New Year's Scorecard 2005

I now undertake the painful exercise of revisiting this year's resolutions.

Will complete bulk of thesis research so I can graduate in 2005-2006 academic year.

Verdict: Done. There's some significant (though mostly uninteresting) technical work that remains, but I plan to defend in summer 2006. I'd probably be done with the technical work, and in the midst of writing the actual dissertation document already, if it weren't for the fact that I've been simultaneously looking for a job. Blech.

Will donate larger fraction of income to charitable organizations.

Verdict: Failed. I did not donate a significantly larger fraction of my income to charity this year, and I'm not in a position to remedy this in the next two days.

Will limit blog-reading to two times a day, or fewer.

Verdict: Mixed. Most days I keep to this. Having my XML feed reader hacked into a more usable form helped. Some days I don't read blogs at all, because I'm confident that I can catch up on anything interesting the next day via my feed reader. However, other days I fall off the wagon, particularly if I'm trying to avoid something.

Will strive for more generosity of spirit and less petulance in my blogging.

Verdict: Failed. My most popular post this year, by far, was this rather obnoxious post on Intelligent Design (which, incidentally, is now the #5 hit on Google for "intelligent design debate" [UPDATE 6 Jan: And also one of BlogPulse's top 25 blog posts of 2005. Wow/argh.]). The runner-up for most popular post was this evisceration of Jonah Goldberg. And, although most of my posts are less vitriolic than these, I feel that my tone has been splenetic more generally as well.

Will publish one, preferably two conference papers.

Verdict: Failed. I did, however, get two conference papers rejected, which is something of an achievement I suppose. On the brighter side, my reviews have been decent, and I'm presenting at a workshop in the coming year.

Will learn how to deal with conference deadlines without skipping the gym, spurning my friends, and generally dropping off the face of the Earth.

Verdict: Mixed. The degree to which I drop off the face of the Earth has diminished somewhat, but I do live a pretty lousy life in the month leading up to a deadline. On the other hand, my papers get rejected, so maybe I should learn to drop off the face of the Earth more. Paul Graham has interesting things to say about this.

Will buy and read:

• Spikes: Exploring the Neural Code, by Fred Rieke, David Warland, Rob de Ruyter van Steveninck, and William Bialek.
• The Pentagon's New Map, by Thomas P. M. Barnett.
• Rational Herds: Economic Models of Social Learning, by Christophe P. Chamley.
• Collapse: How Societies Choose to Fail or Succeed, by Jared Diamond.
• All the books I bought over the holidays.

Verdict: Mixed. Of the above, I only read Collapse, and about half the books I bought over the holidays. I did read some other interesting books (though I didn't finish as many as I'd have liked), of which I'll recommend a couple:

• Iron Council by China Miéville. The folks at Crooked Timber have said volumes about this, and I don't have anything new to add for now, but it's a lovely book. I also read Perdido Street Station on a plane; it's less good, but managed to keep me up on the red-eye from Seattle to Newark. I'm going to be flying around a lot in the next couple of months, and I'll most likely be reading more of Miéville --- at least The Scar, to complete the New Crobuzon novels.
• Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States, by Albert O. Hirschman: I'm still in the middle of this, but if you're even vaguely interested in politics, economics, or game theory, then you really owe it to yourself to read this sometime. It's a slim little volume, easy to carry around, and easy to read, though digesting the ideas fully will probably take me a lot of mulling (which, of course, I don't really have time to do these days, but whatever, I'll probably do it anyway).

Well, all in all, it was a pretty mixed year. I suppose I'll be making new resolutions (or renewing the old ones) soon. Have a Happy New Year all, unless I'll be seeing you in the next couple of days, in which case I'll wish you Happy New Year in person I suppose.

Manual blogspamming via Technorati: Grrrr.

I'm writing today to note a mildly noxious form of spam that's cropped up a couple of times on this blog, and that popped up again tonight. My last post about the Massachusetts Little Red Book/DHS hoax got a comment that was only marginally on-topic, and frankly struck me as a total non sequitur. It was evident that the post's satirical content, for example, completely sailed over the commenter's head. Or, more likely, the commenter simply hadn't read my post at all, since the comment consisted solely of two links to his own blog posts about the Little Red Book hoax. Hence, the comment seemed like some kind of self-promoting spam. So I deleted the comment, of course, but I also got annoyed enough to fire up Sitemeter and go a-hunting. The comment was posted at about 8:40 p.m. PST, and bingo, here we go:

The IP address is located in Wisconsin, and the blog in question states prominently that the author's located at Marquette University, so this is probably our man. Note that this guy spent a grand total of 15 seconds reading my post before composing his perfunctory, self-promoting comment. Gee, either he's a speed reader, or... hey, let's look at that referer --- it's cut off in the picture, but he came from the Technorati search for "student mao massachusetts". I wonder if this person's spammed his comment onto other blogs on this search?

Hey, that comment looks kind of familiar... (I'd go on, but I think I've proven my point.)

Therefore, "John McAdams", I christen thee spammer, sad sack of dittohead wingnuttery that thou art, so desperate for traffic that thou goest commenting onto random Technorati blogs for attention.

And let that be a warning to other would-be Technorati spammers out there: this blog's not a platform for you to misdirect traffic towards your own site. If you're not here to have a conversation, then crawl off and spread your ant droppings somewhere else. Or better yet, use comments as they were meant to be used, and don't try to promote yourself that way at all.

On newspapers and trust

On December 17, a regional Massachusetts paper called The Standard-Times reported that agents from the Department of Homeland Security visited a student's house after he checked a copy of Mao Tse-Tung's Little Red Book out of a library. Today, we learn in the Boston Globe that this story was actually false. Now, clearly, this is an example of the incredible dangers inherent in the technology of newspapers, which permit reporters to publish things that are false! It's true that the error was eventually corrected, but the story was out there for a whole week, and it's far from clear that people who read the original article will see the follow-up reporting.

And look --- the Standard-Times posted a corrected article, but the old article's still available in the newspaper archives, where any schoolchild could come across it! Why don't these irresponsible editors send the archived article down the memory hole, for the children?

Emboldened by the unencumbered freedom of the printing press, reporters are out there spreading scandalous untruths about our national security apparatus. It's clear that the system of reporters and editors digging up facts and reporting them, without the oversight of a national Ministry of Information, cannot be allowed to continue. Or rather, I am personally all in favor of newspapers, but unless newspapers voluntarily submit themselves to scrutiny by a private-sector Ministry of Information, which consults with the government to ensure accuracy and fairness, then in the next election, who knows, maybe some reporter will report something inaccurate about a politician, and the political pressure to regulate newspapers will become overwhelming. And people like John Seigenthaler will have to step in to defend them.

Respectable arguments against same-sex marriage

Random late-night browsing whilst procrastinating w.r.t. proofreading a camera-ready copy of a paper: as a thought experiment Belle Waring constructs an intellectually non-ridiculous argument against same-sex marriage; and (linked from the above comment thread) a plea for intellectual humility from Jane Galt, of all people, that's also not entirely ridiculous.

I have two reactions. First, any intellectually honest person who opposes same-sex marriage for the linked reasons should also support denial of marriage benefits to childless heterosexual couples. But opponents of same-sex marriage won't come out against childless heterosexual couples. Why not? Because society views childless heterosexuals (unlike homosexuals) as first-class citizens, and therefore, it would be incredibly unpopular to infringe on the autonomy of heterosexuals. People believe that heterosexual couples have a right to choose both marriage and childlessness for themselves. Besides, some childless heterosexual couples are childless for painful personal reasons; adding insult to injury by denying them marriage benefits seems beyond the pale. I could say closely analogous things for committed homosexual couples --- that they have the right to choose marriage, and that denying them benefits is cruel --- but the anti-same-sex-marriage forces wouldn't care. Therefore, let's admit the role that power, stemming from bigotry, plays in the acceptance of even the relatively respectable arguments proposed by Galt and Waring: although we would not treat heterosexuals this way, we will treat homosexuals this way, because it is popular to do so, and therefore we have the power to do so.

Second, on a somewhat different track, I want to make a larger point, which is that sometimes the well-being of society simply must defer to individual rights. Individuals are ends-in-themselves, whereas society is a means to furthering the ends of individuals. If guaranteeing individual rights commensurate with our values ultimately leads to the breakdown of society, then so be it.

It is conceivable that a society built upon individual autonomy, civil liberties, and equality before the law cannot be sustained indefinitely. It is conceivable that on the grand scale of history, each era when such institutions prevail is a brief interlude, like foam on the crest of a wave, which inherently rises and then subsides, whereas discrimination and oppression are a sturdy bedrock which endures. It is conceivable that the American experiment in constructing ever-widening circles of social equality contains the structural causes of its own collapse, and that it will therefore recede and be replaced by something more morally repugnant and more sustainable.

So what? Happiness is always finite, but at least it is happiness. Most love affairs contain within them the seeds of their own destruction, but this does not compel us not to love. If equality before the law cannot endure forever, then at least we're lucky enough to have lived in a time when it prevailed. Maybe in some distant future, after same-sex marriage and America's million other concessions to the pursuit of individual happiness have destroyed America, and some dark millennia have passed without us, some other people will, for a fleeting few centuries, be lucky enough to live in such an era again. If so, then I hope that they, too, refuse to destroy their happiness in order to save it.

The death penalty vs. war

A propos of nothing, I was thinking today about a comparison I've frequently seen made between the death penalty and war --- viz., that if you believe killing in war is sometimes justified, then you cannot be philosophically consistent and also categorically oppose capital punishment. I find this comparison rather fatuous. Under conditions of war, there is no reasonable way to accomplish your goals --- like stopping the Axis powers from taking over the world ---- except by killing large numbers of people. If you try to stop the Axis powers from taking over the world, many of their people will shoot at you, and will not stop shooting at you until they are dead. And the goal, in this case, is unquestionably good: if the Axis powers take over the world, then a lot of people will die or suffer horrific oppression.

Capital punishment is a completely different beast. At the point where capital punishment comes into play, the person to be executed is sitting inside a steel-barred concrete cell. However horrific we find this person's psyche, (s)he is essentially a helpless worm in the unyielding grasp of the state's fist. Now, consider what further objectives society has, and whether these objectives can be accomplished by nonlethal means. I'd claim the only unquestionably good aim here is to defend society against future violence, which can be accomplished by keeping that person in the cell forever. There's an array of other, far more questionable objectives, like vengeance or emotional satisfaction, that may require you to execute this person, but it's hardly inconsistent to believe that such objectives do not justify killing.

Or, to put it another way, justifiable killing requires that the aim be just, and that there be no other way to accomplish that aim. One can therefore consistently believe that for a modern nation-state, killing in war is justified, but capital punishment is not.

Incidentally, note that the above implies that if a nation can attain its war aims without killing people, it is obligated to do so. It also implies that if the governing power is too insecure and weak to guarantee incarceration for a complete human lifetime, then it may be justified in killing instead. The latter's interesting for two reasons. First, it probably describes the conditions under which Old Testament moral codes were crafted. Second, if you believe, as I do, that we may enter a future phase of history when human lifespans exceed those of nation-states, then capital punishment may be justified again.

On Wikipedia and trust

So, the Seigenthaler affair is getting a lot of coverage, and now that my deadline's past I feel that I can weigh in. Katharine Seelye, in a recent Week in Review, asked:

According to Wikipedia, the online encyclopedia, John Seigenthaler Sr. is 78 years old and the former editor of The Tennessean in Nashville. But is that information, or anything else in Mr. Seigenthaler's biography, true?

Seelye then went on to discuss the inaccuracies in Wikipedia's Seigenthaler bio, of which Seigenthaler complained in a recent USA Today Op-Ed.

Ah, the irony: Katherine Seelye wrings her hands about people making stuff stuff up. The Wikipedia entry can be fixed by anybody. Seigenthaler could have just clicked the "edit" link and corrected the mistake. But the damage Katharine Seelye did in her 2000 election coverage can never be undone. I find Seelye's distorted coverage of the distinction between the Gore and Bush budgets particularly irksome, but perhaps you'd prefer Seelye's deceitful coverage of the Gore/Bradley primary debates.

So do I trust Wikipedia? About as far as I trust a newspaper, or Brittanica, or a recently published peer-reviewed scientific result; which is to say, not really. All knowledge I obtain through all these sources is merely provisional and subject to change. But absolute trust isn't a useful benchmark of value here. If I held all my information sources to the standard of absolute trust, then the only things I'd believe would be machine-checkable proofs of mathematical theorems (which, by the way, excludes nearly all nontrivial published theorems in the math and computer science literature). But that would be a ridiculous way to live. The right question here is: are we better off because this resource exists? Or, in other words, is the net value positive? And for Wikipedia, the answer's undoubtedly yes, at least for me.

Therefore, my answer to Wikipedia critics is: if Wikipedia's net value to you is negative, then don't use it. On the other hand, if you want me to stop using it, you must convince me that I'm mistaken, and that I'm worse off now than I was before Wikipedia existed. That's a pretty tough sell, because I'll require empirical evidence --- not that Wikipedia has errors, which it obviously does, but that Wikipedia's meaningfully worse than the alternatives. I don't believe that's the case. In my area of expertise (programming languages), Wikipedia's relatively reasonable. It's at least as good as any coverage in the popular press, or even most of the computing industry press. Of course, Wikipedia isn't TAPL, but then nothing is.

Finally, let's remember that Seigenthaler hardly needs our pity or assistance here. In his Op-Ed, he painted himself as a helpless victim, but he's a heavy player in media circles. He's heavy enough to get an Op-Ed published in USA Today simply because he found some random nonsense on the Internet. Think about that: if I got an Op-Ed published every time I found nonsense on the Internet, I'd have more bylines than Krugman, Dowd, Brooks, and Friedman put together. He's heavy enough that the Wikipedia prankster was forced to resign from his (non-journalism) job, presumably because the prankee (a co-worker of the prankster) was a friend of Seigenthaler family and wanted to remain in their good graces. Ask yourself this: has anybody ever forced someone out of a job simply to remain in your good graces? And he's a heavy enough player that the noises he's making about Wikipedia inviting regulation, e.g. in this CNN interview, might be taken seriously. Seigenthaler says:

Next year we go into an election year. Every politician is going to find himself or herself subjected to the same sort of outrageous commentary that hit me, and hits others.

I'm afraid we're going to get regulated media as a result of that. And I -- I tell you, I think if you can't fix it, both fix the history as well as the biography pages, I think it's going to be in real trouble, and we're going to have to be fighting to keep the government from regulating you.

Elsewhere in the interview, he disavows wanting the government to regulate Wikipedia. However, Seigenthaler's tone bears no small resemblance to a mobster's when warning a shopkeeper that if he doesn't pay for protection, hey, it's a dangerous neighborhood, and somebody might come along some night and burn something down. The truth is, people like Seigenthaler are a much greater threat to a free society than Wikipedia or its vandals.

Incidentally, the fantasy nightmare scenario that Siegenthaler spun on CNN bears no resemblance to reality. Wikipedia existed during the 2004 election. The Bush and Kerry articles were vigorously vandalized, and vigorously corrected, and today they're probably among the most factually accurate public sources of information about the two former candidates. There was no Wikipedia Chernobyl, no meltdown of defamation and reaction. The process worked. Of course, veteran newsman and paragon of journalistic integrity John Seigenthaler could not be bothered to investigate what happened during the 2004 election before speculating on national television about what could happen during the next election. Which, if you think about it, sort of undermines his point.

Further reading: John Seigenthaler Sr. Wikipedia biography controversy and talk thereof.

Friendster vs. Google

I recently logged onto Friendster, for the first time in months, on a lark, and the next day I got a Friendster message from an old high school acquaintance who's on the service. This puzzled me, because Friendster's messaging interface bites hard, and because you can get my email address within seconds via Google. Surely anybody who wants to get in touch with me, and who knows my name, should be able to email me...

Then I realized that my time in academia's showing. I have a personal home page, as does everybody else in (North American) computer science academia. When you want to look someone up, the first thing you do is Google their name. But the vast majority of people still, a decade after the web revolution, do not have home pages. (Maybe this is a good thing for those of us who do, since it makes it easier to find us.) Most people don't live in a well-indexed world. They live in a world where the people from their past simply disappear, and it takes something like Friendster, or Facebook, or whatever, in order to reconnect them.

It seems obvious that this is both bad and fixable...

24 hours to certify Diebold source code?

Via IP today, I learn that the North Carolina State Board of Elections claims to have reviewed and certified the reliability and security of Diebold election machines' source code in less than 24 hours after it was placed in escrow.

All I can say is, North Carolina's State Board of Elections must have software quality assurance engineers the likes of which the world has never seen, engineers who can conjure comprehensive test suites out of thin air faster than mere mortals can even type them up, and who can construct proofs of program correctness in Hoare logic on-the-fly while reading source code, and who don't need symbolic model checkers because they can simulate thousands of Büchi automata in parallel in their heads while filing their fingernails.

Or, hey, here's a thought: maybe the board conducted some kind of sham certification process and rubber-stamped the software without really examining it. Nah...

My previous posts on election machines: Voting Machine Abuse Link-O-Rama; Close your eyes and hum pretty songs and maybe the code will fix itself; Further links on voting machine manufacturers' arrogance; Krugman on voting shenanigans [etc.]; Electronic election idiocy from Tom Zeller.