” in the modern American context, describes a range of political views. It includes hard core libertarians who would like to reduce the size and scope of government to something between a minimal state and no state at all. It also includes the much larger number of Americans–polls suggest ten to fifteen percent of the electorate–who favor substantial, but not radical, reductions in government involvement in both economic and social matters. A libertarian in the second sense is probably in favor of legalizing medical marijuana, might well be in favor of decriminalizing marijuana use, is probably not in favor of complete drug legalization. He is likely to view education vouchers favorably, is probably opposed to national health insurance and may support some privatization of social security, but is unlikely to take seriously proposals for a fully private system of schools, health care, or old age insurance.

The Libertarian Party is mostly made up of hard core libertarians. Its campaign strategy has been to spread their views, to appeal to those who share them, while hoping for some votes from soft core libertarians who see voting for it as a way of pressuring the major parties to shift in a libertarian direction.

This year is going to be different. Bob Barr, the LP’s presidential candidate, is an ex-Republican congressman whose past (and, arguably, present) views are no more than moderately libertarian. He supports the legalization of medical marijuana but allows interviewers to assume without contradiction that he opposes more general marijuana legalization. He limits his opposition to the War on Drugs to opposing federal involvement–which, given that the War on Drugs is mostly conducted on the state and local level, is very nearly an irrelevancy masquerading as policy. Judging by what I have seen of him, his version of libertarianism will correspond, at best, to the views of the libertarian wing of the Republican party.

An optimistic view of this year’s strategy is that it will finally get the LP a significant number of votes. Currently Barr is polling at about 6%, roughly twelve times the best electoral outcome the party has ever achieved in a presidential election. Some of those are doubtless Republicans, both libertarians and fiscal conservatives, who feel this year that their party has abandoned them. But many others may be Republicans, Democrats and Independents who are libertarian in the moderate sense and have in the past been repelled by the LP’s more extreme version. If Barr actually ends up with five to ten percent of the votes that will put pressure on both major parties to modify their positions at least a little in a libertarian direction, which is surely a good thing.

A pessimistic view of the current strategy is, first, that it won’t work–that the vote totals will not match the polling totals–and second that, to the extent it does work, it will send the wrong message. The positions that Barr is likely to push, after all, are already held by quite a lot of people in the major parties. A few years ago, I suggested to a local Democratic congresswoman that one way in which her party could pull libertarian voters out of the Republican party was by coming out in favor of a federal policy of respecting state medical marijuana laws. She replied that she and some colleagues had already introduced such a bill.

Seen from this standpoint, Barr’s candidacy seems chiefly designed to push the Republican party back towards something more like its pre-Bush position as a coalition containing, but not dominated by, libertarian elements. It may also strengthen the position of the more libertarian elements in the Democratic party. But it is unlikely either to spread libertarian ideas in any strong sense or to encourage the major parties to consider more than minor adjustments in their policy positions.

The poster child for the more radical strategy is the Socialist Party of the first half of the 20th century. It won almost no elections, in part because it supported extreme positions. But most of those positions were eventually adopted, and implemented, by the major parties.

I have no confidence in my abilities as a political prophet, so am reluctant to predict which strategy will work better. As an individual voter, however, knowing that a single vote has no appreciable effect on the outcome of a national election, I regard voting as a symbolic act. Being myself a hard core libertarian, I am disinclined to vote for a Libertarian candidate who is, so far as I can see, no more libertarian than the better Republicans.

It looks as though I get to sit out this election.

Wanted: Better Egosurfing Software

One attractive feature of the internet is the ability to spot conversations I would like to be part of, most obviously posts that mention me, and join them. I can and do use Google for the purpose, doing a search for my name plus one of a group of words likely to show up in references to me, minus a string that signals a post on one of the Usenet groups that I'm already reading, and limiting the search to the period of time since I last did one. It's entertaining, useful--on one occasion I found someone attributing a page or two of text to me that was actually written by someone else--and provides at least a modest boost to my ego, so long as what I find isn't too hostile. And I occasionally use the same approach to try to find new information on some topic of interest to me.

While the available tools are vastly superior to anything I have use of in realspace, they are still pretty clumsy. Google checks to see how recently a page has been updated but doesn't tell me whether the update introduced the particular terms I'm searching for, so many of the pages I find are ones written long ago and seen by me long ago. And my search doesn't distinguish between a page that says something about me and a page that merely contains a link to my blog, a link that is likely to have been there for a long time.

Is there a better way of egosurfing?

One obvious approach is to figure out a better search string; so far I have not managed to do so. Another, if (as I suspect) there are a fair number of people online who engage in similar searches, would be dedicated software for the purpose. The main thing I would want it to do is to filter out any page where the reference to the term being searched for was present during the last such search, while leaving in a page that was spotted before that now contains a new reference.