/usr/share/morlock

Blog moved

2010-11-15T13:48:00.002-05:00

I'm rather fed up with blogger. Go see http://jaapweel.posterous.com/

Another Culture War? No Thanks | Brink Lindsey | Cato Institute: Commentary

2010-09-06T04:49:00.000-04:00

Brink Lindsey hitting some nails on the head: "Another Culture War? No Thanks"

High Resolution Fundraising

2010-09-04T20:02:00.000-04:00

Paul Graham in his latest piece on High Resolution Fundraising: "Fixed-size, multi-investor angel rounds are such a bad idea for startups that one wonders why things were ever done that way." And yet they were. For a while.

And that's how economics works in the real world.

It's not the investors colluding by Marxian magic to do what's good for the entire Capitalist Class, as Graham briefly suggests but then rejects.

But neither does a market find its equilibrium immediately in one great flash of implicit Walrasian auctioneering.

Rather, the market's efficiency is limited by the rules it follows, which are fluid and developed by trial and error. Startup investing is fraught with uncertainty and asymmetric information. There's more than one way to skin those cats, so it's no wonder that the system keeps evolving.

In old Mexico

2010-07-22T14:44:00.003-04:00

"In Old Mexico," one of Tom Lehrer's lesser-known songs, is surely among the most fun to perform. Like "Lobachevsky," but unlike many of the others, it is fully durchkomponiert: there are no repeating stanzas, but rather a rhapsody of different tunes that support the humor of the lyrics, interrupted twice by parlando sections where Lehrer speaks while continuing to play the piano (something that, I must admit, I have never been able to pull off).

But what I did not know about the song is that all the fancy-shmancy Spanish words are actually technical bullfighting terms. The matador comes first, then the picadors and the banderilleros, and finally the matador returns to kill the bull with a sword. Even the bit where "my brother's dog Rover" gets killed "with such grace and artistry that the witnesses awarded the driver both ears and the tail" is a reference to the traditional practice of awarding one ear of the bull to a good matador, two ears to a very good one, and indeed both ears and the tail to one who is truly outstanding. Read all about it on Wikipedia...

A hack

2010-05-23T08:50:00.003-04:00

We had an interesting problem in the lab yesterday. The Snowglobe 2.0 Second Life Viewer allows "Shared Media," which comes down to web pages and the like being displayed on objects in Second Life.

It appears that whenever the user walks their avatar up to an object (called a "prim") that has a web page displayed on it, a process is forked called SLPlugin. As the avatar gets moved away from the object, at some point, the process is terminated. Usually. Most of the time.

But occasionally, some of these processes survive. They seem to be very busy reading from a TCP connection to localhost that has been closed on the other end and gives an EAGAIN error over and over again. Anyway, whatever they do, they take up memory.

At some point, these SLPlugin processes fill up RAM. On our machines, it takes about 100 of them. The Linux OOM killer then kicks in and kills the main Second Life process. Kaboom.

The solution comes from an option to pkill: -o. It lets you kill the oldest of the processes matching some description. If you want to kill all but the 50 most recent SLPlugin processes, using zsh:

while [[ `pgrep SLPlugin | wc -l` > 50 ]]; do pkill -o SLPlugin; done

Wrap that in another loop to execute it every 10 seconds. Voilà. It seems that in our application, there are never more than 50 non-stale plugin processes, and it all works.

Google maps around DC

2010-05-15T21:18:00.005-04:00

Roads around DC are a godawful mess and Google can't really be blamed for not getting everything right the first time, but it would be nice if they could fix some bugs.

First of all, in Arlington, there are streets named "n STREET N" and "n STREET S" that are nowhere near each other. This is similar to the DC system, where each street exists up to 4 times: "x STREET NE", "x STREET NW", "x STREET SE" and "x STREET SW". But, unlike in DC, there is a local tendency to write "3333 N 3RD ST" instead of "3333 3rd ST N". Google Maps standardizes these addresses correctly... some of the time. In particular, try to search Google Maps for "1521 n 12th st arlington va" and you'll get it standardized to "1521 12 st s"! This messes up things like padmapper.com, which take address data from places like craigslist where it is often listed in the informal form with the direction before the street name.

Then try this query: "22030 to dc". Why does it not take the interstate? Because some of it is HOV-only some times of day.

View Larger Map

Next up: some metro stations are presumed to be on freeways. Try "caron dr falls church va to west falls church station".

View Larger Map

In defense of André Rieu

2010-05-14T03:47:00.005-04:00

I have seen some Australian reviews of André Rieu that interpret him as the conductor of an elite symphony orchestra and chide him for not being on par with the Concertgebouworkest or the LA Philharmonic. I can only imagine what sort of confusion of expectations must be at the root of some highly strung upscale-newspaper music critic walking into an André Rieu concert and trying to form an opinion about the authenticity of the interpretation of the tempo in the third movement or some such. Not that Rieu is a bad musician—far from it. For those who are wondering who the hell I'm talking about, a short video. If this doesn't do anything special for you, pay special attention to the blissful facial expressions in the audience. To say they love it would be a gross understatement.

Rieu will do anything to please a crowd. In that, he's not alone. But what he has going for him is the ability to actually find out what will please the crowd, and the skill to put it on stage, whatever it is. A typical Rieu concert can involve music written by anyone from Schubert to Michael Jackson, and artists ranging from local Maastricht amateurs to famous professional opera singers and, well, Heino (a picture says more than 1000 words, see photo.) But everything he plays has one thing in common: every number is essentially an encore. In Maastricht, he plays Wie sjoen ós Lèmburg is en Waar in't Bronsgroen Eikenhout—in Australia, it's Amazing Grace and Waltzing Mathilda. Do the locals know how to waltz? He'll play another waltz. Do they like the sirtaki? Hava Nagila? Rieu will comply, with all the flowing dresses, shiny tuxes, bouncing fiddlesticks, booming bagpipes, and flashy fireworks he can muster.

What is striking about Rieu is his odd mix of cultural rootedness and versatility. When Rieu says he loves Maastricht, he means it. He is perfectly sincere when he writes about his home town in sentimental prose filled with exclamation points ("I’m sure there’s no other city where so many people would have sung along, so musically, with every piece. It’s no wonder my heart lies here!"). Limburg oozes from every note Rieu plays. And yet he is a cultural chameleon, jumping from Vienna to New Orleans to Moscow and back within 5 minutes, performing local favorites in ways the locals like better than their local performers.

This is not just a personality trait. To understand Rieu, you have to understand Limburg. Here goes. Dutch Limburg, much as the Limburgers may romanticize the landscape, is a fairly run-off-the-mill piece of Europe. It hasn't been of any great importance since Charlemagne and it is mostly composed of mid-20th-century mass-produced mining towns amid rolling hills. Maastricht, admittedly, is pretty, and so are a few of the smaller towns in Belgian Limburg: Tongeren, say, or Hasselt. But what's special about Limburg is the culture.

It's culture full of paradox, deep and distinct and intensely local, yet mongrel and cosmopolitan to the bone. Dutch Limburgers who could scarcely conceive of living among Hollenjers think little of moving across the national border that divides the two Limburgs. (The reverse is less common because of relative real estate prices.) They speak a collection of dialects that are incomprehensible to the surrounding speakers of standard Dutch, High German, and French and pride themselves on local traditions like vastelaoves (mardi gras) with its endless preparations and extensions, and local organizations like the herremenie (brass bands; Rieu once used 500 amateur brass players in a concert) and sjötterie, which involves dressing up in traditional uniforms that don't follow any particular tradition and using hand-cast bullets in enormous but clumsy guns to shoot small bits off clay off a structure that looks like an oversize telephone pole.

On top of all this, most of Limburg practices a form of Roman Catholicism long on ritual and family gatherings and short on theology and an attitude to life described in contrast to that of their Protestant Dutch brethren as "bourgondisch"—that is, alien to the Calvinist concept of food, drink, merriment, and relaxation being inherently sinful.

In short, there is a lot of random assorted folk culture, and relatively tightly knit family bonds. What else is new?

It is that this particular tightly knit, quaint, cutesy local culture absorbs foreign elements haphazardly and enthusiastically. When I say "absorb," I mean it. Rieu's shtick is Viennese waltz, yet he has managed in short order to turn his annual open-air performances into the kind of "typical" thing that you reminisce about in exile—just like the steady stream of Czech folk dancers making the rounds of small towns. Local popular music includes styles derived from Mexican (norteño), Austrian (Tiroler), and American (tex-mex) folk music, as I noticed to my surprise when I moved to LA and found the local Mexican radio stations to broadcast the sounds of home. If you don't believe it, here's same Austrian Limburgers:

I am going to go out on a limb now: I think that Limburg culture is an instance of a more general phenomenon that shows up in border regions. Limburgers have been governed over the years by many different jurisdictions, and to some extent they have become immune to attempts to assimilate them to one national culture or another. (This to the great annoyance of the government in Den Haag, which would at least like to see the slightly corrupt and strongly family-based political culture replaced with the less corrupt and strongly party-based political culture that prevails elsewhere in the country.) As such, the strong division between indigenous and exogenous that exists in the seats of nations, where the locals speak the same language as government and identify strongly with the nation state, does not emerge.

(This theory was suggested to me years ago by my uncle.)

Voice in secondlife

2010-03-15T02:19:00.003-04:00

I use secondlife a lot for my work. But in a strange way: I mostly use it as a programming environment. This is because we use secondlife in our economics experiments as a ready-made 3D interface, and I write the scripts to make it do what we want, and the only way to edit those scripts is to use the secondlife client.

Anyhow, Linden Labs recently released the 2.0 version of their viewer, and they made a big deal out of the new version voice chat, which supposedly just works and requires no setup. I have seen this claim made about many voice chat, PC/phone, etc., programs, and it's never true. For Skype it might be true on Windows, but on Linux it sure still is a big mess.

This is why I am delighted to report that with Secondlife 2.0 Beta on Ubuntu GNU/Linux 9.10, running under xmonad and with pulseaudio for an audio server...

...I had no problems whatsoever. Clicked on the "speak button," said "can you guys hear me?" and the answer came back "yes."

Sometimes software does work ;-)

(BTW, apparently the way Linden Labs got this to work was by licensing something from a company called Vivox, which makes a library for voice chat in games.)

Poor man's single-sign-on

2010-02-22T21:59:00.006-05:00

How to allow the users of your research application to log in with their existing email passwords:

smtplib.SMTP('the.college.mail.server.edu').login('user', 'password')

If they entered the right password, you get some tuple, and otherwise, an exception.

City planning

2010-02-08T13:52:00.001-05:00

Literary genius hiding in VA EZ-pay office

ffax city 2/8

2010-02-08T13:46:00.001-05:00

compare and contrast

ffax county 2/8

compare and contrast

Best bumper sticker ever

Rubber feet

2010-01-20T19:22:00.000-05:00

I will try to get around to blogging more high-brow things soon, but first a pet peeve: the little rubber feet on electronic devices that fall off. Case in point: a DYMO® LetraTag® label printer sitting right in front of me on my desk, barely 24 days out of the box (or rather, the blister-pack.) Came with two little rubber feet glued to the bottom of the battery compartment cover. One is gone. These little rubber feet are well-intentioned. It is hard to manufacture cases that will sit stably and without scratching on a variety of surfaces, and adding a few bumps made of a slightly springy artificial rubber material fixes the problem. In fact, I have sometimes had to add them back to devices that had lost them, such as one portable hard drive that wouldn't stop making annoying noises, which got a substitute made of a little lump of rubber eraser held on by a patch of duct tape. But why do they always have to fall off in the first place? It is perfectly well possible to make these things properly. Because the material is flexible, you can mold them into little rivet-shaped things that snap into the casing. Or you can use the kind that are held on by a screw. Or whatever. Now you might say that I silly to complain about the quality of cheap things, which are cheap precisely because low quality is cheaper to manufacture, and I should just shell out for quality or alternatively shut up. But that won't fly: the problem exists even with expensive and, otherwise, high-quality devices. The hard drive I mentioned had its case designed by none other than F.A. Porsche, which must have cost a pretty penny.

End of rant.

Consumer credit officially no longer tight

2009-11-17T05:10:00.001-05:00

Consumer credit is now officially no longer tight, if it ever was. This I have proven very scientifically in the following way: a company I have a credit card with sent me an offer for a $30,000 car loan. I am a grad student. They are trusting me to buy a new BMW on credit. QED.

American food no longer so shabby

2009-10-28T09:52:00.001-04:00

Once upon a time, Europeans—or at least those in the traditionally Catholic countries—found it easy to dismiss any and all American accomplishment: what good is an air-conditioned "home" with two land yachts in the driveway if you can't get a decent cup of coffee, a palatable glass of wine, or a loaf of Real Bread? But the time for culinary smugness has long since passed. Campbell's soup cans still come with recipes for "casseroles" produced by combining the contents of three cans of Campbell's soup with some breadcrumbs and imitation cheese, but they have been supplemented by generous helpings of delicacies both foreign and domestic, both in grocery stores (the biggest!) and at restaurants (the cheapest!). All this and more is eloquently documented in "America’s Food Revolution" by Jerry Weinberger, City Journal Summer 2009 (HT: Arts & Letters Daily). The piece is not really meant to be an act of social science, but it does leave me with one question: why didn't all this happen earlier? What is it that caused Americans from the 1950s onward to spend so much of their rapidly increasing incomes on more convenient, rather than tastier, foods, for so long, and what is it that turned fine dining into a national pastime after all? It can't be income alone, can it?

Popper on the treason of the clerisy

2009-10-19T04:53:00.001-04:00

"Now it is interesting to see that some of those who denounce reason, and even blame it for the social evils of our time, do so on the one hand because they realize the fact that historical prophecy goes beyond the power of reason, and on the other hand because they cannot conceive of a social science, or of reason in society, having another function but that of historical prophecy. In other words , they are disappointed historicists; they are men who, in spite of realizing the poverty of historicism, are unaware that they retrain the fundamental historicistic prejudice—the doctrine that th esocial sciences, if they are to be of any use at all, must be prophetic. It is clear that this attitude must lead to a rejection of the applicability of science or of reason to the problems of social life—and ultimately, to a doctrine of power, of domination and submission."

"Why do all these social philosophies support the revolt against civilization? And what is the secret of their popularity? Why do they attract and seduce so many intellectuals? I am inclined to think that the reason is that they give expression to a deep-felt dissatisfaction with a world which does not, and cannot, live up to our moral ideals and to our dreams of perfection. The tendency of historicism (and of related views) to support the revolt against civilization may be due to the fact that historicism itself is, largely, a reaction against the strain of our civilization and its demand for personal responsibility."

(Karl Popper: The Open Society and its Enemies, vol. I, Introduction, p.5)

A spirited defense of SICP

2009-10-11T21:39:00.001-04:00

I wrote this as a comment on Robin Hanson's blog just now.

I think there are things wrong with teaching Java in the first place. (I notice your son writes his games in Python. Maybe he’s onto something.)

One of the most common problems with beginning students of computer programming is that they do not treat the programming language as a system they fully understand (because they don’t), but rather as a system that contains mysteries (rules about implicit typecasting and the order in which to execute the initialization code of various classes etc.) they’ll never understand. This leads to voodoo programming, defined in the Jargon File as “The use by guess or cookbook of an obscure or hairy system, feature, or algorithm that one does not truly understand.” To add insult to injury, the approach to software design that comes with languages like Java is full of things that have the smell of ancient rituals with origins long forgotten, at best—or tax forms to be filled out in triplicate, at worst. The introductory student has no concept of the problems (often deficiencies in the language) that led to the invention of the Factory Visitor Singleton and the like. Hence, once they leave the classroom, they’ll think of such things as bureaucratic fluff and omit it and they are left to rediscover the virtues of modularity all over again.

How, then, do I propose that Computer Science be taught? Structure and Interpretation of Computer Programs, the classic MIT textbook.

Abelson and Sussman make a point of “making the magic go away.” (I was told by the person responsible in past years for the smoke effect that at some point during the semester teaching from their book at MIT, one of them used to show up to class wearing a wizard’s hat, which would then disappear in a puff of smoke.) They use a language so simple that they can explain all of it. You learn not one, but several mental models of what is going on, in increasing levels of detail, culminating in an interpreter of the language written in itself (sounds scary? it fits on the blackboard!) and (admittedly in a usually omitted last chapter) a compiler. Scheme has no structured data types or classes or any of that, but in the course of the book, all of those are constructed as facilities within the language, the student gets to see precisely how they work, and why.

A student that understands SICP understands most of what we can teach about programming. (One exception is the use of type systems.) Of course a CS major may also want to know about Fibonacci heaps and shading algorithms and neural networks and branch prediction and TCP and how to schedule processes in O(1), but none of those are about how to program so much as about what to program, which is a different question.

So why do high schools teach Java? Because colleges teach Java. Why do colleges (though not, by and large, the best CS schools) teach Java? Because someone has gotten it into their head that computer programming is not a new way of thinking, but a thoroughly practical clerical task, a sideshow to the real thing, be it the Analysis of Really Clever Algorithms, or Corporate Data Processing, or what-have you. And that attitude translates into teaching the “industry standard” cruft and the complaint that if you teach students using a language they actually understand, like Scheme or Machine Code (cf. Harvard) or even Python (but that one’s dangerously close to voodoo-land already), then they’ll have to “learn it all over again.” I even know of colleges that teach C++. I have two degrees in Computer Science, I have written several compilers, and programmed in a dozen different languages, and I cannot wrap my head around C++: I do not understand why on earth we expect freshmen to be able to look at a 3-page Standard Template Library error message. If I had to teach CS1 (and not being likely to ever have to do so is frankly the greatest of my regrets in switching from CS to economics) I would want sophomores to tell next year’s freshmen, “computers aren’t mysterious black boxes; once you’ve taken CS1, you’ll understand it’s all really fairly simple; it’s a big heap of NAND gates and all the rest is but conveniences to help us think about them.”

There may be another reason, which is that teaching CS well has two drawbacks: it requires better teachers than teaching it poorly, and it fails to give those students who will just never get it the impression that they’ve actually understood it about as well (or as poorly) as the ones who do. As for the latter problem, I can’t really see a solution. As for the former, I think we need to realize that there simply aren’t enough good CS teachers to go around, or rather, that most schools cannot afford them, given how much they can often earn in industry. That is why it’s important to keep thinking about ways to improve textbooks, languages, and other low-marginal-cost educational tools.

End of rant.

Food irradiation

2009-10-06T04:43:00.001-04:00

I had a post a long long time ago about the names of technologies and how they influence their popularity. Here's an example that I think I missed then: food irradiation. Despite attempts at rebranding it as "cold pasteurization" and the like, it is still a big bugaboo. Which is a pity, because it's a great way to disinfect meat and dairy. I'd gladly pay extra for irradiated eggs. I could use them for Caesar dressing, tiramisu, mayonnaise, and other things with raw eggs in them, without fear of salmonella. (HT: Bureaucracy or Irradiation: Which Should You Trust More?, by Ronald Bailey) As far as I know, though, grocery stores do not sell irradiated eggs, so I have to make do with pasteurized egg whites from a box (a.k.a. "egg beaters.")

Postbestiality

2009-10-05T09:29:00.001-04:00

If you have access to the The Chronicle of Higher Education, this is a good read: Heteronormativity Is Hot Right Now. Best word: "postbestiality."

1066 and all that

2009-10-04T05:11:00.001-04:00

"It was Williamanmary who first discovered the National Debt and had the memorable idea of building the Bank of England to put it in. The National Debt is a very Good Thing and it would be dangerous to pay it off, for fear of Political Economy." (Thanks to Deirdre McCloskey for quoting that in here recent talk at GMU.)

Maximal gerrymandering and proportional representation

2009-09-30T09:43:00.003-04:00

I like to think of proportional representation like this. Suppose that we stopped worrying and learned to love gerrymandering, but we mandated that politicians did it properly and shamelessly. We adopt a rule that says any politician can draw any district, no matter how funny-looking, as long as he wins it unanimously. The Democrats draw funny-looking, disconnected districts around all the Democrats, and the Republicans draw funny-looking, disconnected districts around all the Republicans. Now every Democrat gets to be represented by a Democrat and every Republican by a Republican, and there is proportional representation. So those people who would like to see proportional representation in the US should also love gerrymandering. I know, I know, I made some stark assumptions, some of them hidden, but that's the intuition, anyhow.

Gordon Brown makes apology to Alan Turing's surviving relatives

2009-09-27T16:45:00.004-04:00

Here's the story:

For Alan Turing, a real apology for once

Here's a short documentary about Turing, his achievements, and the circumstances surrounding his death.

Here's a joke:

The next step: repeal Don't Ask Don't Tell.

Rubber rooms in The New Yorker

2009-09-22T23:24:00.001-04:00

Steven Brill's The New Yorker piece about the New York teachers union would just be yet another rubber room exposé if it were published in, say, Reason. But this is The New Yorker, the venerable but ever fashionable left leaning magazine of Culture, Art, and Other Associated Snobbery. So here's to yet another rubber room exposé! It's a good one, by the way, with lots of Real Reporting, and meticulously fair to the claims of the union and its allies, yet nowhere quite managing to make them appear sympathetic.

Sign, signified, and nuclear waste

2009-08-29T10:08:00.002-04:00

There has been much thought given to designing signs that would signify the presence of a nuclear waste dump to a reader with no cultural connection to the dumper.

I propose a very simple solution. The only thing that reliably signifies radiation in a way that is not dependent on any cultural or linguistic context is nuclear radiation itself. Hence, one should not simply have an isolated nuclear dump surrounded by pristine material, such that one that missed or misinterpreted the sign might dig unsuspectingly, and dig some more, and then suddenly hit upon something rather toxic.

Instead, the degree of radioactivity should be ramped down gradually, such that the general area of the dump would be sufficiently radioactive to make an observer with a Geiger counter mildly surprised, and as he gets closer and closer in his digging, increasingly alarmed. Regardless of cultural context, if radiation levels steadily increase as some point is approached, an intelligent observer will at least suspect that they will continue to increase as the point is further approached.

You may counter that the future civilization getting the message may have entirely lost the concept of nuclear radiation, Geiger counters, or anything like that. But in that case, you're stuck anyway. There is no way that a sign that merely conveys "danger," without being more specific, is going to scare people away. Nothing that the ancient Egyptians could have written on their pyramids in any language would have kept people from being intrigued by what could be inside them.

In particular, Edvard Munch's painting The Scream, which I believe has been proposed as a nuclear warning device, probably does universally signify fear, at least to humans that share our innate emotionally determined facial expressions. But that would only make the disposal site more interesting to future archeologists, not less.

Reviews of Anathem

2009-08-29T09:40:00.002-04:00

I am disappointed by mainstream reviews of Anathem. Reviewers appear unable to judge speculative fiction on its own terms. They want books to be about character psychology. To which my response is, let them read Proust. I, for one, find Proust dreadfully boring. When Michael Dirda writes that much of the book sounds like "transcribing intellectual conversations that sound like really nerdy Caltech grad students schmoozing at 3 a.m." and he takes that to be a bad thing, he hits the nail on the head: books like this are not written for people like him, much like A la recherche du temps perdu was not written for people like me.

"Sex is referred to, but never seen," complains Dirda, while lamenting the detailed descriptions of "buildings, machines, and events." To me, when Stephenson does write about sex (the Baroque Cycle is full of it), I usually skip it. Not that he does a bad job of describing sex, but I already know how straight people go about rubbing their private parts together, purely from hearsay, but that's quite enough, thank you very much. Same with fights: Stephenson does a great job of fight scenes, but there's little novelty in the idea of a fight scene.

The descriptions of buildings, on the other hand, are fascinating. They have nothing to do with the usual scene-setting descriptions in an otherwise psychological novel. They are an integral part of the theme of the book.

The theme is long-run institutional design. How do you design an institution such that it would not only be useful if it were to persist, but such that it will in fact persist? It is the problem tackled in The Federalist, and it is the problem faced by those seeking to design nuclear waste disposal facilities or the Clock of the Long Now.

Much of the intricate description of building layouts in Anathem is an attempt to solve such a problem by architectural means. The architecture of Saunt Edhar's largely determines the modes of communication. There are overlapping interests and mutual expectations related to sticky architectural features that are hard to renegotiate because of limitations imposed by those same architectural features.

I suppose Dirda probably finds the Federalist Papers rather dry, too, lacking as they are in sex scenes and character development. But Anathem is a novel of ideas, and any particular reviewer's failure to appreciate those ideas is no reason to condemn it.

Stephenson is an idealist

2009-08-29T09:06:00.002-04:00

From Anathem, it seems clear that Stephenson is an idealist, not in the sense of one that puts abstract ideals before personal comfort, but in the sense of one that believes in the influence of ideas on history.

The problem with idealism, from a scientific point of view, is that it's a messy hypothesis. If ideas and beliefs influence history, that means that history is influenced by one of the hardest systems for historians to understand, that is, the human mind. But the reason why idealism is nonetheless alive and well is that the alternative is worse. Materialism, as it is called, holds that history is a consequence purely of material reality and the ideas that people have about that reality are irrelevant to it. Materialism is neat and tidy and easy to understand, but also ostensibly wrong.

Moreover, few of the people that have expounded materialist ideas have taken them seriously. Marx may have claimed that history is a function of the "material forces of production," but he definitely wrote and acted as though communist agitation might do some good for the world. Similarly, the Public Choice school of political economy, in its most extreme forms, models political decision making as a function of voter self-interest narrowly conceived and the structure of the political process. And yet, Public Choice economists have long dispensed liberal helpings of particular policy advice at the slightest provocation, because like virtually all economists, they believe that the advice given by people who actually study policy questions has at least some effect, on the margin, sometimes.

The avout in Anathem made a study of all the different mental models the saeculars have had of them over the ages. They believe, and quite rightly so, that these mental models have real consequences for the way they interact with the saecular world, and they are quite justified in that belief. At the same time, it seems unlikely that these oconographies are the product of environmental factors in any straightforwardly deterministic sense, so they are entirely correct in relying on empirical observation in order to establish which iconography is prevalent at the beginning of a given apert. I apologize for all the in-world jargon, but the bottom line is that the community of people that the book's main character is part of is idealist and not materialist, and rightly so.

Stephenson can write endings

2009-08-29T08:40:00.004-04:00

Neal Stephenson has gotten frustrated by the prevalent idea that "Stephenson can't write endings," because he feels that it has become such an idee fixe that people will find something wrong with his endings no matter what. I think he is being unfair to his critics. To be sure, he has made serious attempts to write more traditional codas to his books, moving away from having events just speed up and spin out of control so as to end the book in a singularity.

But what people are, I think, upset about is not so much that the ending is bad as that it happens when it does, that is, before all the ends are tied up. We have come to expect of novels, and speculative fiction novels in particular, that (in the words of W.F. Hermans) not a sparrow falls or it has some significance. Note that this is not at all true in the real world. I do things all the time that have no relation to anything later in my life; I start projects I never finish, not because they become irrelevant but because I forget; I meet people that I never see again; and so on. But I am not a character in a novel. In Anathem, the main character's girlfriend is separated from him, starts a relationship with some a good friend of the main character's, and later becomes involved with the main character again. There is much tension over this between the main character and the other boyfriend, but we never find out before the book ends what happens to this tension. And that is simply not allowed by the Rules of The Game, where "The Game" is the game of novel writing.

Stephenson might retort tha gossip is immaterial to the point of speculative fiction. Speculative fiction is not about psychology but about sociology, not about character development but about history and ideas. A "good yarn" is important, but from the way Stephenson frames his apology for yarn spinning, you get the idea that it is mostly sugar to make the medicine go down. Rounded characters are still appreciated, but mainly because they help paint a more vivid picture of the systemic story. And the systemic story is wrapped up, or at least left hanging from a plausible cliff: peace is established with the aliens, and the strict separation of avout and saecular, having been found wanting in times of exogenous crisis, comes up for grabs.

I can live with that, but I guess it is fair to say that books like Anathem aren't really novels in the modern sense. Which is fine. Anathem is interesting and enjoyable and stands firmly within the epic tradition, which is far older than the novel, for what that's worth.

BBC wit

2009-08-23T05:07:00.003-04:00

"Mr Kim won the Nobel prize in 2000 for his work to foster better relations with North Korea and his death seems to be having a similarly positive effect."

http://news.bbc.co.uk/2/low/asia-pacific/8215928.stm

In the Loop

2009-08-13T07:49:00.001-04:00

Saw this movie. It's hilarious.

Okay, a little background. First, there was Machiavelli, and later, Public Choice Economics. And Whitehall Farce. Not the kind performed at the Whitehall theater, but the kind going on at Whitehall: Yes Minister, Yes Prime Minister, and The Thick of It. Now, In the Loop.

Pentagon to world: "Get lost!"

2009-08-13T05:03:00.005-04:00

It is a little known but crucial feature of US military strategy that in the case of invasion by an army composed of anything but Northern Virginia natives or Defense Department bureaucrats, the Pentagon will never be at risk. The simple mechanism employed to this end is a network of roads so tangled and twisted that even with the best areal photography, there is no way to find out how to actually get to the place.

Pay special attention to last photo. Coming from I-395, which ramp gets you to the South Parking Lot?

Using Google Spreadsheets for psychometrics

2009-08-12T18:33:00.005-04:00

We decided (for some reason that this post is not about) that it would be useful to add some basic psychometric data collection to our experimental economics experiment thing. In other words, when undergrads come into our lab to play simple game theory games with each other, we first want them to fill out a list of questions from which we will deduce how extroverted, neurotic, agreeable, &c they are.

But this is not the main thing we are researching, and we don't want it to take too much time, of our subjects or of us.

Which is where Google Docs comes in. There is a feature in the Spreadsheet part of Google Docs where you can create an online form linked to a spreadsheet. Looks like so:

The responses feed into a spreadsheet automatically, where you can then score them using some simple spreadsheet equations. We just tell everyone in the room to fill this thing out, and the moment they hit "Submit", we have their Big 5 personality scores in a spreadsheet.

To be sure, I could have just programmed this myself as a little web application. The same may be true for you, dear reader. But this is much simpler than that. It requires about the same skill level as using a spreadsheet, or a statistics package, which means that there is no messing around with servers and interpreters and programming languages and other bits and pieces that turn programming these sorts of things into a specialist's job even if the task is fundamentally pretty simple.

The first pancake

2009-05-27T03:28:00.004-04:00

The first pancake never works. This true of small, fluffy, American pancakes. It is true of big, thin French crepes. And it is true of big, slightly thicker, Dutch pannenkoeken. (When I was a kid, they were "pannekoeken," but a spelling reform intervened.) And you can find many explanations of why this is so on the world wide interwebs! Too hot. Too cold. Too much grease. Not enough grease. Enough grease, but not coating the pan in the right way. Whatever. It never works. Which is no big deal if you are making a stack of kaaspannenkoeken (Dutch-style thick crepes with Gouda cheese filling, glad you asked) half a foot tall. But if you make just a single pancake, you're down to a pancake efficiency of 50%.

Nonetheless, pannenkoeken are great. The way to cook them has been engraved in my memory since the time I sat on top of the counter watching my dad make them, but I choose my words with care: "way to cook" does not a recipe make. Pannenkoek cooking is not like making a cake, 2.5 cups of this, three tablespoons of that, bake at 330 degrees for 27 minutes unless you have a gas oven, in which case it's 28. It's more of an intuitive thing, involving you-know-it-when-you-see-it every step of the way.

For authentic Dutch pancakes, beat 1 cup flour per person with 1 egg (always one egg), a little salt, and enough milk to make it runny. You may want to add more eggs, actually, but the way I have the recipe stuck in my head, there's one egg regardless of the amount of flour. No eggs at all and your pancakes just don't work, although the dirty little secret is that if you substitute soy for cow's milk, the egg is somehow superfluous. Heat and lightly grease a skillet (Blue Band margarine was the traditional thing in my birth household, but cooking spray will do), pour a ladle full of batter, and swivel the pan around. If the batter does not coat the entire skillet, your batter wasn't runny enough. That's what I mean by you-know-it-when-you-see-it. Making sure the batter the batter is just runny enough to coat the entire pan when you swivel it, but no runnier, is the only indication I have of how much milk to use.

Also, we tend to use "pannenkoekenmix," or "pannenkoekenmeel," or some such, rather than flour. It's a contraption of which the commercial availability probably does not extend more than a few hundred kilometers from the Koopmans Corporation of Amersfoort's headquarters, but rest assured: as far as I recall, it's just a mixture of wheat and buckwheat flours, with maybe a little baking powder added, depending on which type you get.

But for added authenticity, be sloppy about the whole recipe. Dutch cuisine is not supposed to be pretentious. Just don't be so sloppy that it doesn't work. I'll now the nature of your sloppiness when I see it. Or rather, when I taste your pancakes.

And yes, you are supposed to flip the pancake when the portion of the top that's uncooked is small enough that the batter will not run all over your stove as you flip it. And no later than that. As with the consistency, that's an ex post measure of when to take an ex ante action, so the real solution is, again, practice.

Next step: kaaspannenkoeken. This is tricky. You put the batter in the pan like before, and after the bottom gets cooked but before the top gets cooked (and you have no way of knowing when the bottom cooks!), you add some slices of Gouda cheese, cut with that most unpronounceable of kitchen implements: the kaasschaaf. Now, you move the slices of cheese around a little, the point being that they need to be covered in batter. You may find it necessary to use some extra batter; the amount stuck to the bottom of the ladle tends to be just right. But don't use too much, because you will still not be able to flip the thing before the top is mostly cooked, and the top will now cook slower, which means the bottom might burn before the top cooks. Oh, oh, and you thought that Dutch cooking couldn't be all that hard, given how prosaic it tastes, well, we've got you fooled. Pancakes aren't exactly an art, but they are a habit, which takes time to acquire.

Actually, I think many dishes are much more like pannenkoeken than cookbooks let on. Cookbooks like to take, well, a "cookbook approach." They make it all seem as though cooking is about following a series of steps. All ratio, no muscle memory. Bullshit, of course. But marketable bullshit. That's why I like YouTube recipes. You see people do their cooking the way they actually do it, before it gets reduced to the near-mathematical abstraction of a formulaic recipe-book ingredient listing, "add one medium julienned, caramelized, pureed red onion and heat au-bain-Marie until bubbly near the edges."

Just remember: the first pancake never works.

Beer again, and Will Wilkinson, and Behavioral Public Choice

2009-05-21T22:07:00.002-04:00

Will Wilkinson says all I would want to say about conservatism and its problems, complete with metaphors I could only dream of, so go read him.

Also, not all Trader Joe's store brand beer sucks. Their "Bohemian lager" is entirely inoffensive, if maybe a little bland.

Finally, behavioral public choice is the way of the future. More to come later.

Word of the day

2009-05-11T16:59:00.001-04:00

Cisgender. Apparently, it started out as a witty neologism and is now actually fairly widely used. Reminds me of the way I like to think I'm pretty funny when I call the non-overpriced tomatoes at Wegman's “inorganic,” but apparently people are using that word, too, to denote non-organic food, without any apparent sense of irony. Hint: the only actually inorganic foods I can think of are water (H2O), salt (NaCl), diet Salt (KCl) and salmiak (NH4Cl). Tell me if you know any others.

Making It Through Graduate School

2009-05-03T12:54:00.001-04:00

From "David Romer's Rules for Making It Through Graduate School and Finishing Your Dissertation," a.k.a. "Out In Five":

Don't try to game the profession, work on what interests you.

I would love to believe that is true, because gaming the profession is booooring, and it also defeats, at least in my mind, the point of doing science in the first place, which is to Figure Things Out.

But... isn't there a potential source of bias here? What if working hard on what interests you without regard to "gaming the profession" happens to work for those people who happen to be interested in things that happen to yield papers that happen to fit the prejudices of the profession? What if people of that type are the people who end up in positions where they get to pontificate about How To Make It Through Graduate School And Finish Your Dissertation?

Mind you, my intuition is still that Romer is right. I think that at least for me the added productivity coming from working on something interesting with people that you get along with far outweighs the potential career boost from trying to game the profession by working on things that interest "the profession" with people that are good at impressing "the profession."

But... I'm in a poor position, too, when it comes to generalizing my attitude into advice to give other people. After all, I have a decent backup career in case I don't make it as a professional economist—I can program computers. Also, I neither have nor plan to have children to support, which increases the risks that I can take without feeling like I am neglecting a duty to others. Such things inevitably lead me, if only subconsciously, to be less risk-averse than I otherwise would be, and it's all too easy to take one's own levels of risk aversion and time preference and project them onto other people.

And, of course, the inevitable economist's disclaimer: the choice between Interesting Work and Gaming The Profession is not an absolute one. It is a tradeoff, and we should seek to equalize the net benefits of moving in one or the other direction at the margin. Obviously, nobody would argue that it is a great idea to work on topics so unmarketable that you'd starve trying to pursue them, or on topics so boring that you'd be happier taking up custodial engineering as a career instead.

"Frugal Joe's Ordinary Beer" from Trader Joe's [REVIEW]

2009-04-27T19:10:00.002-04:00

May be drinkable if you have a severe cold or are willing to hold your nose. Not quite as bad as long as it’s really cold, but if that’s what you’re relying on, you’d better drink fast, before it warms up any.

Scala considered useful

2009-04-26T23:36:00.004-04:00

Programming language features have a very long trickle-down time. If you want a programming language feature to make it into the kind of languages used for things like large-scale corporate data processing 30 years from now, you'd better invent it yesterday. There are several reasons for this, all of which come down to the fact that inventing a language feature in isolation and providing all the things necessary for its benefits to outweigh its costs in practical applications are very different tasks.

Lisp, for instance—after FORTRAN the programming language still in use—has long had features like automatic garbage collection that took a long time to make it into commonly used languages. The language that probably did most to popularize garbage collection was Java. But otherwise Java never seemed like a good idea from the perspective of a programming languages geek. It did not seem to buy you anything fundamentally new, and at the same time, it still lacked many desirable features of advanced but uncommon languages. Guy Steele, in an effort to convince fellow Lispniks of the value of Java, had to put it like so: “We managed to drag a lot of [C++ programmers] about halfway to Lisp.”

With Scala, there is a chance that many of the same programmers can be dragged even further toward Lisp, and even beyond Lisp, into the realm of things that up until recently were only available in those select few languages to have made significant progress beyond Lisp, like ML, Haskell, and Erlang. True Lisp fundamentalists will no doubt complain that Scala is not homoiconic and has no macros. But ML programmers should be delighted: here is a language that truly integrates object-oriented and functional programming, rather than bolting one onto the other as in OCaml.

There are many other good things, and a few bad, to be said about Scala, but the two that are perhaps most important are the ones that make it feasible to mix Scala with Java, replace Java with Scala, and replace Scala with Java, all at once, or a bit at a time:

Scala compiles to JVM bytecode .class files.
You can link Java and Scala against each other.
Scala can call Java.
Java can call Scala.

Just like that. Look ma! No FFI!

chortle

2009-04-26T16:13:00.001-04:00

chortle, v. [neologism from Lewis Carroll's poem Jabberwocky] To emit a brief snoring sound while chuckling. Even though Ray Magliozzi chortles whenever he hears his brother say it, this is NPR, National Public Radio.

The neat thing about microinnovations...

2009-04-22T10:57:00.001-04:00

...is that they rarely come with a good excuse for why they were not invented before.

There is no particularly good reason why, for years, pedal trashcans without holes in the inner bin were manufactured and sold all over, aside from the fact that nobody thought of making the bins with the holes in them. The opportunity had been staring trashcan manufacturers in the face for decades, and they simply never noticed.

But it is exactly because it seems, in hindsight, so easy that, say, the shipping container never gets the sort of attention that is lavished on nuclear energy or the sequencing of the human genome or the space race. (Admittedly, I believe there is a book on the shipping container, but it's clearly considered a kind of quirky thing to write a book about.)

And that (the story must have a moral!) clouds the way we think about science and technology policy.

Microinnovation of the minute

2009-04-22T10:47:00.000-04:00

My cell phone has a tiny little mirror next to the camera lense. This lets you take pictures of yourself, with or without other people or buildings in the background etc., even though you can't watch the screen while you take the picture.

Microinnovation of the minute

2009-04-22T10:45:00.001-04:00

Our metal pedal trash can comes with an inner plastic bin that has a hole near the top edge, about an inch in diameter. This lets you neatly use bin liners that are too big by stuffing the excess liner through the hole.

Microinnovation of the minute

2009-04-22T10:44:00.001-04:00

Caribou has better coffee cup lids than 7/11. You can push them on without spilling coffee and they don't come off by accident. This is because of a tiny difference in the design, where the lid basically clamps the edge of the cup from two sides rather than from the outside only. And this feature is extensively advertised in the catalog of the supplier that Caribou gets their cups from.

Hayek and propaganda

2009-04-21T11:06:00.001-04:00

“Public opinion, however, cannot decide in what direction efforts should be made to arouse public opinion.” (Hayek, Constitution of Liberty §8.6)

Actually it can, but it probably shouldn’t. If it is allowed to, we obviously set ourselves up for a positive feedback loop. That's why it's bad to have the DEA fund local ballot initiative campaigns. I would even say that it is a good argument against the Dutch government's well intentioned Postbus 51 system of public service ads, which regularly crosses the line between educating people about policy and actively advocating for the status quo or even for particular changes to it.

My new favorite German word

2009-04-21T04:42:00.001-04:00

Zwetschgenschnecken

My browser collects its own garbage...

2009-04-18T21:33:00.002-04:00

...sort of.

Actually, my browser leaks memory left and right. By the time the memory is dripping all over my desk, I shut down the browser, and I open it up again. The browser is pretty good at remembering all the interesting bits of its state—what pages I was looking at, what cookies were being kept—but power-cycling the whole thing neatly gets rid of all the uninteresting bits of state: the garbage. Now, given that I have to do this anyway, what danger is there to adding a real garbage collector? I understand that garbage collection has overhead (although the relative performance of GC and malloc/free actually depends on a bunch of details of the program's behavior), but I can't imagine that automatic garbage collection is really any different from "manual" garbage collection by restarting the program, now is it?

Then again, I think the real problem is slightly nastier. Even a garbage collector is no guarantee against poor design decisions. You can write a program in a nice garbage-collected environment and keep live pointers to some big data structures around in a global table and you'll leak memory anyway.

This is an instance of a more general problem: abstraction in computer science is ever really an abstraction. No matter how nicely isolated and modularized and packed up with a little ribbon on top your code is, the caller will notice what's inside, if only because it influences performance. A garbage-collected heap is not really an infinite supply of cons cells, a remote procedure call is not really a procedure call, and so on. That's why the best programmers are the ones that can think of clean and elegant modular structures, but at the same time keep in the back of their mind a decent model of what's underneath, a box they can open up if they have to but that won't clutter their thinking otherwise.

This, incidentally (but now I'm really starting to rant), is also the reason why in Computer Science education, it is crucial to use languages that the student can build a complete, no-compromise mental model of. I recommend Scheme for introductory programming, along with a book like SICP that builds an interpreter and a compiler and everything, and C for systems stuff, along with a good text on computer architecture. Anything else and the student will end up treating computers like magic boxes, inscrutable oracular deities that cannot be understood but only supplicated. Of course, there might exist some students that you can just throw C++ at in their first year and they'll form a complete understanding of the language syntax and semantics and the underlying implied machine model and the extent to which that model actually maps on the machine they have... but I've never met anyone like that. Languages with sophisticated type systems, like Haskell and ML, or complicated object-oriented semantics, like C++ and Java, are not suitable for introductory programming. In fact, I could argue that C++ is not suitable for anything, because unlike with ML, which is clever but not complex, so that it becomes easy once you "get" it, nobody ever "gets" all of C++ down to the last little rule about the order in which things are initialized and the like. The best any mere mortal can do with C++ is to become confident of the semantics of a subset and scrupulous of staying with the subset's boundaries.

As I said, a rant. Done kvetching now. Back to work.

Bank takeovers in Google Street View

2009-04-18T06:28:00.001-04:00

If you search for "TD Bank in midtown," you get pictures like this:

View Larger Map

I wonder how often Google updates their Street View images...

Most awful pun ever

2009-04-12T05:22:00.001-04:00

By Guy Steele.

Re: need for macros (was Re: Icon)

And some people prefer not to commingle the functional, lambda-calculus part of a language with the parts that do side effects. It seems they believe in the separation of Church and state. :-) :-) :-)

If you can't beat 'em...

2009-04-08T21:47:00.001-04:00

On TechCrunch today:

"Through its Content ID program, [YouTube] digitally fingerprints the video libraries of 600 content partners and offers those media companies and video creators the option to place ads against any video that contains their copyrighted material. They also have the option to automatically take those videos down, but more than 90 percent opt for the additional ad revenues."

An ML programmer's rant

2009-04-08T08:56:00.001-04:00

Algebraic data types are not exactly the hardest concept in computer science. Given two types A and B, we want to have a type A*B that contains a value of type A as well as a value of type B, and we want to have a type A+B that contains either a value of type A or a value of type B. These are the two simplest ways in which you can put types together, and together with procedural abstraction and suitable primitives for getting values back out of these types, they are really all you need to have a workable type system.

Most language support product types pretty well. Sometimes, there actually is a tuple type, and sometimes, you have to define a record or structure or class of some sort, but something or other tends to be there.

Sum types, on the other hand, are sorely lacking in many languages. The only way to create an A+B type in most object-oriented language is to make A and B both subclasses of a superclass S. For one thing, this cannot be done after A and B have already been declared, so that the only way to make a discriminated union of two library types ends up being encapsulating both of them in containers that are subclasses of a supercontainer (ugh!).

But what's worse, it forces a very artificial style of programming. Normally (and by that, I mean, in a “civilized” programming language...) when you have a value of type A+B, and you have some operation of type A->C, and some other operation of type B->C, you can turn your A+B into a C. Note how nicely this corresponds to logic, by the way: if you know that either A or B is a true proposition, but not which, and additionally you know that both A and B imply C, then you know that C is true, albeit not why.

Object-oriented dogma has it that this kind of dispatching on the tag of a tagged union is somehow dirty and that it needs to be sanitized by having the A->C function tied to the definition of A and the B->C function tied to the definition of B. Aside from the fact that, again, this makes it hard to do anything to library types you have no control over, it can also force the programmer to separate code that logically belongs together over many different class definitions.

There are some profound-sounding philosophical arguments for why this sort of inside-out programming is a good idea, but I think that what is really going on is that the lack of any simple, type-safe implementation in popular languages like C++ and Java is the good reason to avoid it, and any more philosophical arguments are mostly rationalizations for why what would otherwise be an obvious gap in the language is actually Right and Proper.

As far as I am concerned, the Right and Proper style depends on the algorithm. If I have 20 different flavors of Ticket objects and I want to be able to convert them all to strings for easy display and debugging, I have no objection to a style of programming where each stringification function goes with the corresponding type. The stringify functions are not, usually, mutually recursive, and do not conceptually form a single algorithm.

If, on the other hand, I have an abstract syntax tree on which I want to do some complicated transformation, say a continuation passing transform, or lambda hoisting, or common subexpression elimination, or some other arcane compiler algorithm, then I find it hard enough to keep one of these heavily mutually recursive algorithms in my head if it's all right next to each other. The idea of having a lambda hoisting algorithm spread out over 20 different class definition files is the sort of thing that makes me wake up in the middle of the night and scream.

Fortunately, the Gang of Four comes to the rescue and has invented a Pattern. My rule of thumb for Patterns is that in 9 out of 10 cases, they are workarounds for an obvious language design flaw. In the case of the Visitor pattern, which requires an RSI-inducing amount of boilerplate typing just to accomplish a tag dispatch on a tagged union, we are clearly looking at one of the 9, and not the 10th. I will spare the reader's stomach for now; if you must learn about the Visitor pattern, I recommend googling it.

Another recipe

2009-04-04T01:44:00.001-04:00

Plain tofu at the Korean store is $1 a pound. Nice, marinated Tofu from Trader Joe's that actually tastes good is easily twice that for half the amount, if not more. Unfortunately, it is not that easy to make it yourself. Or, at least, it did not seem that easy until I figured out the trick.

The trick turns out to be that for tofu to absorb marinade, it first needs to loose some water. There are many tricks to this end. The very least you can do is to drain it properly, put it between layers of (paper) towel and press with a big cutting board, drain it some more, and so on. I've heard kosher salt proposed, but I find that the salt from the soy sauce in the marinade already makes my tofu at least as salty as I want it.

Which brings me to the next trick: dry frying. Cut the tofu into slices, drain and press as much water out of them as you can, and then “fry” them, without oil, in a non-stick pan. The water drips out and boils off. Be sure to move the pieces of tofu around every once in a while, and turn them over when they start to brown. They will develop a tough, brown layer that is, as far as I can tell, the “tofu skin” that you get at sushi places.

It's a time-consuming process, but it dries the tofu out to the point where it will absorb marinade like a sponge. To make the marinade, mix 1 cup soy sauce, 1 cup strong stock, 1 chopped onion, 2 tbsp chopped garlic, 1 tbsp ground coriander. Microwave for a minute, pour over 3 pounds of sliced and dry-fried tofu in a shallow dish of some sort, let sit for a few hours, retrieve and freeze the tofu, and recycle the marinade for another batch. Wax paper works well to keep the tofu slices from freezing to each other. (Saran wrap does not work. Trust me, I tried.)

Dinner

2009-04-04T01:26:00.001-04:00

I'm not sure if my recipes are really at all interesting for anyone, but here's another one.

Sautee 1 tbsp chopped garlic and 1/2 chopped red onion in plenty olive oil, add 1 can stewed tomatoes and 1 splash of beer, simmer, mix in 1 handful of shredded mozzarella, serve over spaghetti boiled al dente in salted water.

Three cheers for Hal Abelson

2009-04-02T21:10:00.001-04:00

His list of accomplishments already involved co-authoring the best intro CS textbook ever written, but now he has gotten MIT to require its faculty to make all their papers publicly available.

Peter Suber, “MIT adopts a university-wide OA mandate”

The Faculty of the Massachusetts Institute of Technology is committed to disseminating the fruits of its research and scholarship as widely as possible. In keeping with that commitment, the Faculty adopts the following policy: Each Faculty member grants to the Massachusetts Institute of Technology nonexclusive permission to make available his or her scholarly articles and to exercise the copyright in those articles for the purpose of open dissemination. In legal terms, each Faculty member grants to MIT a nonexclusive, irrevocable, paid-up, worldwide license to exercise any and all rights under copyright relating to each of his or her scholarly articles, in any medium, provided that the articles are not sold for a profit, and to authorize others to do the same. The policy will apply to all scholarly articles written while the person is a member of the Faculty except for any articles completed before the adoption of this policy and any articles for which the Faculty member entered into an incompatible licensing or assignment agreement before the adoption of this policy. The Provost or Provost's designate will waive application of the policy for a particular article upon written notification by the author, who informs MIT of the reason.

How do you like that there econ department?

2009-04-02T21:05:00.001-04:00

The George Mason economics department is home to the full spectrum of political thought from outright individualist anarchism to Fox News style conservatism—and also to the full scale of political rhetoric from the sort of discourse that does not shy away from the hard bits and takes opponents seriously to the sort of pseudo-philosophical rambles that do well in columns that web sites can run with Ann Coulter ads alongside.

Take this:

“Our Problem is Immorality” by Walter E. Williams

The reason why your college professor, politician or minister cannot give a simple yes or no answer to the question of whether one person should be used to serve the purposes of another is because they are sly enough to know that either answer would be troublesome for their agenda. [...] I am all too afraid that a historian, a hundred years from now, will footnote America as a historical curiosity where people once enjoyed private property rights and limited government but it all returned to mankind's normal state of affairs—arbitrary abuse and control by the powerful elite.

And this:

Cafe Hayek: I Am a Liberal

I love and care for my family and friends more than I care for mere acquaintances, and I care about most mere acquaintances more than I care about total strangers. But the nationalities or physical locations of these people's residences are irrelevant to me. I care no more for a stranger in my town of Burke, Virginia, than I care for a stranger in Beijing, Beirut, or Berlin. If this claim sounds harsh, let me say the very same thing differently: I care as much about a stranger in Beijing, Beirut, or Berlin as I care about a stranger in Burke, Virginia. I accord all strangers the same rights and respect. I regard the well-being of strangers in foreign countries to be no less important than I regard the well-being of strangers in America.

I trust that I don’t have to go to the length of personally insulting one or the other professor for my dear readers to find out which of the aforementioned quotes fits into which of the aforementioned categories.

A major step for open-access education

2009-04-01T20:46:00.001-04:00

Years ago, Phil Greenspun and some of his co-workers at ArsDigita (an early web startup) put together a one-year computer science training program. The thing was targeted toward people who already had a college degree, but not in CS, and the intention was to cram all of the most important parts of a CS major at a top-20 school into a year. They shot video of all the lectures and it is still available on aduni.org, but only in a really strange flavor of RealVideo that turns out to be nearly impossible to actually play on a lot of modern software.

About a year ago, I started the process of transcoding all these videos and uploading them to Google Video, but I got bogged down mainly by the fact that Google Video made me enter all the metadata by hand.

Fortunately, someone else with more patience has finished the job for me. And such are the benefits of the zero-marginal-cost information economy that the rest of the world now has, at the price of a broadband internet connection, access to a first-rate one-year intensive sequence of computer science courses.

I hope that this will be picked up sooner or later by YouTube EDU and/or academicearth.org; the video quality is not the greatest, but this does represent a more complete CS curriculum than you can find anywhere else on the web.

Coors between devil and deep blue sea

2009-03-30T04:18:00.001-04:00

Here's the kind of quote I really don't get.

Beer Wars Movie

While making the film, I was surprised that many of the brands on supermarket and grocery store shelves that appear to be made by small craft brewers actually come from BudMillerCoors. OK, I wasn’t really surprised because I knew how things worked having been in the beer business but the proliferation of “craft style beers” was staggering. How is a consumer supposed to know what’s a “real” craft beer. Is it the type of beer? Or the brewer that matters?

Why on earth would I care whether Blue Moon is made by Coors? The reason we disliked Coors to begin with is that the aromas of the beers they used to make had a distinct resemblance to those of feline urine, to put it mildly.

Which is why Coors decided to make some better beer. (They don’t exist, after all, to purposefully insult our taste buds, but to make a profit by making stuff we are willing to buy.)

So now Coors makes some beer that actually tastes good, in this case a sort of knockoff Hoegaarden.

And now we blame them for doing precisely what we used to blame them for not doing?

Besides, most of those lovely Belgian beers are made by one subsidiary or the other of Anheuser-Busch InBev. Just so you know.

All I know about politics I learned from Sir Humphrey!

2009-03-27T09:18:00.001-04:00

Many people are blaming the latest dispensary raid on Obama, or Holder, or both being a hypocrite. Now I cannot deny that I am quite inclined to believe that hypocrisy is indeed a job requirement for any politician as powerful as either of them, but here is another possibility:

“It would be different if the government were a team, but in fact they are a loose confederation of warring tribes.” (Yes Prime Minister, Series 1, Episode 3)

And while we are at it... I just dug up an even better quote from that same episode:

“Notwithstanding the fact that your proposal could conceivably encompass certain concomitant benefits of a marginal and peripheral relevance, there is a countervailing consideration of infinitely superior magnitude involving your personal complicity and corroborative malfeasance, with a consequence that the taint and stigma of your former
associations and diversions could irredeemably and irretrievably invalidate your position and culminate in public revelations and recriminations of a profoundly embarrassing and ultimately indefensible character.”

While we're on the topic of OTC drugs...

2009-03-27T09:10:00.001-04:00

...here’s my quote of the <insert-unit-of-time>.

Teen Strip Searched ... for Ibuprofen

"Remember," he says, "this was prescription strength Ibuprofen."

Melatonin

2009-03-26T17:35:00.001-04:00

Melatonin is an OTC drug that is very useful in adjusting circadian sleep rhythms that have been knocked out of whack by intercontinental travel, night shifts, a graduate student lifestyle, and the like.

But for some reason, US drug stores tend to sell it in much too high of a dosage (3mg). Dutch drug stores sell it in 1mg tabs, which works fine for me, but apparently even that is too high, since clinical studies have estimated the effective dosage as low as 0.3mg.

There seems to be a tendency among Americans to want to use OTC drugs at high dosages “just in case.” It’s not as bad as in France, but it’s there.

To some extent, I like that. As liberal as the Dutch are about recreational drugs, our attitude to the medical ones is often overly puritanical, to the point where anesthesia for childbirth is generally thought of as a very naughty thing that a woman has to plead for. (As far as I understand, in France you’d have to plead not to get it, and in the US, it’s a common option. But I have never given birth to a child in any of these countries, so YMMV ;-) )

Anyway, back to the melatonin. With drugs that act on the central nervous system there’s likely to be all sorts of strange nonlinearities in dosage response, and more is not necessarily better. There is no point in taking more if it’s going to make you drowsier the next morning without actually helping you sleep.

From here, I could conclude either on a ranting note (the damn drug stores don’t give a damn about me), or I could point out the business opportunity (sell “non-drowsy works-better MIT-tested formula” melatonin in a fancy bottle and see if you can beat the store brand.)

But I don't know which of those moods I'm in.

Maybe they’ve got a pill for that?

God Hates Virginia

2009-03-25T11:15:00.001-04:00

Or so Fred Phelps would have us believe.

My piece about his upcoming picket at GMU was published by thenewgay.net:

Fred Phelps is coming to Fairfax on Monday, March 30, to protest George Mason University’s celebration of the annual Pride Week, when attention is called to LGBT issues and students.
Now it is nothing extraordinary for Fairfax to play host to an elderly, self-proclaimed Baptist minister from Kansas. But this man is no ordinary preacher. Phelps’s is a gospel of hatred, and his preferred way of preaching it is the street-corner picket.
When he pickets the funerals of US soldiers killed in Iraq, members of his church carry signs saying “Pray For More Dead Soldiers.” On his Web site, Phelps explains that “the IED is God’s weapon of choice in avenging Westboro Baptist Church by blowing America’s kids to smithereens in Iraq.”
Phelps’s God, if we are to believe his picket signs, hates Catholics, Mormons, Jews, Muslims, and anyone with a Christmas tree. Westboro Baptist runs Web sites named godhatesireland.com and godhatessweden.com, and Phelps has wished especially large helpings of fire and brimstone upon the unsuspecting nation of Germany. But his bitterest bile is embodied in his most notorious slogan: “God Hates Fags.”
Which is where Fairfax comes in, for underneath the peaceful surface of our fair city dwells—according to Phelps—one festering sore: George Mason University. Where others see an up-and-coming college with a surprisingly good basketball team, Phelps discerns Sodom and Gomorrah all stuffed into one tree-lined suburban campus.
It is Mr. Phelps’s inalienable first amendment right to contribute his dissonant voice to GMU’s Pride Week celebration. At the same time, it says nowhere in the first amendment that the citizens of Fairfax should be particularly pleased to see their sidewalks decorated, however temporarily, with signs shouting “Thank God For Aids” or “God Blew Up The Troops.” This is why GMU’s Pride Alliance has worked with Driving Equality, a Boston-based group that specializes in helping communities respond to Phelps’s antics, to set up a very special kind of charity pledge drive.
The idea is simple: you visit www.phelps-a-thon.com to pledge a contribution to Pride Alliance for every minute that Fred Phelps’s protesters disfigure the intersection of Braddock Road and Route 123 with their picket signs this coming Monday morning. The contribution can be as small as you want it to be, but even if you pledge just a penny a minute, that means that for every minute that Phelps is here, he will be sending another penny to the university’s volunteer-run LGBT undergraduate club.
We hope that everyone in our community, regardless of their personal beliefs, will be able to join us in condemning Fred Phelps’s attempts to sow discord in our midst.

$PAPERSIZE

2009-03-14T17:05:00.001-04:00

There are way too many ways one can set the standard paper size on my Ubuntu Linux box. After going through what I thought was all of them, I finally found an innocent-looking

export PAPERSIZE=a4

in my .zshrc ...

No, really, wait

2009-03-14T04:30:00.001-04:00

Ran across this... (The punchline is at the end.)

Drive New York City like a New Yorker

Many locals, when stopped for a red light, have the habit of watching the lights on the GO side (clearly visible at night, or at daytime, by watching the lights reflecting on the covering shades). Then they would start going as soon as they see the red light turn on for that other side. There's actually a one-second delay before the green comes on for you. This works for those who need to get an advantage over others, like taxis racing for a fare , or a hothead who wants to put another hothead "in a rightful place". But it's dangerous and illegal. Besides there are crossings that don't always work that way - some lights let pedestrians go first, some let turning vehicles first, and so on. There will be signs warning of this that say, guess what, "Wait for the Green Light".

Some ideas about democracy in China and similar places

2009-03-13T17:15:00.001-04:00

All else being equal, democracy seems to be among the more reliably decent systems of government so far invented. Contrary to popular mythology, it does not guarantee that the government reflects the “will of the people,” both because there is no coherent way to define the “will” of a collective (except in special cases), and because spreading out the responsibility of decision-making over millions does not create a particularly strong incentive for anyone to actually understand any given policy proposal. But, equally contrary to popular mythology, all forms of government rest on the mass psychology of legitimacy, and not on abstract sources of Law with a big L—which is where a mature democracy shines. After all, in a political system where by force of custom the rulers accept the popular vote as a means to initiate transfers of power without bloodshed from one faction to the next, egregiously exploitative or unpopular regimes usually don't last, and violent means of deposing the rulers are rarely resorted to. Even without the guarantee that democracy contains a secret sauce for picking the Right Policies, democracy—once it is firmly entrenched—can confine competition for power to a civilized framework of shuffling around votes and money rather than bullets and political prisoners. And that, I submit, is a Good Thing.

It is not impossible for systems other than democracy to lead to similar outcomes, but it is rare. The major competition in the Good Government tournament is rule by an elite that happens to be reasonable. The British Colonial Administrative Office's rule over Hong Kong is probably the best—in fact, some would say, the only—example.

There appear to be two major problems with the enlightened elite system. The first is that there is usually no way to guarantee the continued enlightenment of the elite; even if a particular enlightened monarch or ruling clique is found that won't budge for ideological reasons, and remains firmly in power, it is inevitable that rulers die like everyone else, which is where the trouble usually starts. (Hong Kong cheated on that one; the colonial administrators, even if the people of Hong Kong could not depose them legally, were ultimately at the service of the British civil service, a somewhat aristocratic institution, but tempered by centuries of strong and strengthening democratic influence.)

The second problem with enlightened elites is that, if they are indeed to stay in power, they have to limit political competition. This is because even (one might go as far as to say: especially) in societies that are well governed by one group, there will always be another group eager to take over, on account of some difference of opinion or even just desire for power. This makes it hard for legally non-deposable elites to stay enlightened for long. Policy differences between democratic and authoritarian regimes are surprisingly hard to pin down statistically, but the one difference that is significant is precisely when it comes to policies that affect political competition.

Political competition may seem like a minor point; if a regime is otherwise working well, why would it need competition? The obvious answer is that, as we saw before, there is no guarantee that the regime will continue to do “well,” whatever (within reason) your standard of “well” may be. The other answer is that limits on political competition are usually limits on activities that are important to things other than politics: things like the development of technology and the basic human desire to develop and express one's personality, the higher rungs of Maslow's hierarchy, if you will.

Where does all this leave us when it comes to the widely voiced demand for democracy in China?

First of all, I have a hard time believing that there is anything special about China that would make it unsuitable to a democratic system of government. But I think that the Chinese should not be misled by the revolutionary rhetoric of many existing democratic regimes. The key to a successful democracy is not any particular piece of paper, but the implicit understanding that an elected regime is legitimate and should be obeyed, while an unelected regime is to be resisted. As long as the population, and especially the armed forces, civil service, and other assorted elites, believe that whoever occupies the presidential palace is the president, usurpations will happen. Only a strong tradition of peaceful transfers of power appears able to
consistently get the state on the side of the elected government.

Propaganda often portrays democracy as something that came to a country in one fell swoop with some revolution or another, but history shows that it rarely sticks unless there is some of it already. The best example is Britain, which suffered an attempt at democratization (the Commonwealth) in the mid-17th century that utterly failed, and then went on to have the Glorious Revolution, which is more properly thought of as the starting point for several centuries of evolution, with fits and starts, but generally speaking in the direction of greater democracy. By the time the American revolution came around, the power of elected bodies in cities, in the settlement colonies, and in Westminster had been eroding royal prerogative for a while, to put things mildly. On the other extreme there is, of course, Weimar Germany, a democracy instituted to replace the sovereign rule of a warmongering racist, only to lead to the more-or-less democratic appointment of another warmongering racist to an even stronger form of autocratic power.

Now all this is commonplace, especially in the context of “nation-building” foreign policy failures. But why should the Chinese care?

In the very hypothetical situation that anyone in China got an opportunity to rescind the constitution and replace it by something democratic, I wouldn't blame this hypothetical person for taking a chance and hoping for the best. But such opportunities are unlikely to arise, and if you think of them as the only way to improve the current regime, the upshot is inevitably desperation.

If, on the other hand, you think of democracy-on-paper as being the apex, rather than the base, of a set of political achievements, you are much more likely to get somewhere. There are, to my knowledge, numerous opportunities to improve the Chinese political situation. Chinese government is not an autocracy so much as a big, messy, complicated oligarchy, where oligos (Greek for “few”) means “a few million,” easily. Within such a system, there are all sorts of degrees of freedom that can be manipulated, at least by some people.

Take the role of judges. To my knowledge judges in China traditionally think of themselves as law enforcement personnel in the service of the government, but that hasn't prevented the intrusion of a Western legal ideology where judges see themselves as serving abstract principles of property, tort, and contract. In a country as big as China, the private ideologies of officials such as judges are bound to have some influence on the day-to-day workings of government that cannot be controlled with arbitrary precision from on high. Conclusion: judicial training programs for Chinese judges that acquaint them with Western legal principles are probably a Good Thing. I, for one, am glad that there are some.

Similarly, I have heard that Communist Party training programs have started to drift away from high theoretical Marxism to teaching more practical matters, and that the elite status conferred by Party membership has been attracting many for whom the old ideology is more of an annoying formality than a zealous passion. To be sure, as we discussed before, it is hard for any monopolist ruling class to act in an enlightened way while maintaining its monopoly, but within the boundaries set by power politics, there is room for ideological influence on policy making. I can imagine worse things than a Communist Party staffed by people who actually understand some basic neoclassical economics and cost-benefit analysis: for instance, a Communist Party staffed by people whose education consists of learning Marx and Mao forwards and backwards.

There is, of course, a chance that the evolution of the regime toward more responsive policies and more flexible institutions will crowd out attempts to impose Western-style democracy overnight. If one's goal were to “get to Denmark” on paper ASAP, damn the consequences, one would have to discourage anything that might make the current regime work better, the quicker to have it break down. But if your goal is to ”get to Denmark” in a sustainable way, you have to look at what you've got and start from there. You need to look at the existing system, and at your own position—as an official, an intellectual, a philantropist, a businessman, whatever—and see what you can do, not dream up the blueprints for an imaginary divine intervention.

If anyone who actually has some training in political science and/or actually knows something about China reads this: feel free to comment, this is a ramble, not an edict, your mileage may vary ;-)

The same is true for anyone without such specialized knowledge, actually.

C++

2009-03-10T15:31:00.001-04:00

For the amount of time I have spent doing computer programming, it is perhaps really surprising that I never learned any C++. But now circumstances have finally arisen that have made pick up a (virtual) copy of Bjarne Stroustrup's The C++ Programming Language. I still don't think I'm going to like what I see very much, but I'm encouraged by the following:

Remember that much programming can be simply and clearly done using only primitive types, data structures, plain functions, and a few library classes. The whole apparatus involved in defining new types should not be used except when there is a real need. (p.15)

Not use classes for absolutely everything all the time as much as possible? That sounds very different, much more imaginative, and much more pragmatic than the object-oriented fundamentalism I remember from intro-to-Java-101.

Why blog when Wilkinson does?

2009-03-06T21:29:00.001-05:00

Read him.

Place. Limits. Liberty.

Anyway, I was drawn in by this amusing passage in this Daniel Larison post: "[L]et us reflect on the fallen state of man. How did it happen, [...]" I read this to Kerry who submits that “it sounds like he’s talking about Dungeons and Dragons or something,” [...]

Limited liability

2009-03-05T12:46:00.001-05:00

Limited liability insurance companies, to me, have always seemed like a poor substitute for syndicates of elderly notables liable to the last penny and acre. Unfortunately, the last remnant of the old way of doing things collapsed under the weight of strange tax policy, fraud, and other nasties back in 1996. The new(ish) way of doing things involves regulating a limited liability corporation into sufficient prudence that bankruptcy never actually hits. We've seen how well that works. At the same time, many other mechanisms we have for dealing with judgment proof defendants require precisely that people insure themselves—the canonical example being mandatory car insurance. That is why you cannot simply trust customers to police insurance companies, either.

Now let us imagine a world without limited liability laws: all property and all contractual obligations are held by individuals or by jointly and severally liable partnerships. Could we have a financial crisis like the current one in such a world? On the other hand, how much of a hit would we take in terms of economic growth?

I don't have the answers ready, but I think those questions ought to be thought about.

Rauch gets it right

2009-03-05T12:24:00.001-05:00

Was I too harsh on Jonathan Rauch, two posts ago? Maybe. I actually entirely agree with this:

Independent Gay Forum - Shaping the Battlefield

I talked the other day with a California-based political consultant who explained that the problem we faced with Proposition 8, and other anti-gay-marriage ballot fights, is that short-term tactics and long-term strategy work at cross purposes. In the short term, the election outcome is decided by a narrow group of swing voters, and these folks are turned off by appeals that feature gay people or gay couples (especially with kids). But running vague, de-gayed ads that appeal to this group means we never make the positive case for marriage, which is the key to moving public opinion and mobilizing support in the longer run.

Culinary update

2009-03-03T17:07:00.001-05:00

Cheerios (or their Wal-Mart store brand knock-off) make excellent salad croûtons. They really do.

The logic and psycho-logic of self-oppression

2009-03-02T18:50:00.001-05:00

Here's Jonathan Rauch:

Independent Gay Forum - Why We Lost Prop 8...

I've been arguing for some time that we will not win marriage by dismissing opponents as haters and contrary arguments as proof of bigotry. We must make a positive case and respond frankly and respectfully to opponents' qualms. We must be prepared for the obvious attacks, instead of believing that justice will prevail. Above all, we must not talk to the voters as if we were entitled to their support ("Don't put rights up to a vote," etc., etc.). If you don't believe me...ask Frank Schubert.

On the one hand, Rauch is most likely a much better political strategist than me. It doesn't take much to be a much better political strategist than me. And he is probably right, too.

On the other hand... I seriously doubt that a lot of LGBT folk are emotionally up to the task of treating their opponents the way that Rauch proposes.

There is a real psychological tension between the alleged (and plausible) fact that it's good tactics not to claim entitlement to rights and the equally plausible idea that we are, indeed, entitled to them. It's not logically inconsistent to believe you are entitled to something and talk like you're not, but it's psychologically challenging.

The issue here is what an earlier generation of crypto-Marxist gay liberationists called self-oppression: the tendency of the oppressed to internalize the norms of the oppressor. Whatever else you may think of the rhetoric of that bygone era—too abolute, too deterministic, too... Marxist—they had a point when they said that "the ultimate success of all forms of oppression is our self-oppression. Self-oppression is achieved when the gay person has adopted and internalised straight people's definition of what is good and bad."

There is no way, I intuit (but cannot prove), to win the respect of a society without winning self-respect. And it's awfully hard to win self-respect by talking to people who do not respect us on their terms rather than ours. The uncontrollable facets of human nonverbal communication have a way of making self-hatred reek for miles in every direction, and it seems unlikely that the stench will inspire respect.

Dale Carpenter, also writing at Independent Gay Forum, seems to understand this much better, or at least much more the way I do, in a piece in which he discusses the the no-on-8 campaign—a remarkable example of superficially strategic genius implicitly acknowledging oppressive values by scrubbing all actual gay people out of any advertising material.

Independent Gay Forum - Know on 8

Second, there's a paradox here. Almost everyone agrees that victory in the gay-marriage struggle ultimately requires the deep cultural shift brought on by coming out, by acquainting Americans with the real problems faced by real gay families, and by showing them how gay people are no threat to their own churches, families, and values.
We aren't going to fool people into supporting gay marriage. We can't just coldly claim legal rights. We may persuade gay activists that it's "wrong" and "unfair" to eliminate rights created five minutes ago by four judges. But most people don't believe there's a right to something that's not right. And they need to know a lot more about gay families over a long period to reach the conclusion that gay marriage is right.

How can that be done without talking about actual gay people? And when will it be done on a large scale except when we have the resources and energy to do it, as we do in a ballot fight? Winning in the end may depend on losing a few preliminary rounds in a way that progressively erodes the opposition. Instead, in every single ballot fight in thirty states, we have squandered the opportunity to educate voters for the future. Losing smartly now means winning later; losing ignorantly just means endless losing.

Of course, I have no hard evidence on the political effect of sanitizing the gay rights movement. All I can say is that I'm from a place where the movement has been largely successful and where it has been more often open and confrontational than sanitized and tactful. What works in Holland need not work in the United States. But intuitively, I have a very hard time picturing a story wherein an oppressed minority wins popular respect by hiding its own self-respect.

My op-ed in The Broadside

2009-02-11T22:32:00.001-05:00

I wrote an op-ed for GMU's campus paper The Broadside. It was about Don't Ask Don't Tell and it was called Don't Waffle, Don't Backpedal. For the benefit of anyone who reads that article and is sufficiently surprised about an incorrect factual assertion to look up my name on the internet, I have here two corrections.

The military employs about 1%, not 10%, of the US workforce. (My bad.)
I am not an American citizen. (I didn't write that; the editor did.)

Here's the original, unedited piece:

Don't Waffle, Don't Backpedal

Jaap Weel

When Stephen Benjamin joined the US Navy, he knew what he was getting into. Every day of his life as a naval officer he would live in fear of becoming a former naval officer.

Of course everyone in the military fears a premature end to their career. Those who join the military sign up for a risky job voluntarily, they know it, and they are well rewarded for it.

But it was not foreign enemies with bullets or shells that threatened to end Stephen's military career as an Arabic translator. It was his superiors, wielding a policy known as “Don't Ask, Don't Tell.”

In 2007, Stephen Benjamin was dismissed after inspectors digging through a chat system had found references to the fact that he had committed that gravest of military errors—he had told.

No, he had not told a lie, or an official secret. He had told someone that he is gay.

Never mind that Arabic translators are in short supply, and crucial to the already disappointingly difficult war effort in Iraq. Never mind, either, that military personnel are in short supply generally: so short, indeed, that recruiters have been told to relax all sorts of rules, ranging from age restrictions to criminal record checks.

~~The armed forces employ nearly one tenth of the entire US work force. As of today, that means that one in ten jobs is inaccessible to anyone who is gay, lesbian, or bisexual.~~ The US armed forces employ over a million people. As of today, that means that over a million jobs are inaccessible to anyone who is gay, lesbian, or bisexual.

Don't Ask, Don't Tell sounded like a sensible compromise when Bill Clinton signed off on it. No longer would the military make a big deal out of the sexual orientation of its personnel. Nobody would ask about it, nobody would talk about it, and all would be quiet.

Reality turned out different. In practice, Don't Ask Don't Tell means that “an honest statement of one's sexual orientation to anyone, anywhere, anytime is grounds for discharge,” as the Service Members' Legal Defense Network puts it. Gay service members need to make a choice: live a lie, or leave the service.

A majority of the American public want service members to be judged based on the sacrifices they make to serve their country. Americans support a military that selects its employees by their qualifications and their commitment, especially in times of war.

With so much disapproval of current policy, there is only one way to sustain it: refer the issue to further investigations, to be carried out at a leisurely pace interrupted only by the occasional interdepartmental memo. Which is—you guessed it—precisely what the Obama administration has seen fit to do.

The matter has, of course, been investigated at length before. Many times, in fact. Other industrialized democracies, including Israel, have realized long ago that there are more important qualifications for a military job than the gender of your significant other.

In politics, “further study” is less often conducted to study matters further than to make sure that matters stay in the process of being studied further, preferably indefinitely. While matters are being studied further, nobody asks questions, and nobody tells the answers. At least, that is the idea.

The principles of freedom and equality before the law that have made America a place where people can prosper in peace are precisely what the US military is supposed to defend. It is not more than reasonable to demand that the military start its fight at home by guaranteeing to its employees that they are free to choose their romantic partners and that they will be treated equally by a defense force that selects and promotes its members based on their ability to help defend the country.

Political appointments as emancipation

2009-02-05T11:11:00.001-05:00

Iceland has had an interesting taste in prime ministers for a long time. For many decades, they had the strange habit of replacing Foo Barson by his opponent Baz Quuxson, only to revert in the next election to erstwhile loser Foo Barson, or, if Foo Barson was at that point dead or at least senile, his son, Bar Fooson. (The Icelandic spurn newfangled surnames in favor of the old-fashioned convention of patronyms ending in -son and -dottir, preferably with a few ðs and þs thrown in for good runic measure.)

Moreover, Iceland was, to the best of my knowledge, the first to allow the lamentably underrepresented world minority of California Institute of Technology graduates to break the glass ceiling on the road to high office. And they had the first female president.

The recent report that the person to take the unenviable job of salvaging what is left of the Nordic outpost in the Atlantic after the collapse of its banking system is a lesbian set me to thinking a bit. Just how important a goal should appointment to public office be as a goal for an emancipatory movement?

I don't want to deny that it's a decent indicator of success. When people think about Margaret Thatcher, the first thing to come to mind is more likely "firebrand conservative reformer and union-buster" than "woman." And once we get over the initial phase of TV commentators asking every 5 minutes just how "historic" the election of Barack Obama—in fact, it seems like we've gotten there by now—people are most likely to associate him with this, that or the other policy or bit of political intrigue than with "black."

The ability of a culture to conceive of some powerful person as a powerful person rather than as a member of some oppressed minority seems more consequence than cause of a successful emancipation. It is probably more useful to consolidate a turn of public opinion that has already happened, lend it the official stamp of approval that pulls some of the remaining skeptics over the line, or close enough to it that at least they can be shamed into hiding their bigotry.

Of course, one can always claim that ceteris paribus, a few political appointments never hurt anyone. But in the real world, all the cetera aren't always para and there do exist tradeoffs.

Which brings me inevitably back to the US gay rights movement. I have heard awfully little criticism of the fact that the new US president still hasn't repealed Don't Ask Don't Tell—and don't tell me that it cannot be done at the stroke of a pen, that it requires interdepartmental commissions and task forces and consultations, all things considered, in the fullness of time, taking everything into consideration, at the appropriate moment, when the obstacles technical, administrative, and legal have been overcome. At the same time the HRC seems awfully happy that some of its friends have received political appointments. Not even high-profile ones that might conceivably have at least some symbolic cultural influence, as far as I can tell.

Consolation from history

2009-01-29T13:02:00.001-05:00

I don't have much time to do any writing of my own at the moment, but here is Brad DeLong eloquently reminding us of the wonders of long term economic growth (HT: Will Wilkinson):

Our economic appetite - THE WEEK

Nearly eighty years ago, John Maynard Keynes did the math on economic growth and concluded that within a few generations—by the time his peers' great-grandchildren came of age in, say, the 2000's—the persistent economic problem of too-scarce resources and too-few goods would no longer bedevil a substantial portion of humanity. He was right—even in the midst of our current hard times, he is right.

Parkinson's law

2009-01-18T13:37:00.001-05:00

Parkinson's Law by C. Northcote Parkinson was meant to be a work of satire, parodying both the inefficiency of public administration and the faux precision of social scientific models. But Parkinson's genius continues to feed actual research. Take this one, in the latest issue of Kyklos:

Service Quality in Modern Bureaucracy: Parkinson's Theory at Work
Beate Jochimsen

Parkinson's Law states that work expands to fill the time available for its completion and that the number of administrators in an office is bound to increase over time. A unique laboratory to test Parkinson's ideas are vehicle registration offices in Germany. Using their data we found empirical support for Parkinson's Law: First, service quality is no better in offices that have more staff per case. Second, service quality is worse if the service procedure is disaggregated into multiple smaller sub-services. Third, the staff size is a convex function of the number of customers. These results are robust to specifications in various alternative models.

Google good, FDA mweh...

2009-01-12T17:50:00.001-05:00

The patients in this story survived, but only barely.

Santa Cruz doctor helps save lives of family who ate poisonous mushrooms

... Two years ago in January at Dominican Hospital, Mitchell and Dr. Wendy Knapp treated a family of six who had eaten tacos made of death cap mushrooms they picked at Wilder Ranch State Park.

Searching Google, Mitchell found a treatment used in Europe, an intravenous milk thistle preparation called Legalon-Sil. He had persuaded the Food and Drug Administration to allow its use as an emergency investigational new drug. He arranged for an air courier to deliver the medication to the San Francisco hospital where four of the six patients had been taken after developing liver failure and needed transplants. ...

Google clearly comes out as the good guy. The FDA... well, they caved in at the right moments, but I think the story serves as a good reminder of the pros and cons of drug regulation. Regulators can prevent medications from doing harm, but they can also prevent them from doing good. How many cases of mushroom poisoning occur that are not treated by a doctor as gung-ho as this one?

Win for the participatory panopticon

2009-01-10T19:38:00.001-05:00

Custodiens ipsos custodies...

I have read that police tried to confiscate cell phones with possibly incriminating evidence against them after this shooting. Oughtn't there to be a law against that?

Anyway, there was still evidence from security cameras, and some cell phone video after all.

Quality reporting

2009-01-10T12:05:00.001-05:00

Radio Netherlands has a headline that reads (my translation): "tourists nearly as polluting as industry." The main article states something very different. Dutch tourists emit as much CO2 annually as the entire chemical industry. But who cares. There's a scary press release, and they report on it. What matters is the fact that some halfway prominent organization said something scary, not precisely what they said. Right?

A journalist to watch in 2009...

2009-01-08T14:05:00.001-05:00

...James Kirchick. This guy appears refreshingly able to think outside political boxes. Read his damning but still fairly measured criticism of Sean Penn's hypocritical politics in The Advocate:

It’s not surprising that Sean Penn, thanks to his star turn as Harvey Milk, is becoming a hero of the gay community—likely to be showered with acting prizes, and deservedly so. But his outspoken admiration for the Castro and Chavez regimes should make everyone think twice.

History and the politicians

2009-01-07T12:04:00.001-05:00

Bryan Caplan writes:

Part of the reason, no doubt, is that excessive attachment is the enemy of good history. Ever notice how wretched historical surveys become in their closing chapter on "recent events"?

In a sense, there is no reason to study economic history. All these quaint historical conditions that are so hard to study for lack of data are actually pretty similar to present conditions in different parts of the world. But what's going on in the world today is so, well, so... political.

You can write a paper about the decline of the Mongol empire and everybody, except for half a dozen opinionated experts and a few Mongolian nationalists (who live far away and don't get much listened to) will respond will a heartfelt "hmm." But say something about the causes of economic collapse in Zimbabwe and you're taking a political stance in a messy, messy debate about race relations.

This is not to say that history isn't political—sometimes I wonder if history is much besides politics. But the contemporary political relevance of historical events is, in most cases, an invention far more recent than the historical events themselves, and therefore it can be re-invented if necessary.

Eric the half a civil servant

2009-01-06T11:56:00.001-05:00

The Dutch civil service makes a distinction between the status of ambtenaar (civil servant) and semiambtenaar (semi-civil servant). The latter are not distinguished, as their name might suggest, by a 50% reduction in civility, but rather by the fact that their employer is independent from the government in a very particular, if murky, sense that I shan't bore the reader with. Suffice it to say that Dutch financial institutions offer personal loans to civil servants at a special discount, presumably because their job security reduces the risk of their defaulting. And sure enough, the lenders make no distinction between making a loan to a full civil servant or to half a civil servant.

Why there's hope for Britain

2009-01-05T21:16:00.001-05:00

“...and that's why public choice economics, which explains why all this was going on, was at the root of almost every episode of Yes Minister and Yes, Prime Minister.” (Sir Anthony Jay)

Blog stats

2009-01-05T18:30:00.001-05:00

It's been a while since I last checked the Google Analytics page for this blog.

And the most visited page on the blog is...

...the recipes for speculaaskruiden!

Queen James, Queen William, and the boys

2009-01-04T14:03:00.002-05:00

Reading about early modern European history, it strikes me just how many kings are rumored to be gay. Even in modern times in tolerant urban communities, I doubt more than about 1 in 20 men choose to pursue same-sex relations (Kinsey’s 10% figure is now widely acknowledged to be overestimated). And yet, in an era where religiously motivated sexual taboos were still virtually unquestioned, “ce savant roy d’Angleterre / Foutoit-il pas le Boukinquan?” (Pardon my French, I'm quoting a famous poet!)

What explains the prevalence of same-sex relations in royal bedchambers if they were so thoroughly taboo elsewhere?

You could blame the rumors on revisionist historians engaged in "queering hegemony" or some such dadaistic experiment, but most of the time the rumors seem to predate the contemporary gay rights movement.

I have a different explanation, based on the idea that the rumors are true.

Let’s engage in some economic imperialism and consider the market for royal romance using the hypothesis that sexual preferences are not absolute, but that indeed demand curves slope down.

People, kings and otherwise, have preferences, sometimes strong, sometimes weak, for relations with persons of one or the other sex. But it is not those preferences alone that will determine who they will actually end up in bed with. Rather, it is the interaction between their preferences and the cost of indulging them.

I will argue that for a king who enjoyed male relations, even if he actually preferred women, the attendant problematic consequences (“costs”) of male relations may have been sufficiently lower that he ended up en foutant le Boukinquan.

The first observation to be made is that demand for sexual intimacy with the king must have been fierce. For one thing, any expedient to get close to the king, figuratively or literally, can readily be explained as a means to improve a person’s social and political standing. And for another: power is, at least for some people, sexually attractive in and of itself. As a contemporary case in point, consider the excuse that Italian model, pop singer and all-around sex icon Carla Bruni gave to the Times of London for marrying the rather plain, middle-aged president of France: “I want a man with nuclear power.”

In other words, an age-of-enlightenment king had a lot of options, and it is unlikely that the problems of search and matching that plague other mating markets were much of a problem for a king. Nor were the strictures of bourgeois morality a major impediment for these exalted aristocrats, as is evidenced by their fondness for that most bourgeois of sins: waste. (In Dutch, I may add, the language of that most prototypically bourgeois of nations, waste and sin are both zonde.)

No, constraints on royal intimacy had little to do with Searching for the One or hiding from the prying eyes of neighbors. But kings, of course, had their own problems. For one thing, they were typically married. And royal marriage was an act of diplomacy, not love. Therein may lie the key to the puzzle of the gay kings.

Under the complex politics of inheritance and succession, a king’s relationships with a woman other than his wife had as many political implications as royal marriages themselves. In some cases, depending on the religious and legal environment, a king could divorce and remarry, and even if he didn’t, royal bastards frequently made claims to the throne. (Birth control measures are, I believe, commonly assumed to have been comparatively ineffective at the time.) For that reason, even a queen that had long given up, or never gotten started, on actually having a loving relationship with her spouse, could have all sorts of political reasons to resent mistresses. Not to mention the queen’s relatives, who were often the rulers of some foreign country that had married her of in the hope of gaining some influence or currying some favor.

Extramarital same-sex relationships, notwithstanding the general moral disapproval of any same-sex relations at the time, would not have carried the same political implications. There was no option, under the marriage laws of the time, for James I to marry the Duke of Buckingham and provoke a diplomatic crisis with his first wife's relatives at the Danish court. Nor could the secret passageway linking James and George’s bedchambers at Apethorpe, whatever the activities may be that it facilitated, have led to the production of any bastard offspring.

From a political perspective, a royal boyfriend was no more threatening than any other particularly trusted courtier: that is, threatening for sure, but not in anywhere near the same way as a mistress.

Political economy cannot explain why James I was so infatuated with Villiers, or William III with Keppel. But it may well explain how kings, and high aristocrats in general, even those who had only mildly bisexual predilections, may have faced very different incentives, compared to the prudish burghers in their realms, to cultivate and indulge one variety of romantic preference over another.

I welcome comments, as usual, but especially from anyone who actually knows something about the customs at early modern European courts. The story's only half in jest. Feel free to prove me wrong.

ScribeFire a great success

2009-01-04T12:45:00.001-05:00

ScribeFire is great. It really makes it possible for me to have the idea for a blog post, push a button, start typing, click publish, and be done. And only now did I realize how much the sluggish interface at blogger added to the effort involved in blogging.

Killer apps

2009-01-04T12:43:00.001-05:00

I remember when I was in high school, in the early days of the World Wide Web, one of the first types collaborative interactive web sites that caught on here in The Netherlands was... repositories of summaries of Famous Works of Dutch, English, French, and German Literature, particularly those typically read for the comprehensive oral exams on literature we had to pass at the end of high school. Some say the real killer app for all computer technology is porn, others say it's violence, but I say it's cheating on homework ;-)

Capitalism, Logic, and the Historians

2009-01-02T22:15:00.002-05:00

From Margaret Jacob's review of The Familial State: Ruling Families and Merchant Capitalism in Early Modern Europe by Julia Adams in 79 J Modern History 884. (Why am I reading this? Vacation. I'm bored. ;-) )

Once the male head of an early modern family acquired wealth and translated it into state offices, he fought to keep that patrimony for sons, nephews, and generations yet unborn and only imagined. When the state possessed insufficient offices to be sold—as was the case, for instance, in the relatively impoverished English state—it created a situation in which elites wedded to commerce, or rallied by religious zeal, could pose great risks. While appreciative of the Marxist contribution to our understanding of the English revolutions, Adams argues that "political, economic, familial, and religious motivations or logics of action were typically fused in patrimonial politics; they were not independent causal streams" (135). Class by itself just cannot do the work of explaining the complexity of social relations found in these societies. Similarly, only masculine power—the sinews of birth, blood, and sexual identity—can explain how the Dutch elite, once constituted as an oligarchy in firm control of its urban localities, what Adams calls its entrenched and "fractious localism," held on to power so firmly that only the French invasion of 1795 was capable of restructuring the constitution in ways that gave the United Provinces a chance at modern statehood, arguably saving the nation from a level of corruption and ossification that would have spelled disaster in the coming century.

Only masculine power can explain how the Dutch elite was able to to hold onto power. That means, in my white male logic, that masculine power explains how a particular group of men was able to hold onto power, and that nothing else can explain this. Let's look at the first claim first. If it were claimed that masculine power explains how men were able to hold on to power, we would have to respond that the statement reveals a lamentably low standard of what constitutes an explanation—but we are saved by the specification that masculine power allowed this particular group of men to hold onto power. This theory sounds attractive until we realize that there is nothing about masculine power that makes it particularly likely to sustain this group's power and not that of the thousands of other male-dominated elites that did not make it, or indeed of the Dutch elite after 1795. The theory doth explain too much. It can "explain" how just about every other elite in history has held onto power, or would have held onto power even in cases where it actually did not. Hence, the theory fails the first criterion for a scientific theory, which is that it should have some nontrivial consequences.

So much for the power of the observation that the elites were male to explain why they were successful, and, somehow, why the ceased to be successful in 1795 rather than earlier or later. On to the claim that nothing else can explain how they held onto power. I sure hope not. If that is so, given that the male-power-causes-male-power theory is rather flimsy, we would be left with nothing at all to explain a phenomenon that is actually, by all appearances, crucial to the rise of the Western world and by extension to world history.

I would like to think that in order to do economic history well as an economist, you have to be open to the perspectives of non-economist historians on economic history. But whenever I try to be open in this way, I run into statements that are indisputably about economics, often written by people at prestigious universities, that make absolutely no sense. In fact, it is not infrequent that I run into theories that are simply logically incoherent, or I see opposite effects explained by the same cause (disease-induced labor shortages caused relaxation of serfdom in France and tightening in Russia!)

This is not to say that there isn't good history, of course, nor that there are not professional standards in history. Obviously, there are professional standards in history. But they appear to have more to do with your ability to find surprising facts in obscure archives than with the coherence of your theories.

On an optimistic note: I am reading Doug North and it's delicious.

Maximize that!

2009-01-02T18:54:00.001-05:00

The standard Chicagoesque rebuke to exaggerated claims from behavioral economists, which Deirdre McCloskey calls, I believe, “The American Question,” is to say: “If you’re so smart, why aren’t you rich?” After all, if people consistently behave irrationally in ways that are exploitable for profit by others, you would expect those others to go and do some exploiting.

It seems that with swoopo.com, someone has taken up that challenge with a business model that specifically exploits a well known behavioral “anomaly” known as the endowment effect. The endowment effect occurs when a person lets their decisions be influenced by sunk cost. Maximizing economic agents don't cry over spilled milk, let bygones be bygones, cut their losses, and move on. The mere fact that you’ve already spent some amount of money on the “same” item should not influence whether you are willing to spend more yet to keep it—Max U compares marginal cost to marginal benefit and that’s that.

The way in which swoopo.com exploits this is ingenious. They have an auction where each bid increases the current price by a penny (or 15 cents), it costs 75 cents to place a bid, and each bid extends the auction by 15 seconds. Think about it. Or read Jeff Atwood's article. (HT: Tyler Cowen.)

Coding Horror: Profitable Until Deemed Illegal
I was fascinated to discover the auction hybrid site swoopo.com (previously known as telebid.com). It's a strange combination of eBay, woot, and slot machine. Here's how it works:
You purchase bids in pre-packaged blocks of at least 30. Each bid costs you 75 cents, with no volume discount.
Each bid raises the purchase price by 15 cents and increases the auction time by 15 seconds.
Once the auction ends, you pay the final price.

What to make of this? As far as I can tell, it is strictly speaking not fraud, because the rules are clearly explained, and not gambling, because the procedure is entirely deterministic. Simple signals that one might use to judge whether the whole thing is legit have been taken care of: the graphic design of their web site is fairly professional-looking (which signals an intention of the company to stay around for a while) and the presence of lots of other bidders, which signals that apparently there's something good to be had here, is carefully directed toward those auctions that make the front page.

There is a tendency to want to regulate these sorts of not exactly fraudulent but uncomfortable business practices away. I am not sure that this is a good thing. I fear that you can never succeed in going all the way, while at the same time an environment in which you can assume that everything that looks legit actually is will encourage people to be less cautious, which will prompt demand for more regulation of ever more doubtfully shady business practices.

Who says that French political writing is boring and pompous?

2009-01-02T15:09:00.001-05:00

Who says that French political writing is boring and pompous?

«Il y avait en Westphalie, dans le château de monsieur le baron de Thunder-ten-tronckh, un jeune garçon à qui la nature avait donné les mœurs les plus douces. ... Monsieur le baron était un des plus puissants seigneurs de la Westphalie, car son château avait une porte et des fenêtres.» (Voltaire, Candide)

A paradigmatic parable

2009-01-02T15:06:00.001-05:00

Suppose we have a set of social phenomena described in a grand Encyclopedia of Social Science, ranging from Actuarially Correct Life Insurance to Zulu Creation Myths. Professor Jones has devised a simple and elegant theory that accurately describes all the phenomena from A through X, but utterly fails on Y and Z. Professor Smith has an equally simple and equally elegant theory that accurately describes all the phenomena from C through Z, but utterly fail to account for the phenomena described in volumes A and B.

Professors Jones and Smith are both convinced that they have found the One True Theory, and that their competitors' theories are fatally struck by the anomalies in, respectively, the first and the last two volumes of the Encyclopedia. On the occasion of a Festschrift for Professor Doktor von Thunder-ten-Tronck zu Thunder-ten-Tronck, their common Doktorvater, professors Smith and Jones invite professor Popper, the famous philosopher of science, to apply his theory of Critical Rationalism to their competing theories.

Poor old Sir Karl decides that there is no way to arbitrate between Smith and Jones, except for the creation of an even grander, even more elegant theory encompassing both of theirs. He picks up a copy of The Structure of Scientific Revolutions, re-reads it, realizes that it does not actually contradict his Logic of Scientific Discovery in the least as long as you think of competing paradigms as being like Smith and Jones, explains his findings to professor Kuhn, and saves the world a lot of meaningless "debate."

Douglass North talk

2009-01-01T09:38:00.004-05:00

I'm going to take a class on European Economic History next semester, and I'm pretty excited about it, and spending some time reading in preparation for it. You cannot take an interest in Economic History without tripping over Douglass North. Despite the renewed fashion for demographic determinism, my best hunch is that North is onto something. Watch for yourself.

The Beeb and YouTube

2009-01-01T09:07:00.001-05:00

It has been reported that the BBC has been warming up to internet distribution of its content and has even set up its own YouTube channel. But what strikes me even more is that it appears to be silently condoning the distribution of some of its stuff by others. YouTube is full of British sitcom: Blackadder, 'Allo 'allo, Absolutely Fabulous, Keeping up Appearances, they're all there. It's neatly categorized by season and episode and not exactly well hidden. The Beeb can't possibly be policing this very strictly... right? I wonder if there is a design in it. These things are still rebroadcast, not just by the BBC itself but also by many TV channels outside Britain. Besides, there's DVD sales. But maybe, just maybe, exposing a larger audience to the joys of René Artois and Hyacinth Bucket (it's boo-kay!) increases DVD sales. Or just doesn't influence it much at all. Frankly, I have no idea why the BBC is so apparently lax about its copyrights.

Are people wise or just lucky?

2008-12-29T16:31:00.001-05:00

From Deirdre McCloskey’s “The good old Coase theorem and the good old Chicago school: a comment on Zerbe and Medema.”

It must have been around 1968, in the coffee room of the Social Science Building at Chicago, that Friedman and Stigler had a jovial and public conversation about being economists. It made a big impression me. Milton was lamenting the stupidity of tariffs, to which george broke in, from a foot above, saying something like this: “Milton, you’re such a preacher! If people want free trade they'll get it. If they don't want it, no amount of jaw-boning by economists will change their minds.” “Ah: that's where we differ, George. We admire markets, but you think they’ve already worked.” “And why not? People are self-interested, voting their pocketbooks—that's enough to make the market work. The people bought the tariffs; tariffs must be what they want.” “No: they pursue their interests but often do not know what the interests are. People need education. The average citizen has no idea that a tariff hurts him.” “Education! Try educating a lobbyist for the textile industry.” “As I said, that’s where we differ: I’m a teacher, and think that people do some things because they are ignorant.” “And I’m a scientist, an economic scientist: people do what they do because they are wise.”

Separation of Church and State

2008-12-29T11:03:00.001-05:00

In the previous post, I wrote that I am pretty tolerant of retaining harmless and possibly useful elements of religious traditions even as I continue to oppose those traditions' general dogmatism and specific odious doctrines. I want to add something to that that is important enough to warrant a separate post.

When it comes to state endorsement of religion, I think there is a very good reason to be fastidiously careful. The foundation of the open society that makes it possible for culturally diverse communities to prosper is that all but the most necessary features of culture should not receive the backing of an organized monopoly on violence. We can debate till the cows come home where precisely to draw the line between those institutions that must be uniformly enforced if peace and prosperity are to be maintained and those that are merely incidental features of one cultural community or another. But the details of religious doctrine are clearly on the wrong side of that line.

This means that I am actually quite radical on a lot of traditional church-state issues. References to God still abound in the judicial procedure, coinage, and ceremonial features of many Western nations, and many otherwise civilized countries still have established churches. Ironically, the US, that most religious of Western nations, has worked the harder than most to eliminate religion from government and policy—though by no means from the politics that underlies it. And I think that this is a good thing and has helped the US be better able to integrate a diversity of ethnic and religious groups than many European countries.

Atheists for Christmas

2008-12-29T10:49:00.001-05:00

David Kopel on The Volokh Conspiracy writes:

Therefore, it is my recommendation that you use the phrase "Happy holidays" if and only if you are speaking to an ultra-sensitive African-American atheist.

I, for one, am an atheist and I am not offended if you say "Merry Christmas."

In the Frankish lands where I grew up, not far from the grave of old Charlemagne, Christmas far predates Christianity as such. Maybe it would be better if we called it by a pre-Christian name like Yule, as indeed Scandinavians still do. But I have no particular reason to dislike the fact that in
traditionally Christian cultures many that have long given up on Christian dogma still do celebrate the winter solstice
with family gatherings and greeting cards and the indoor display of decorated pine trees.

Look. I do happen to think that the doctrines of Christianity roughly break down into those that can be found in other religions, too, and those that are so ridiculous that only a millennium or two of politicking could have commended them to Western civilization. But the idea that it having regularly scheduled days of celebration that provide an excuse for family gatherings is in the former category, not the latter.

And anyway, I do not subscribe to the French version of Enlightenmentism that says that all social customs and traditions ought to be, in all their details, the product of rational, conscious design. They never have been and cannot be, which makes the question whether they should be rather uninteresting.

Every culture has its traditions. Tradition is not the enemy. Particular traditions can be harmful or helpful, but the presence of some tradition or another is a given. That is why I dislike the tradition of teaching the infallible supremacy of the papacy, and the tradition of teaching that faith is a virtue more sublime than reason, and many other traditions that I was exposed to in Catholic elementary schools—but I have no particular truck with Christmas.

ScribeFire blogging plugin

2008-12-29T10:17:00.001-05:00

I write this post using the ScribeFire blogging plugin. If that works out, I'll be using that more often. It sure seems to be a lot faster than the fairly clumsy web interface that blogspot provides.

Post office polka

2008-12-27T08:42:00.002-05:00

In the concluding section of a recent paper, "The Distribution of Post Offices in Italy and the United States," the authors speculate about the effects of postal liberalization on post office distribution. As it happens, the Dutch post office was sold and the restrictions on competition slowly removed, and indeed the pattern of retail outlets has changed. But in a slightly different way than they predict...

The postal bank, the phone company, and the post office, which used to all be part of the same organization, were separated. The post office, after merging with an Australian package delivery firm, became TNT N.V., the phone company, Royal KPN N.V., and the postal bank became part of ING Group N.V. The phone company had been developing its own retail chain (Primafoon) and no longer has anything to do with the post offices, which are run by Postkantoren B.V., a joint venture of TNT and ING.

Postkantoren soon figured out that it was much cheaper to have a postal service window inside an existing and independently profitable retail store. They opened 1300 postal "service points" inside grocery stores, and all but a rapidly shrinking 250 of their 800 "post offices" are themselves inside stores, too: typically branches of a chain of stationers (Bruna) that Postkantoren has bought.

Another development is that ING has been pulling out of the whole thing, preferring to offer all but the simplest banking services from their existing ING-branded chain of storefronts instead of through the post offices. I would not be surprised if they were to cut loose from Postkantoren altogether at some point.

Finally, Postkantoren has become a contractor offering minor administrative operations to business and government. They do ID verification for transit passes, and car registration, and the like, they pay out sweepstakes prices and rebates for merchants, sell concert tickets, and handle various municipal services for local governments. Some of these are tasks that had been handled by the government post office in the past, but it is clear that Postkantoren is aggressively, and somewhat successfully, expanding their contracting business.

In short, it is not just the geographic distribution of post offices that has changed. It is also the definition of a "post office".

Even the tiniest coin

2008-12-26T13:17:00.004-05:00

"Descriptive" historical writing, especially about economic history, often has more theory hidden in it than meets the eye. Take this sentence near the beginning of Crouzet's History of the European Economy: "Specie circulation was restricted to the elite and to estate traders, who were selling high-value products. Transactions of lesser folk were too small for the use of even the tiniest coin: payments to their lords were in kind or labor, and barter deals were frequent." It's there near the end of a paragraph, immediately to be followed by a change of topic to invasions in the east of the Holy Roman Empire. But there is an interesting theory there—and a theory, moreover, which seems intuitively implausible, or at least incomplete to me.

The theory says that the reason that medieval lords taxed their serfs in kind rather than in specie is that there was not enough specie to go around. More in particular, the theory counters the obvious objection that one could mint smaller coins, or cut them in pieces like Spanish gold dollars were later cut in eight. It states that for a given supply of silver, you can end up with transactions of which the value in silver is simply too small to coin.

It's a clever theory, but it appears to me too clever by half. Silver is not the only feasible money. The Romans, in fact, had copper money as well as silver and gold. Adam Smith reports iron nails being used as money in a village in Scotland, and anthropologist have examples of shells, gemstones, salt, and various other portable, nonperishable, and reliably scarce commodities being used as money. Was there no such commodity available in Europe? If there was, what prevented people from using it?

In general, "culture" does not count as an explanation. To say that in European culture salt was not used as money is to restate the problem, not to solve it. Cultures sometimes change and sometimes don't, and the interesting question is when they do and when they don't and why.

We can easily think of an obvious solution to the problem of transactions "too small to coin": use a cheaper metal. Moreover, we can adduce examples, from cultures both Western and non-Western, primitive and sophisticated, of enormous creativity in finding commodities to use as money. That means that any real explanation of the in-kind nature of early Medieval trade and taxation must involve some sort of credible obstacle to that sort of innovation.

My feeling is that the lack of bullion, or its "hoarding," is a red herring. The real reason for in-kind taxation and the breakdown of trade, which of course are indeed related to each other, probably has more to do with issues of political economy. About which more later.

Undergrad economics education

2008-12-17T19:21:00.004-05:00

In the United States, economists very often get a chance to convey to impressionable undergrads something about their interesting, important, counter-intuitive, and inherently politically incorrect field of research.

We typically do not engage in the kind of academic politics of trying to get our courses into the general ed requirements, so that unlike the race-class-and-gender crowd, we cannot rely on a captive audience. This is made up for in part, though, by the common, if questionable, presumption that knowledge of economics is useful in business.

The upshot, anyhow, is that we still end up with masses of undergrads to educate in a field where there is very little in the way of outside constraints on what gets taught.

In the light of those facts, it is kind of sad that, to the best of my knowledge, the teaching of undergrad economics classes is not based on any sort of systematic evaluation of what teaching methods are most likely to convey economic principles in a way that makes them actually stick. Rather, we blindly follow the example set by Samuelson's 1948 Economics or, for those of us who are by education or by temperament more "Chicago" than "MIT," Stigler's 1942 The Theory of Price. Of course there are many other texts in use, but most of them follow one or the other of these archetypes pretty closely.

Bryan Caplan has a good section in The Myth of the Rational Voter on economics education, where he states that "you must respect the Laffer Curve of learning: They retain less if you try to teach them more." That sounds like a good starting point to me. If I look at a textbook like Perloff, with dozens of little chapters each treating a model even more obscure than the previous, and asking questions that bear no relation to real economic phenomena but are always framed in terms of the models themselves, my gut feeling is that I see a textbook that is way on the wrong side of the educational Laffer Curve. It may be a good reference for the student or two that goes on to do a PhD, but I could easily see it leading to people who cram the whole thing into their heads for the test and who would be at a loss to explain comparative advantage to you a year later.

But there is more to good teaching than restraint in curriculum expansion. If we look at fields where the fruits of education are directly applicable to students' careers and there is considerable outside pressure on universities to really teach certain facts and skills, what we typically see is a strong emphasize on practice. Whether it is civil engineering or law, the realization in professionally oriented education is typically that what is taught is a skill at least as much as a body of knowledge, and the only way to develop the procedural memory of how to perform that skill is to drill. The way I learned physics at Caltech was by doing physics and applied math problems, oftentimes 40 hours a week, until my neurons grew in the sorts of patterns that are required to do that sort of thing effectively. Economic thinking, similarly, is a skill, but I have the distinct impression that it is implicitly thought of by many educators as a body of factual knowledge instead.

My current best guess at what a good undergrad economics syllabus would look like is something like this. First of all, there would be a very strong emphasis on a finite number of basic principles, all of them understandable and understood in words as well as graphs and math: gains from trade, comparative advantage, equalization at the margin, rents accruing to fixed resources, and so on. Secondly, there would be a lot of practice applying the simple principles, questions of the sort "the price of cheese falls compared to milk; have interest rates gone up or down?" or "why is French wine in the US typically better than French wine in France?"

Some might say that this approach is pretty close to the old Chicago approach, or indeed to that of my own Micro Theory I instructor (Walter Williams). But I would counter that in order to take that approach beyond the glass ceiling of elite institutions and charismatic individual instructors, it is necessary to transform it into the sort of ready-made workbooks and test banks that the Samuelsonian approach is already blessed with to a larger extent.

Most importantly, though, all of this is just a rough guess. Really what we need is some serious empirical work on what works and what doesn't.

The credit crunch

2008-12-17T11:06:00.003-05:00

There is a credit crunch!

Consumers can get no credit!

That is why poor foreign graduate students get credit card offers in the mail.

Right?

I know, I know, the plural of anecdote isn't data, and the singular sure isn't.

First semester as a PhD student

2008-12-16T09:00:00.004-05:00

I just finished my first semester at GMU. Having worked on my thesis for VU up until the last week before leaving for Fairfax, I had had next to no time to prepare for GMU and I went in on the basis of the two economics courses I ever took and a large helping of general mathematical maturity. I did just fine. If you have a science or engineering background and you are wondering whether you could keep up with the econ majors in a graduate program, my answer would be: give it a try. The gap between the average undergrad program in economics and what you learn in grad school is much bigger than the equivalent gaps in the "hard" sciences. That makes economics a relatively easy field to get in on at a later stage. I don't want to say grad school starts from scratch, but a little preparation goes a long way.

...and the other nineteen do not have to be published...

2008-12-10T10:45:00.003-05:00

"My impression is that the best and brightest in the profession proceed as if economics is the physics of society. There is a single universally valid model of the world. It only needs to be applied. You could drop a modern economist from a time machine—a helicopter, maybe, like the one that drops the money—at any time, in any place, along with his or her personal computer; he or she could set up in business without even bothering to ask what time and which place. In a little while, the up-to-date economist will have maximized a familiar-looking present-value integral, made a few familiar log-linear approximations, and run the obligatory familiar regression. The familiar coefficients will be poorly determined, but about one-twentieth of them will be significant at the 5 percent level, and the other nineteen do not have to be published. With a little judicious selection here and there, it will turn out that the data are just barely consistent with your thesis adviser's hypothesis that money is neutral (or nonneutral, take your choice) everywhere and always, modulo an information asymmetry, any old information asymmetry, don't worry, you'll think of one."

For those of my readers who are familiar with economics, I would challenge you to take a guess at who wrote that paragraph. I promise you it is not a usual Austrian, "post-autistic" or otherwise heterodox suspect, but someone who figures big time in your macro textbook, somewhere near the first chapter, most likely. Also, despite the definite McCloskeyan flavor of that last sentence, no, it's not McCloskey.

Speculaaskruiden

2008-11-28T01:35:00.003-05:00

According to various sources, speculaaskruiden always contain cinnamon, cloves, nutmeg (and/or mace), and white pepper (or allspice). I have grouped together nutmeg and mace because they are similar in taste; the same with white pepper and allspice. Dutch recipes sometimes mention mace as a milder substitute for cloves in the recipe, though. Recipes vary as to whether we are told to use ginger, coriander, anise, and cardamom. The amounts of each spice also vary by recipe. Typically, there is lots of cinnamon, less nutmeg/mace and cloves, and less yet of everything else. Cardamom and cinnamon, by the way, are commonly seen as substitutes, although in my opinion cardamom is much tastier.

Radio Netherlands (RNW) has 15g cinnamon, 2g cloves, 2g nutmeg, 1g white pepper, 1g ginger, 0.5g cardamom. about.com's Dutch Food section has 4tsp cinnamon, 1tsp cloves, 1tsp mace, 1/3tsp ginger, 1/5tsp white pepper, 1/5tsp cardamom, 1/5tsp coriander seed, 1/5tsp anise seed, 1/5tsp nutmeg. Trifles.nl has 6g cinnamon, 2g cloves, 2g nutmeg, 1g white pepper, 1g anise, 1g coriander. Witchin' in the Kitchin wants 30g cinnamon, 10g allspice, 10g ginger, 10g mace, 5g cloves, 5g coriander, 5g nutmeg, a pinch of cardamom, and anise to taste. Multiple sources mention that the amount of cloves determines the intensity of the mixture and/or the difference between "speculaaskruiden" and "koekkruiden."

All in all, nobody really knows what makes speculaaskruiden. Virtually everybody who uses them presumably lives in The Netherlands, where each grocery will carry one of about three brands of them, pre-mixed. In practice, that determines what speculaaskruiden really are. Unfortunately, the producers do not reveal the proportions they use. Maybe some day somebody with a chromatography machine and too much time on their hands can reverse engineer Silvo's offering and the world will finally know what Dutch Grocery Store Standard speculaaskruiden are like.

Valery Gergiev turns perfectly good Beethoven into Russian mush

2008-11-24T20:18:00.006-05:00

I grew up with the Limburg Symphony Orchestra. I harbor a certain nostalgia for the times I saw them play in the cozy intimacy of Maastricht's Theater aan het Vrijthof, but setting that aside, they frankly weren't much good. A typical performance of that orchestra would have you wait with great expectations for the moment when you got to chuckle at the brass section's first entrance into the piece—invariably a stuttery affair. A good concert was defined as one in which everything was in tune and on time.

My expectations having been permanently set on a provincially low level in my formative years, I was baffled when in later years I got to hear the LA Philharmonic and the Concertgebouw Orchestra. There simply are no technical flaws or trip-ups to speak of when these people play.

But that means that you get to judge their musical interpretation. When I heard Valery Gergiev conduct his Mariinsky (formerly Kirov) orchestra at George Mason University last week, I heard all the technical perfection I could have hoped for, but I was also hit by the impression that the 90s never happened in Russia.

To be sure, as the Washington Post reviewer correctly points out, the piano was awful, and awfully American in its awfulness. But that is no excuse for drowning some of the most sparkly sprinkly Beethovenly phrases that Beethoven ever wrote in oozing puddles of pedal, only to cut off the final note before the audience has so much as registered its being played. Nor is the fact that the orchestra is Russian any excuse for interpreting the intricately cerebral polyphony of the 4th piano concerto as if it were some sort of Wagnerian "tone picture."

My criticism of Gergiev's Beethoven can be summarized rather bluntly: I didn't hear all the notes. That may be perfectly appropriate when you are performing Prokofiev, as indeed he did before and after Beethoven's 4th. But it does not do justice to old Ludwig, who was the last of the Classics at least as much as he was the first of the Romantics, and who would never have written a note he didn't intend to be played and heard.

I do not intend to criticize a very able musician whose artistic vision obviously did not include the fetishistic "correctness" of the historical performance movement for failing to tune his violins to 430 Hz or playing a tad slower than Beethoven's breakneck metronome markings or anything like that. A musical score should be a basis for creating something beautiful and enjoyable, not a recipe that must be followed to the letter regardless of whether the result is edible.

But for crying out loud, it's 2008, and we expect a little more Glenn Gould and a little less elevator music.