Unmistakable Marks

Elmer's Programming

Mark Pilgrim, having solved a problem for himself by slapping together a bunch of freely available open-source tools, predicts:

In the future, there will be so much open source software available, programmers will be judged by how much they know about it and how well they can glue it together to build solutions.

In a lot of ways, that future is already here today. The combination of the open-source movement and the Internet has revolutionized (and I say that with no hyperbole) the process of software development. Whenever I encounter a problem these days, the first thing I do is look to see if somebody's already written open-source software to solve it. As often as not, they have. And if they haven't, then I can usually take existing open-source tools and piece them together into a solution with a minimum of coding; I'm just gluing together components into a larger structure. The completely custom software I use to publish this blog, for instance, has fewer than 500 lines of my own code in it.

This sort of thing, rather than any complex moral or economic rationales, is why I'm convinced that open-source is going to be even more of a success in the long term -- it's just too damn useful to go away.

Comments | March 31, 2003

Meet the Court

Most people -- even those who disagree with him -- say that Supreme Court Justice Scalia is a brilliant legal scholar. Not being much of a legal scholar myself, I'm unable to assess the truth of that claim. What's obvious to even a lay observer, though, is that Scalia is a complete prick. Consider this wonderful exchange, from a Scalia intent on keeping homosexuality illegal:

"It's conceded by the state of Texas that married couples can't be regulated in their private sexual decisions," says Smith. To which Scalia rejoins, "They may have conceded it, but I haven't."

The hell? Who would say that? I can't believe that even the staunchest Republican approves of that sort of federal regulation. But there's Justice Scalia, defender of the government's right to micromanage the sex lives of married couples as long as it keeps them gay folks down. If this is the best that right-wing jurisprudence can do, I submit that right-wing jurisprudence isn't all its cracked up to be.

Comments | March 27, 2003

Bubble? What Bubble?

I'm being slightly premature here, because final papers aren't signed, inspections aren't completed, and mortgages have not met with final approval, but: It looks like Anne and I are probably going to be homeowners.

This has been a slightly surreal last few days. A brief timeline of this house-buying experience:

Friday: House listed for sale.

Sunday: We saw house.

Monday: We decided to put offer on house, found out there was already another offer on it, and put a slightly better offer on house. Offer accepted in principle, with a few modifications that we need to go and sign to approve today.

The amazing thing is that this isn't amazing. The average time on the market for houses in this neighborhood (which is on the nice side of normal, but nothing magically special) is six days -- and they frequently sell for above the asking price. Unreal. It's like San Francisco in 1999, except without the jobs and the climate.

At any rate, now I'm reading up on home inspections, which is scaring the hell out of me. I had no idea there were so many things that could go wrong with houses. I need to worry about how they mounted the flushing on the chimney, now? (Well, no. The inspector does. This keeps me lightly sane.)

Comments | March 25, 2003

Food and its Discontents

Kevin Drum has realized with some amazement that butter on popcorn is better than margarine. I know the feeling. I grew up on faux foods and have only in the past few years discovered that the real thing usually tastes better. Mayonnaise is better than Miracle Whip brand salad dressing; whipped cream is better than Cool Whip brand dessert topping; butter is better than margarine; and macaroni and cheese made with actual cheese is arguably even better than that made with tastylicious Kraft brand orange powder.

So the question is: Why the heck do so many people use these fake foods in the first place? They're not all that much cheaper, they're not all that much more convenient, and they obviously aren't nearly as good. Was there some kind of anti-organic backlash before my time that conditioned people to seek out synthetic molecules? Or are we just seeing the power of advertising brand names?

Comments | March 24, 2003

Mondegreens of News

On the radio yesterday, I heard a report about the strain that the Iraq War has placed upon our alliances.

"A torn NATO..." the story began, "...touched down in Georgia."

Oh.

Comments | March 21, 2003

What He Said

I always have a strong temptation to point everyone I know in the direction of Paul Krugman's latest article, whatever it is; I'm kind enough to generally refrain. But when he starts talking about hard numbers, he's really at his best:

Accountants estimate the "actuarial balance" of Social Security and Medicare the same way a private insurance company would: they calculate the present value of projected revenues and outlays, and find the difference. (The present value of a future expense is the amount you would have to invest today to have the money when the bill comes due. For example, if $1 invested in U.S. government bonds would be worth $2 by the year 2020, then the present value of $2 in 2020 is $1 today.) And both programs face shortfalls: the estimated actuarial deficit of Social Security over the next 75 years is $3.5 trillion, and that of Medicare is $6.2 trillion.

But how do these shortfalls compare with the fiscal effects of recent and probable future tax cuts?

The new study, carried out by the Center on Budget and Policy Priorities, estimates the present value of the revenue that will be lost because of the Bush tax cuts . those that have already taken place, together with those that have been proposed . using the same economic assumptions that underlie those Medicare and Social Security projections. The total comes to $12 trillion to $14 trillion . more than the Social Security and Medicare shortfalls combined. What this means is that the revenue that will be sacrificed because of those tax cuts is not a minor concern. On the contrary, that revenue would have been more than enough to "top up" Social Security and Medicare, allowing them to operate without benefit cuts for the next 75 years.

So, just this once, I'll say: Go read him.

Comments | March 21, 2003

Little To Say

I've started, and deleted, many posts about Bush's War. Ultimately, I've decided that I simply don't want to talk about it. Much of the harm that I feared would come of it has already come to pass; the rest of the harm I fear is probably yet to come. But none of this will be apparent for some time -- it took a decade for the consequences of the first Gulf War to come back to us in the form of Bin Laden -- and I have little appetite for talking about the immediate details of war.

So, I won't be talking about the war here; if this seems like ignoring the elephant in the living room, so be it. There are plenty of places to get all the latest updates on the elephant, and few enough places where one can think about matters non-elephantine.

Comments | March 20, 2003

Notes Toward a Methodology of Evaluation for Interactive Entertainment

Some computer game reviewers can't understand why critically lauded titles fail and really lousy games sell a zillion copies. The folks at Avault have put up a pretentiously-titled article examining the phenomenon.

It's pretty dire. The writer begins by considering the possibility that casual gamers are just stupid; the possibility is quickly dismissed, for thin reasons: "I don't believe the majority of today's game consumers are morons." This isn't exactly a compelling argument; one gets a sense that the writer doesn't quite believe it himself, but couldn't get an article out of the stupidity hypothesis. Having summarily eliminated the possibility of stupidity, the writer desperately casts about for explanations of why people buy bad games.

At this point, someone with a connection to the real world in which people make purchases would be shouting out answers: They buy the games that are $19.99! They buy the games with cool-looking boxes! They buy the games they saw advertised in the Best Buy flyer! They buy things that look silly, or fun, or whatever! In short, they buy things semi-randomly, impulsively, and haphazardly.

This thought occurs to our intrepid writer, who notes somewhat disbelievingly that "Some customers still engage in impulse purchases at software stores, without any rational investigation of the highs and lows of a particular product." But he seems unable to believe that many of these people exist, and goes on to consider other, highly theoretical reasons that people might buy lousy games -- wondering, for instance, whether people "may transfer discernment criteria learned in other entertainment media".

The writer ultimately concludes that casual gamers might not be stupid, but that when it comes to game-buying they are ignorant, credulous, and lazy. This is probably the first sensible thing he's said in the entire article, but he quickly severs his tenuous connection to reality by calling for a revolution in the way that people buy computer games, in a final paragraph that's almost den Beste-ian in its obliviousness:

In today's competitive market, the PC gaming world must strut its best stuff. Since stores will always choose what's popular over what's high quality, the only way to increase the comparative attractiveness of PC entertainment is to get the best titles more visible by somehow increasing their popularity. To accomplish that goal, the bulk of the gaming public needs somehow to become more wary about their purchases, gaining wisdom at a faster rate than the experience of buying deficient titles would teach them. Let's all promise to increase, through whatever means, the proportion of the computer offerings we buy that we can truly be proud of, that are enduring monuments to human creativity. Let us cease to revel in the mundane, and instead reward the imagination of those few coders who actually take us to a higher plane of ecstasy. Above all, while maintaining respect for differences in tastes, let us use the summation of our reasoned individual purchases to help spur developers and publishers to give us their best.

I too hope that we can all gain wisdom at a faster rate, increase the amount of stuff we can all be proud of (through whatever means), and cease to revel in the mundane. Also, I want a pony.

Comments | March 19, 2003

The Power of Cheese

The Sci-Fi Channel this week is showing a miniseries of God-Emperor Children of Chapterhouse Dune: Messiah: House Atreides. I watched a bit of the first episode, but didn't get absorbed enough to stick through the whole thing. In the bit I did watch, though, I saw enough to realize that the capabilities of cheap CGI have utterly transformed the low-budget SF genre.

Everyone knows what CGI can do on the high-budget films: Star Wars, Spider-Man, and The Lord of the Rings all testify to the visual impact you can get if you've got a few hundred million to throw around. But the impact on low-budget efforts may be even more important, because of the larger differential between pre-CGI and post-CGI outcomes.

On a big-budget movie like Star Wars, adding CGI makes things a bit cooler. Now Yoda can get in a lightsaber duel; you can have enormous, elaborate battles and sets. But ultimately, the difference isn't that huge. Yoda as a puppet was just fine, and I don't remember anyone complaining about the battles and sets in the classic Star Wars movies. With low-budget stuff, on the other hand, the difference is unspeakably huge.

I've always had difficulty taking cheap SF seriously. It's not just that it's fakey, it's that it's silly. All the styrofoam "rocks" and chintzy scenery are too pathetic for words. What's worse is that panoramic grandeur and magnificence have been totally off-limits, because there's no way in hell the budget can handle wide shots (unless there are a few static paintings). But now with cheap CGI, that's no longer the case. The Dune miniseries has gobs of dramatic scenery, dynamic shots of spaceships landing in city spaceports, and suitably impressive computer-generated worms.

Now, because it's cheap, the graphics don't look realistic. They look slightly better than a high-end computer game (like Morrowind, say), but aren't in the same league as the high-end effects that Pixar and Lucasfilm can produce. But the thing is, it doesn't matter. No matter how unrealistic they are, they don't look cheesy. They're not embarrassing, and they don't make you giggle. You have no doubt that you're looking at an obvious special effect, but it doesn't keep jolting you out of the story.

Which is why I think that Star Wars: Episode III: Attack of the 50-Foot Bugs From SPACE! should rededicate itself to rubber alien suits and claymation fighting -- anything that can jolt you out of a Lucas story is bound to improve the moviegoing experience.

Comments | March 18, 2003

No More Portals. Please.

Tim Bray should be a telecom exec.

The clueless ones are those who are planning to compete with Comcast by loading up their own offerings with music and other "broadband" fluff.

"For the early adopter, `fast and always on' is great," said Michael Grasso, who is the executive director of Internet marketing at SBC Communications , the telephone company with the most broadband subscribers. But the more typical consumer will want more than a commodity service, Mr. Grasso said. "As we try to push the service to the mass market," he said, "we need to go beyond just selling a fast pipe."

BZZZZZZZZZZZZZZZZZZZZT! Wrong. We don' need no steeekin' advanced features. Fast pipe and always on, you betcha, where do I sign up?

Whenever I hear cable/telco execs talking about how they're going to deliver streaming video portals to entice people to upgrade their dialups to broadband, I have to wonder if they actually use the Internet. I don't want that. Nobody I know wants that. Just give us the damn connection already, and let us figure out what to do with it.

Comments | March 17, 2003

The Use of Allies

Newsweek has an excellent article on American foreign policy, how drastically it's shifted under Bush, and why this matters.

It seems entirely incredible that the Bush administration hasn't yet figured out that working cooperatively with allies gives the U.S. more freedom of action, rather than less; or that actively seeking to prevent rivals from rising to power is the best way to ensure their eventual existence. And yet, here we are, having managed to piss away enormous reservoirs of accumulated goodwill for absolutely no sensible reason. In its boundless zeal for war, the Bush administration somehow managed to lose a popularity contest with Saddam Hussein.

The war presumably starts with tonight's Presidential address and will be over shortly; the era of American isolationism began months ago and is likely to repercuss for years.

Comments | March 17, 2003

Mutton Beret

The most amusing ad I've seen in recent years was, oddly enough, an anti-drug poster. Next to pictures of marijuana were statements of what (presumably) the smokers of that marijuana had done. Most of the statements were about as ham-handedly obvious as you'd think ("I spent the night in jail."), but one of them stood out for pure creative value: "I made a hat out of meat."

As cool a mental image as the meat hat is, I had my doubts that anyone -- stoned or sober -- had ever actually constructed such stylish-yet-perishable headgear.

I was wrong.

Comments | March 17, 2003

Cheap Liquor

Back when I started buying wine, I was flush with cash; I didn't buy expensive wines, but I didn't buy super-cheap ones, either. Most of what I picked up was in the $12-$15 range. Every now and then, I wondered if I was wasting money buying mid-priced wines when I could have gotten cheaper ones that were just as good.

Well, for the past few months, we've been saving up for a house, so have been trying to shave dollars off our budget anywhere we can. The wine budget was an obvious candidate, and we resolved to buy cheap wines. I can therefore report that no, I wasn't wasting money.

Without exception, every wine we've had has been terrible, verging between undrinkable and tolerable. It's not like we're buying blind, either -- our wine store posts little reviews on the shelves, and we're buying things that have gotten favorable reviews, but it doesn't seem to help.

Tonight, though, the whole damn experiment was made worth it, because we finally found a cheap wine that was actually good. It's Borsao Campo de Borja 2001, from Spain at $6.99. I'd never have predicted it, since most pricier Spanish wines I've had have been disappointing, but there it is. It's nice to know that being on a budget doesn't inevitably consign us to Sutter Home-level wines.

Comments | March 15, 2003

Writin' XHTML Good

You've read the XHTML spec, and have decided to start writing XHTML. But then you go and read Ian Hickson's warning against XHTML, and suddenly you're not so sure any more. You don't understand half the stuff that Hickson is talking about, but the stuff you do understand is scary enough to convince you that in you're in over your head. Well, it's not really that bad. Or rather, it is that bad, but smart work practices can help you avoid the problem.

The difficulty in writing XHTML today is that you're effectively working blind. Today's XHTML-compliant Web browsers are capable of working with XHTML in two ways: In backward-compatible mode, using the same lenient parser they use for regular ol' HTML; and in strict-XML mode, using an absolutely unforgiving XML parser. To tell the browser which mode you want to use, you configure your server to send back the appropriate MIME type: text/html for lenient backwards-compatible mode, and application/xhtml+xml for strict XML mode.

You really want to serve your XHTML in strict-mode application/xhtml+xml; only by doing so do you really get the advantages of XHTML. But this proves to be difficult in practice, for two reasons. The first is that legacy Web browsers (all versions of IE, and pre-Mozilla versions of Netscape) don't understand XHTML served as XHTML; the second is that many Web servers are configured to send all HTML/XHTML files as text/html, and page authors may not have the authority to change this configuration. Because of this, page authors are pretty much forced to send their pages as text/html for now. Herein lies the problem.

Because, see, now the author is only viewing the page as rendered by a forgiving (and slightly incompatible) HTML browser, rather than through an XHTML browser. This means that there's a chance for invalid markup to creep in unnoticed, or for more subtle incompatibilities in CSS and JavaScript to pop up.

Is this problem so insoluble that XHTML-using authors should just give up and go back to using HTML 4.01? Fortunately, no. Here's what authors can do to ensure that XHTML stays functional:

1. When editing/authoring, view the page in strict XHTML mode. This is the single most important piece of advice I have; if you do this, almost everything else becomes unnecessary -- because you'll be viewing the page in strict mode, you'll immediately be able to tell when you've got something wrong. If you break your XHTML, you'll get an XML parser error; if you break your JavaScript, you'll get JavaScript errors, and if your CSS isn't right, you'll see what's wrong. Validating by eye might not be the most reliable method in theory, but in practice, it's just about unbeatable.

   If you work on files locally, it's easy to ensure that you view them as XHTML -- just make sure that your HTML files have either .xhtml or .xml extensions and view them in an XHTML-compliant browser like Mozilla. If you work on files remotely, it's a bit harder, but in a pinch you can just save the files to your local machine with .xhtml extensions. You'll almost certainly get lazy about that, though, which makes the next piece of advice more important.

2. Use XML tools as much as possible. Instead of typing your pages up in a text editor or FrontPage, use XML Spy (if you're on a company budget) or one of the many free XML editors (if you're not). These won't catch the subtle CSS and JavaScript problems, but will make sure that your content is always valid.

3. Validate. The W3C's HTML Validator can validate XHTML. This is always handy, but if you're producing XHTML with non-XML tools and are unable to check it visually in strict XML mode, it's essential.

And if all that sounds like too much work... well, go ahead and write your XHTML anyway, if you want. It won't hurt anything. But do keep in mind that when you eventually end up switching to strict XML processing of your XHTML, things are almost guaranteed to break. When that happens and you need to fix it in a hurry, just remember to keep your cursing aimed at yourself and not XHTML. You lazy, shiftless bum.

Comments | March 14, 2003

My Hands Twitch a Little Bit

Whenever I'm cut off from the Internet, I feel like I'm working on half a brain. There's a constant feeling that I could do something easily, if only I had an Internet connection; but without it, it's impossible. Examples:

• Browsing in Barnes and Noble, I pick up an interesting-looking book. Is it any good? I have no idea, since all I can look at is the cover. If I had the Internet, I could read reviews at Amazon and find out whether or not I wanted the book. (And double-check my book database to see if I already own it.)
• Driving through an unknown city, I'd like to find somewhere to eat. Are there any good restaurants around? I could find out in a few minutes if I could google up some results.
• Sitting at a restaurant with friends, we decide it'd be fun to take in a movie. What's playing and where? Give me a few seconds alone with the theatre Web site, and I can tell you. Oh, well, I guess I'll just have to call them up. I'll get their phone number from... shit.

I'm sick of it. I want my full brain at all times, dagnabbit. The problem, of course, is that the technology just isn't there yet. The best mobile brain augmentation device that I've been able to find so far is the Handspring Treo, which combines phone, Internet, and PDA capabilities all in one little device. So far, though, I haven't bought one, because I'm just not sure if it's good enough. Yeah, I can read the spec sheets and reviews, but that doesn't tell me what it's like to actually have one.

But this account of a week spent using the Treo does. Alas, after reading that, I have to conclude that it wouldn't quite do for my needs. I mean, it'd be better than nothing, and if someone gave me one I wouldn't give it away -- but I'm not going to drop my own $400 on the thing. I'll just have to wander around in this half-lobotomized state for the next few years, I guess.

Comments | March 14, 2003

The Perils of Upgrading

So I see that Mozilla 1.3 was released today. Since I'm a rabid tech-head and an avid fan of the Mozilla browser (you should be too: if you haven't tried it lately, download 1.3 and take a look), you'd think that I'd be all set to upgrade, wouldn't you? Well, you'd be wrong.

The problem is precisely that I am a tech-head. Every piece of software I install gets tweaked up to my precise liking. This makes everyday use far more pleasant -- but makes it an enormous pain-in-the-ass to upgrade, because I have to redo every single one of those tweaks. So mostly, these days, I don't bother. Unless there are major flaws or missing features that are fixed in the newest version, I just keep using whatever version I happen to have.

Although now that I think about it, maybe it wasn't such a great idea to use TurboTax 1997 to prepare my taxes for this year.

Comments | March 14, 2003

Getchyer fresh, roasted brain implants!

Over the past five years, I've mostly become inured to "Holy shit, we're living in the future!" news. Cloning? Boooring. Huge worldwide repository of human knowledge available from my cell phone? Well, the screen's a bit small.

But even I'm not quite jaded enough to be blase about scientists reverse-engineering parts of the brain as if it were a microprocessor to be cloned:

No one understands how the hippocampus encodes information. So the team simply copied its behaviour. Slices of rat hippocampus were stimulated with electrical signals, millions of times over, until they could be sure which electrical input produces a corresponding output. Putting the information from various slices together gave the team a mathematical model of the entire hippocampus.

They then programmed the model onto a chip, which in a human patient would sit on the skull rather than inside the brain. It communicates with the brain through two arrays of electrodes, placed on either side of the damaged area. One records the electrical activity coming in from the rest of the brain, while the other sends appropriate electrical instructions back out to the brain.

I'm sure this doesn't actually work right now, and probably won't for a decade or so. But it's boggling that it's possible to even be thinking about this in a real-world context. Of course, the initial tests will be run on rats and monkeys, and even if it's successful then, they'll have to adapt it for people. They say it won't be too hard, though:

The hippocampus has a similar structure in most mammals, says Deadwyler, so little will have to be changed to adapt the technology for people.

It's always nice when nature provides you with a stable cross-platform API.

Comments | March 13, 2003

Progressive Taxation and its Discontents

I was talking yesterday about the nature of the tax system, and my co-conversant referred to our current system as one that taxes the rich more heavily than the poor. While that is, in some sense, an accurate summation, it's not the most accurate one.

The popular view of a progressive tax is one that focuses on people rather than money ("the rich people pay more than the poor people"), and on marginal rates rather than total rates ("the rich pay 39.6%, and the poor pay 10%"). This naturally sounds unfair to a lot of people who believe in equality -- why should the tax system discriminate for or against people, and make them pay wildly different rates? You can answer the critique by appealing to a sense of social justice, but the violation of equality really does bother people.

But the thing is, the inequality is a complete mirage -- everyone really is equal before the tax law. To turn an old saying on its head, the majestic egalitarianism of the law compels rich and poor alike to pay 39.6% on income over $280,000. What a progressive tax system is really doing is not taxing "the rich" more heavily, nor "the poor" more lightly; it's recognizing that money has different personal utility -- that the difference between $0 and $10,000 is more important on a personal level than the difference between $1 million and $1.01 million. This recognition is codified by letting us pay almost nothing at all out of our low income; middling amounts from our middle income; and decent chunks out of our vast wealth. So Bill Gates pays the same taxes on his first $10,000 of income as does someone who only makes $10,000 that year. There's no discrimination there, no punishment of Bill Gates for being rich.

The rich really aren't getting screwed by the poor paying low tax rates, because the rich pay those exact same rates. If we were to further lower the tax rates on income below, say, $30,000, Bill Gates would benefit from that reduction in exactly the same amount that the poor would, and that I would. And yes, Bill pays more on his income over $300,000 than he does on his income under $30,000. This is not punishment for being wealthy, though -- it's just the realization that those fabulous sums of wealth are of less importance to Bill than that initial $30,000; the tax code takes advantage of the law of diminishing returns to draw its revenue where it hurts the least, without ever unfairly punishing one person at expense of another. It's utilitarian and egalitarian at the same time.

Comments | March 12, 2003

Eliminating the French

Mike Shaver (or rather, his wife) has a better suggestion than renaming French fries for those who'd like to reduce the quotient of anti-American Frenchosity in the US:

If they are that pissed at France and want to do something really symbolic, they should return the Statue of Liberty. At least that would be an actual comment on their country's ideals, rather than the needless renaming of tasty but irrelevant foodstuffs.

Comments | March 12, 2003

And so it begins...

CalPundit takes the first step on the nightmare descent into drugs and madness.

Comments | March 11, 2003

Surprising Coolth

So way back when the Mozilla team decided to throw away the Netscape 4 code base and start from scratch, they decided to also write their own custom cross-platform GUI code. They've been raked over the coals for that decision ever since -- it's reinventing the wheel, it doesn't look native, it's too bloated and slow.

Well, it's also pretty cool. XUL (for so their UI language is named) is a custom XML dialect for defining user interface elements (e.g., windows, dialog boxes, buttons, checkboxes), which then use good ol' CSS and JavaScript for their appearance and actions. This makes the Mozilla UI roughly, oh, a billion times easier to modify than separate platform-dependent Win32, Carbon, Motif, and Gtk code would have been.

It also makes it easy for third-party developers to create useful plugins like mouse gestures and preference bars. The importance of this shouldn't be understated: tabbed browsing first came into Mozilla via such a plugin, and experimental UI concepts like pie menus can be tried out this way before being integrated into the core. This is something that you just can't easily do with Internet Explorer or Opera, and certainly not in such an effortlessly cross-platform way.

But none of this is the main reason XUL is cool -- that would be its XML namespace: http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul.

Comments | March 11, 2003

Sharing the Wealth

Lots of people can make theoretical, principled defenses of their positions, and when it comes to income inequality, most of them do. But it's a lot more rare to see a pragmatic, empirical look at the matter. Vaguely Right has a rather nice one, though.

The biggest macroeconomic argument in favor of income inequality is that attempts to restrict it would introduce distortions and slow down growth, making everybody worse off. A rising tide lifts all boats, after all; if the bigger boats get lifted higher than the small boats, well, the small boats are still higher than they were. As Mr. Right points out, even a small impact on growth rates will be enough to make everyone worse off:

Suppose you had the choice to vote for government policies in 1967 that would have had the effect of keeping income distribution constant. For a majority of Americans, this would have been very much in their best interests. Except that, perhaps these policies would have lowered yearly growth rates. How much of an impact would this lowering have had to have to erase your gains? For the fourth bracket, a decrease in the yearly growth rate of 0.016 would convince you not to support the policy. (This corresponds from going from 5.96% average yearly growth to 5.944%.) The lowest fifth would see their gains erased if the impact on the yearly growth rate was slightly more than 0.04 points (i.e. an average growth rate of 5.92%). And the gains of the third fifth would be lost if the impact was 0.052 points. Finally, with an impact of 0.066 points per year, everyone is worse off, and at this point the mean income for everyone is a low $47,063 and the top fifth itself has lost $42,935 worth of yearly income in 2001 due to this policy.

So, it seems that inequality is just the price we have to pay for increased growth, and therefore increased happiness. (Unless, that is, relative prosperity is more important than absolute prosperity.) But somewhere along the way, an unexamined assertion got made -- how do we really know that restricting inequality would hurt growth? Isn't it possible that inequality itself slows down growth? Vaguely Right runs the numbers and finds:

Income growth rates are negatively correlated with the Gini coefficient, suggesting that when inequality is high in one year, growth is low in the next year, and vice versa. ... Years that see large increases in inequality are followed by years of lower than average income growth.

As he's quick to point out, this is pure back-of-the-envelope stuff, and not to be taken as conclusive. But conclusive or no, it is suggestive. Those who favor laissez-faire economic policies need to come up with some actual evidence if they want to keep claiming that continued inequality is best for everyone.

Comments | March 11, 2003

Ghost Town Traffic Planning

What ultimately starts to get to you about downtown Detroit isn't the obvious. It's not the crime, it's not the filth, it's not the excessively high beggar to general passer-by ratio, it's not even the blocks and blocks of abandoned, boarded-up buildings with broken windows. No, what really gets you is the damn traffic lights.

Once upon a time, I'm told, Detroit was actually a thriving metropolis -- a place where people lived and business was done. Nowadays, it's a small town living in the empty husk of a big city. But nobody seems to have told the traffic department, because the lights are still set for Thriving Metropolis. Which means that every day as I drive home, I have to sit at red light after red light, waiting for throngs of non-existent traffic to pass through.

It's depressing, these little reminders that there was once a real city here. And okay, I lied: The sea of abandoned buildings does get to you. Most of them still have their old facades, and you can see that there was once a luxury hotel here, a bank there; and now, nothing. It's depressing to think that the Book-Cadillac Hotel now looks like this. It's even more depressing to realize that it looked like that in 1997, before the city stopped providing a security guard for the building, and that by now the chandeliers and moldings have been looted.

It's not all doom and gloom, admittedly. A site devoted to the ruins of Detroit points out a bunch of buildings that have been brought back from the brink -- and even the poor old Book-Cadillac is slated for renovation soon. Then there's the new stadiums, the imminent move of Compuware downtown...

But I just can't picture a revival. The Compuware move has been all over the news, because it's the biggest employer Detroit's managed to attract since GM relocated downtown. It's a huge, huge thing for Detroit. And yet... I walk past the new Compuware building every day. It's a nice building. There's nice development going on around it. But a block away, it's back to empty buildings, dirt sidewalks, and boarded windows. There's just too much desolation for anything to have a big impact. The weight of the city's desertion makes every bit of development feel like an island under siege.

Even so, I like to think that things could turn around, and that the Detroit of 2010 could be hailed as a model of civic rebuilding. There was a real city here once, and maybe there will be again some day. In the meantime, I think I'm going to start running red lights.

Comments | March 10, 2003

Ergo Propter Hoc

An article about Microsoft's intern program notes:

Maintaining the [intern] program, even during a down economy, is good for business, Challenger said.

Studies, he said, show that companies which are committed to healthy internship programs and hire during recessions have stronger growth potential once the economy recovers.

Studies have also shown that people who buy Jaguars are more likely to get jobs that pay six figure salaries.

Comments | March 10, 2003

Irrational Expectations

Warren Buffett's latest letter to Berkshire Hathaway shareholders (PDF) is worth reading, both for the lucid explanation of how insurance businesses work, and for the novelty of a CEO representing matters accurately even when it reflects poorly on his own performance. (A cynic might note that Buffett's got so much accumulated credibility that he can talk about his failings all day without diminishing his stature, a luxury which lesser managers don't possess. A countercynic might note that much of that accumulated credibility comes from his honesty.) It's hard not to believe that the world would be a better place if there were more CEOs like Warren Buffett and fewer like Dick Cheney.

Even if you don't care about Buffett or Berkshire Hathaway, there are interesting bits in there about the stock market. Buffett's take on the market, which is in line with most sensible analysis that I've seen, is that the market is still overpriced. Explaining why he's not increasing his equity investment in companies he owns, he says:

Though these enterprises have good prospects, we don't yet believe their shares are undervalued. In our view, the same conclusion fits stocks generally. Despite three years of falling prices, which have significantly improved the attractiveness of common stocks, we still find very few that even mildly interest us. That dismal fact is testimony to the insanity of valuations reached during The Great Bubble. Unfortunately, the hangover may prove to be proportional to the binge.

What's most surprising about the Bubble valuations is how much people still seem to set store by them, even while insisting that they knew all along they were meaningless. There's a widespread, persistent belief that if we can get the economy back to where it was in 1999, the market should be back where it was, too: Sure, the Dow fell 4,000 points from its peak, but hey, the market is volatile -- it'll come back when business picks up.

It's just not true. Valuations that were insane in 1999 are still insane. If the economy were today to rebound to its booming 1999 state, the stock market... well, the market would probably be just about fairly valued where it is. There's absolutely no reason to expect a huge market surge when the economy turns around, and every reason not to. There's not going to be any offsetting gains for the huge losses investors experienced from 2000-2003. (Or rather, there already were offsetting gains, from 1996-1999.)

Comments | March 10, 2003

The March of Progress

If you ever start to think that innovation is dead, march on over to your grocer's freezer section and take a look at the selection of frozen pizza. You're certainly already familiar with the basic frozen pizzas; you've doubtless grown accustomed to the "rising crust" pizzas; and the existence of "cheese stuffed crust" pizzas is by now old hat. But relatively new (at least for mainstream national brands) is Chicago-style deep dish pizzas. Let us observe a moment of silence to honor the unfailing efforts of frozen pizza researchers.

Although now that I've had a moment to think about it, it occurs to me that DiGiorno seems to be coming up with these innovations at a predictable rate of about a month after Pizza Hut, so maybe "derivation" is a slightly better word than "innovation." But let's not fuss about details.

(And no, I'm not getting any kickbacks for this. Though if the folks at Pizza Marketing Quarterly magazine would like to talk, I'm all ears.)

Comments | March 9, 2003

Now I need RGB pants

To the perhaps minimal extent that the colors on this site don't make you want to claw your eyes out, you can thank the EasyRGB Color Harmonies tool. Just plug in a color that you like, and get back a handful of potential complements and accents. It's automated fashion sense for Web pages.

Comments | March 9, 2003

In Order to Save the Village

Okay, so this is a bit of a silly article, but it has a great line:

And yet it worries me that Mr. Bush says that one of the reasons he wants to kill a lot of Iraqis is because Saddam Hussein has also been killing them. Is there some sort of rivalry here?

Comments | March 9, 2003

Too Hard For Humans

Following up on yesterday's post on the increasing difficulty of HTML authorship, I've come across an insightful (and humorous) presentation by Steven Pemberton (who happens to be the Chair of the W3C's HTML Working Group). Pemberton illustrates the increased complexity with some concrete examples, including a particularly painful bit of XLink.

The use of XLink as an illustrative example isn't coincidental. XLink has a lot of political baggage. It's supposed to be the next generation of linking technology, replacing the simple <a href> of HTML, but the HTML Working Group thinks it's a horrid mess and was so reluctant to put it into XHTML 2.0 that they invented a competing specification called HLink. Within weeks, the Technical Architecture Group (which sets broad policy outlines for the W3C as a whole, and directs the individual working groups), shot down HLink and insisted that the HTML people use XLink. In the latest working draft, linking is still an open issue.

Politics aside, though, it's nice to see that someone at the W3C is concerned about HTML complification. Pemberton's conclusion is especially worth emphasizing:

Computers are getting more powerful

People aren't.

Comments | March 9, 2003

False Dichotomies

An essay entitled Why Nerds Are Unpopular has stirred up a lot of conversation over at Making Light, most of which makes me feel like I'm living in an alternate reality.

The premise of the article -- which most of the commenters find convincing -- is that smart kids are unpopular because... well, never mind the reason: I don't buy the premise. In my school, at least, smart kids weren't unpopular. I'm not going to say that popularity and intelligence were synonymous, but they were definitely correlated. Most of the kids in the AP classes were among the cool kids, and most of the kids who got picked on weren't particularly bright.

Nor was there any jocks vs. nerds dichotomy: those kids in the AP classes tended also to be on the sports teams. The main dividing line in high school was actually a class boundary: the middle/upper class kids who were into academics and sports on one side, and the working/lower class kids who were into smoking and snowmobiles on the other. But even on that axis, there wasn't a lot of antagonism -- the two groups just didn't mingle very much, because they didn't have much in common.

So why do so many people have different experiences? Why do the "smart and nerdy" and "jocks vs. nerds" stereotypes have such wide currency? My suspicion is that things in high schools have changed in some indefinable way in the last 10-15 years, and those who graduated before that haven't recognized the extent of the change.

Comments | March 8, 2003

How many economists does it take to screw in a lightbulb?

I never took any economics courses in college, which pisses me off to no end, because the further I get from college, the more fascinating it looks.

And I'm not just saying that because Brad DeLong is a funny guy. Although he is.

Comments | March 8, 2003

Standards vs. People

A few years ago, the most important fight on the Web was Standards vs. Software. The W3C had defined a slew of great standards, but browsers were ignoring them and implementing their own proprietary functionality. CSS support was mostly theoretical, and HTML was being corrupted by things like "font" and "layer" tags.

Well, things have changed. The major browsers these days have great support for the standards. Mozilla supports everything from XHTML to CSS2 to most of the DOM Level 2 modules; Opera isn't far behind; and even lowly Internet Explorer is pretty solid if you stick to the basics. So the Web ought to be a paradise of high-quality, standards-based pages, right?

Not quite. Because it turns out that browser authors were quicker to learn the standards than most Web content authors. Most sites, for reasons initially pragmatic but now just inertial, continue to rely on the old-school ways of doing things -- table-based layouts, font tags, broken HTML, and JavaScript that only works in IE. Millions of people learned HTML back in the bad old days, and a lot of them aren't very technical. When non-technical people know how to get something done on a computer, telling them that there's a better way (if they'd just invest a half-dozen hours to learn it) isn't something they're interested in hearing. Ironically, HTML is a victim of its own success.

And it's going to get worse. The specs that are important right now are mostly comprehensible: If you learned HTML in the 3.2 days, it won't take you all that long to get up to speed with HTML 4.01 and basic CSS 2. But the Web of Tomorrow is based on a whole boatload of XML standards, most of which are different from, and more complex than, the technologies that they'll be replacing. To replace the functionality of familiar HTML 3.2 with the newest, still-in-progress standards will require authors to learn: XML, XML Namespaces, XML Schema, XHTML 2.0 (which isn't backward-compatible with any previous version of HTML), XML Events, CSS3, and XForms.

There are a few ways things could play out. The first (and worst) possibility is that these standards will just prove too complex for Web authors and be ignored entirely. The second possibility is that casual authors will increasingly take to using content management systems like Blogger or Movable Type to take care of the technical details for them. The third possibility is that things only seem complicated now because all the technology is so new, and that in five years, casual authors will be as proficient with all those XML technologies as they are with the font tag today.

I'm hoping for the second possibility, but fearing the first -- and dead certain that the third won't happen in our lifetime. Either way, I'm afraid we're seeing the end of the era of widespread advanced HTML literacy. The experts have moved in, and the Web of Tomorrow isn't going to be a casual hobbyist's playground.

Comments | March 8, 2003