21 December 2005

Bush on the importance of court orders

Now, by the way, any time you hear the United States government talking about wiretap, it requires—a wiretap requires a court order. Nothing has changed, by the way. When we're talking about chasing down terrorists, we're talking about getting a court order before we do so. It's important for our fellow citizens to understand, when you think Patriot Act, constitutional guarantees are in place when it comes to doing what is necessary to protect our homeland, because we value the Constitution.

That's President Bush, touting the Patriot Act in April 2004. By the time he made that statement, the President had already personally approved warrantless wiretaps at least sixteen times.

It sure looks like the wiretap thing was illegal to the point of showing utter contempt for democracy and the law, not to mention the Constitution, but I'm not a lawyer. Regardless of its legality, it seems very clear that the program sidesteps checks and balances between secret police and the courts and between the President and Congress.

11 December 2005

Web services standards rated

It's been a few years since I did any Web stuff. Lately I've been skimming through the new open standards from the past 4-5 years.

My ratings, uninformed by significant experience, on a scale of -4 to 10 (anything 0 or less makes me want to bang my head against the wall).

  • XML: 7. Okay, this isn't new, but it's worth talking about how well it's aged. XML scores huge points for usefulness. In hindsight, some of the design decisions were mistakes. As it turns out, the attention to SGML compatibility was misguided. DTDs and external entity references should have been jettisoned. Entity references such as é probably should have gotten the boot, too—get a Unicode-capable editor for crying out loud. CDATA is questionable. Arguably comments should nest, for programmer sanity.

  • XML Schema: 3. The things they decided to implement were implemented with intelligence and taste. But. The spec is unreadable and ugly. It's hard to get a straight answer from it when you have a question. It has so many features that complete implementations are rare. Part 1 alone is 195 pages. Some features depend on XPath, which itself is hard to implement; those features should have been put in a separate, optional module. XML Schema has a lot of features that aren't particularly useful, like <xsd:notation>, fixed attribute values, and simple types derived by restriction (other than maybe pattern). It has two of everything, except when it can get away with having three: two different flavors of final; two completely different takes on subclassing (derived types and element substitution groups); prohibited substitutions, substitution group exclusions, and disallowed substitutions (which are apparently three slightly different things); nillable and optional elements.

    The worst part is, XML Schema doesn't solve the right problems. Validating documents is convenient and helpful. But XML Schema goes far beyond validation to provide a thorough type system for XML. There are two major problems with this: (1) the type system doesn't map to ordinary programming languages or relational databases; (2) no API is provided for programmers to access the type information that's supposedly present in schema-compliant documents.

  • SOAP: -4. Pointless. Presumably I'm missing something.

  • WSDL: 9. Everything a standard should be. Purposeful, simple, easy to implement, readable, concise. It loses a point for refering to SOAP. But you can (and should) use WSDL without using SOAP.

04 December 2005


I meant to say this last Monday before the game, and I meant to say it before today's drubbing started. If Peyton Manning and Edgerrin James stay healthy, I think the Colts are going to have their perfect season. I haven't seen the team that can beat them. They have an offense a geek can really appreciate, by the way. I say Edgerrin James not because he's such a great player, but because their offense banks on flexibility, and James is their running game.

01 December 2005

Maintenance costs and software engineering decisions

It's an adage of software engineering that most of the cost of a project is maintenance after the initial development effort. But is this really true?

Organizations certainly don't seem to act like it's true. Software shops tend to discourage refactoring because the perceived short-term cost (in breakage) is more than the long-term benefit (in maintainability). Possible explanations:

  1. Software shops are shortsighted. They should refactor more.

  2. Engineers don't distinguish between productive refactoring and useless tinkering. Managers can't tell the difference either.

  3. A lot of code doesn't benefit from refactoring. Most of the maintenance costs of a project are concentrated in certain areas of the code. The trouble is, it's impossible to tell which areas until you've already shouldered the maintenance load and it's too late.

  4. The venture capital model of funding software development has an overwhelming cash flow problem that basically destroys any company that thinks long-term (before it gets to the stage where that strategy would pay off).

  5. So many projects fail that the best strategy is to get your project to the point of cancellation as soon as possible. (Too cynical?)

  6. Refactoring really works, but it's not the best use of your time. You can outgrow your debt. Invest your time in work that grows your company (like shipping the product sooner). The maintenance costs will seem insignificant tomorrow, as long as you keep growing. (Too optimistic?)

Hmm. Ideas 4, 5, and 6 all suggest that internal software departments in large non-software companies (intranets, accounting systems, that sort of thing) should have better code hygiene than hotshot startup companies. Is this true?

16 November 2005

U.S. history fact of the day

Back in the 1790s, it was apparently taken for granted that the U.S. government was a secular institution. I didn't know this. But it's declared, in the clearest possible language, in a treaty read aloud on the floor of the Senate and signed by President John Adams.

11 November 2005

Debt vs. GDP endgame

Why do we even care about the national debt?

Here are four long-term scenarios concerning the debt.

  1. We pay it off. This means some suffering, because it won't be cheap. We owe eight trillion dollars. The reward would be that we wouldn't be paying interest on the debt anymore. (There might be an economic reward, as well as the financial one; but I don't know.)

  2. We maintain it roughly at steady state, relative to inflation. This means we'd have to stop borrowing to cover the costs of present-day government programs. Assuming the U.S. economy continues to grow (an extremely safe assumption), this means the debt eventually becomes quite easy to pay off. At some point in the future, eight trillion Frosties will be chump change.

  3. We maintain it roughly at steady state, relative to GDP. This means paying more and more interest every year; but as long as GDP is growing, we're richer and richer every year, so we don't really feel it. Ever. Economists' comfort level would remain roughly level as the debt grows.

  4. We eventually default on the debt. That is, it eventually gets so out of control that we can't roll it over; not enough people will buy our T-bills anymore. Or we make a political decision to bail. From what I hear, this is an extremely bad scenario for the economy.

    Now the nature of debt is that by the time you start to feel the crunch, it's too late. All kinds of economic mayhem will happen before we reach the point of default: interest rates will rise, investment and growth will stall—but these little disasters will not stave off default. By worsening the economy and killing tax revenue, they'll probably accelerate us toward it.

I'm no genius, but it seems like a path somewhere between #1 and #2 might be a good idea. The path we're on right now falls somewhere between #3 and #4.

A coffee snob speaks

I have to say, when McDonald's decides to do a little marketing, they really do it. In New England, they're carpet-promoting their new, politically correct, organic, gourmet, $1.29-for-a-small-cup coffee. For two weeks, they're giving it away to anyone who'll take it. But what sealed the deal with me is that they're getting the coffee from Green Mountain.

So I tried some. It's kind of amazing. It's as though they were afraid that actually selling good coffee would turn off loyal McDonald's-coffee drinkers. The new custom blend is like a gourmet version of the bitter, acrid coffee they sold before. It's an achievement of some sort, in that I wouldn't have thought it possible. Blechh.

10 November 2005

Time and deficit spending

In the software business, time is precious.

Software engineering teams often borrow time from the future at a very steep interest rate. Two examples:

  1. A team is hurrying toward a deadline. The team has painstakingly produced a detailed design document for the product. To save time on the current release, the team stops updating the document as changes are made. The plan is to bring it back up to date after the deadline.

  2. A company has a product for Windows. The company needs to ship the same product for Mac OS X. One plan would be to rework the Windows codebase to make it platform-neutral. Instead, the team chooses an alternate plan. They branch the two codebases. This allows feature development and releases to continue at speed on the Windows branch, while the new branch works on OS X support. Both projects will reach near-term milestones sooner. The long-term plan is to merge the two codebases, a time-consuming task.

In each of these cases, the team gets extra time now, at the expense of time later. In each case, they must either spend the time later to clean up the mess; or lose a lot of work.

I suspect the interest rate on these loans is high (200%-ish), but then the expected ROI for time (as opposed to cash) can be very high as well.

How should these decisions be made?

09 November 2005

Programming language feature request: Point comments

Programming languages still don't have all the features I want. Here's an example.

I want point comments, special comments that briefly explain the point of the following block of code. The language would guarantee that whenever an exception falls out of a point-commented block of code, the point comment is automatically added to the resulting error message.

Here's some pseudocode that shows what I'm talking about. The characters //- denote a point comment.

 //- set up the flogmogger to do ${job}
 if baz

 //- forward the request to the remote agent on ${flogmog.host}
 env = new Rpc.Envelope
 reply = flogmog.send(env)

Here's how this would work. At the point where the comment appears, the system evaluates any variables that appear in the comment (in this case, ${job} in the first comment and ${flogmog.host} in the second). It stores the values on the stack and proceeds. If the chunk of code under that comment fails with an exception, the exception is automatically caught; annotated with the comment, including the pre-calculated value of any variables that appear in it; and rethrown. These annotations could be stored in a field of the exception object.

Then, whenever an exception object is called upon to create a user-visible error message, it includes these annotations to tell what the code was trying to do at the time of the error. Here's a sample error message:

    Failed to forward the request to the remote agent on localhost:
    RpcException: URI not found (/inventory/donuts).

Of course, if the caller had its own point comments, and the RPC module had its own point comments, you would get something more like this:

    Failed to order more donuts from Xylonium Nuclear Bakery:
    Failed to forward the request to the remote agent on localhost:
    Failed to confirm RPC transaction was successful:
    RpcException: URI not found (/inventory/donuts).

It's like a traceback--but instead of source code files and line numbers, you get meaningful English messages. I think both are valuable for programmers, and improved error messages are always welcome from the user's perspective.

A mechanism for achieving this level of error-message coolness exists in some languages: exception chaining. But really using it requires a lot of boilerplate try/catch/throw.

Another flavor of point comment should be provided to allow for internationalization. I had something like this in mind:

 //= flogMogSetup(job)

 //= forwardRequestTo(flogmog.host)

The actual error messages for each language would appear in separate message files, of course. Here flogmogSetup and forwardRequestTo are just keys into the localized message files.

Comments that actually do stuff are something rather, er, different; but pragmatism beats purity. If necessary, the syntax could be changed to look less like a comment and more like code.

A more significant critique is that these error messages still wouldn't be very good. I consider point comments another weapon against bad error messages. Maybe the ultimate answer is for programmers to spend thousands of hours really polishing the error messages their systems generate. I just don't think that's realistically going to happen. I think point comments would improve the baseline case. It's worth a shot.

05 November 2005

J. turns 2

This is J. He's my son, as you can tell; he looks just like me.

(J. behind a book photo)

J.'s grammy bought him a table for his birthday and came up for the party. (That's J. on the right.)

(J. and Grammy photo)

There were a few balloons at the party. One of them was a very large two, or as J. says, “five”.

(J. with ‘2’ balloon

D. & M. gave J. this very cool cardboard box. It was even more fun after we got the green Winnie-the-Pooh chair out of it.

(boys in a box photo)

J. wasn't the only one to get presents for his birthday. O. and I got some Legos. Here's O. putting the finishing touches on an ostrich.

(lego ostrich photo)

We were previously unaware that all you have to do to get Legos is raise a screaming baby to age two. This is an important revelation. Sign us up for another!

03 November 2005

What does this mean?

(chart of U.S. national debt divided by GDP, 1950 to present) Disclaimer - I threw these numbers together myself in Excel from data from two seperate sources. I don't know what I'm doing. If you use this graph for anything serious, you're insane.

Next in line: Iran

The White House is now talking about how military action against Iran is “a last resort”, which I think means the decision has already been made.

I opposed invading Iraq. I'm not sure about Iran. There are a lot of differences. To start with:

  • Iran really is seeking to build nuclear weapons and has already acquired the nuclear material to do so.
  • Iran's government really is a theocratic regime with ties to Islamist terrorists.
  • Iran might be a lot tougher militarily—it's bigger and in better economic shape.
  • If we invade Iran, we have to worry about Iraq disintegrating.
  • We won't have any allies this time. Britain has already publicly and repeatedly opted out.

So: harder, but more justified. The doctrine of preemption certainly would seem to require war. Does anyone see the case for war in Iran as seriously weaker than the case for war in Iraq?

Last night I finished watching the first season of Veronica Mars. Great stuff.

It's a teen girl detective show with amazingly dense plotting. You get a mystery every week, and developments in one or two multi-episode plotlines, and a new twist in the season-long super-mystery. I don't know how they find the time to cram in all this information, but it's expertly done. Solid dialogue, good world design, very good characters, good acting (mostly). Teen angst and romances are part of the bargain, but it's not overdone much.

TV shows are a lot more complex than they used to be.

Reading friends' blogs is risky. I'm worried about spoilers. They're watching the second season on TV now, and I won't see it until it comes out on DVD.

Netflix is great for watching shows like this.

25 October 2005

Bcc etiquette question

Suppose you have a friend J and he knows this other person K. You receive an e-mail that appears to be from J to K (not to you). Apparently J has Bcc'd you on this message.

Without getting into the content of the message itself, does this situation make you feel like you should treat the content of the message as private (just between you and J)?

23 October 2005


Speculation on the purpose of laughter. No worse than any other theory I've heard, I guess.

Spooky answers to weird questions are a staple of old-school science fiction stories. Isaac Asimov wrote a really bad short story about the question of laughter. I don't remember the title. Perhaps “Jokester” is the one I'm thinking of.

22 October 2005

Here's another

The National Flood Insurance Program spends about $200 million annually subsidizing insurance for flood-prone properties. (I apparently pay about $1 of that each year.) I don't see why it shouldn't be scrapped.

In the U.S. law establishing the NFIP, Congress found that “many factors have made it uneconomic for the private insurance industry alone to make flood insurance available to those in need of such protection on reasonable terms and conditions”. Can anyone tell me what these “many factors” are? Private flood insurance is available... on terms that reflect the risk of flood damage.

The NFIP sounds like a typical government insurance program. It is mandated to charge premiums that are sufficient to cover the risks, plus the administrative costs of running the program. But it certainly doesn't do this in all cases—hence the $200M annual price tag—and I can't tell if it does so in the bulk of cases or not.

It sounds like a classic case of government doing something, at a loss, that private companies could just as well do at a profit. With a private market, you'd also get the benefits of competition (on price and service). What am I missing?

Flood plains

I heard an article about the Army Corps of Engineers on NPR earlier this week.

It sounds like Army Corps of Engineers flood control projects (levees), like interstate highway projects, are often pork-barrel job programs, the kind of district-specific spending that is the basic trading currency in Congress (blah).

But unlike highways, levees cause problems. Building a levee encourages people to build homes and businesses in the flood plain below. It can affect the landscape in unexpected ways, as in New Orleans where the levees directly caused the city to sink into a bowl-like shape. And it drains wetlands, removing a natural buffer that slows flood waters when a flood does occur (and, incidentally, killing all the wildlife that lives there).

Proposed solution: drop the requirement that Corps flood control projects have to demonstrate economic benefits outweighing their cost. NPR didn't give the impression there was anything wrong with this clearly insane idea, nor did they interview any opposing interest.

To me this sounds like a program that just needs to be cut.

17 October 2005

It's not an ethical debate

More evidence that the two sides in the culture war aren't communicating. San Francisco Chronicle:

Scientists are reporting two new ways of creating embryonic stem cells without killing viable embryos[...]

In one case, embryonic stem cells were made from a genetically abnormal embryo designed to be incapable of developing. The other method was an attempt to fashion stem cells from an embryo without damaging it.

I'm not sure what this is supposed to fix. The objection is that experimenting on human embryos is an affront to their dignity. Essentially everything having to do with human embryos is off-limits by this standard, including creating them for scientific research.

The opponents of stem cell research are the same people that opposed in vitro fertilization. The only thing to do is wait.

29 September 2005

A broken link is fixed

Someone wrote me mail:

Dear Jason,

http://jorendorff.com/toys/puzzle.html is a 404. It makes me sad
because I like the other foldout toys are very fun and make me happy.

He was right. The broken link appeared on a page of programmer toys, under the blurb: “This one is a bit of a puzzle. It shows you a few lines of code that have profound implications for the way hundreds of millions of people live every day—and asks you to figure out what the code does.”

I had to laugh, because nothing can deliver on that kind of promise, especially after the extra anticipation of waiting for the link to be fixed. Anyway, it's fixed now. If you know a little Python, take a look.

23 September 2005

Unix strikes again

Today at work someone wanted to add some third-party libraries to our version control system. So he typed p4 add vendorlibs/*.so*. Or so he thought.

The command spit out an incredible amount of garbage on his screen. Eventually we figured out he had actually typed pr add vendorlibs/*.so*. The only way this would be funnier is if pr were actually the print command. It's not, despite what the man page says. It's just a text formatting tool.

09 September 2005

Bloggers encourage donations

Crooked Timber will send you a free mix CD if you donate $100 to Katrina relief.

Asymmetrical Information counteroffers: a homemade pound cake for a $100 donation. (Or, a blog entry on the topic of your choice for $250. A risky offer, though the author prudently excludes her private life from the menu.)

Update: What the heck, I'll write on the topic of your choice if you give $50 to Katrina relief. E-mail me or leave a comment. (Not much of a prize, I know, but you don't want my cooking.)

08 September 2005

Hurricane Katrina

As the President says, there will be plenty of time in the months ahead to assess the government's performance. But it seems safe to say a few things:
  • It is correct to hold disaster response to a very high standard. There is nothing special about Katrina (its severity, suddenness, or unexpectedness; or the problems presented by New Orleans' unique geography) that should deflect criticism or excuse mistakes. Handling worst-case scenarios is FEMA's mission.

  • To date, criticisms of the response have been short on plain facts. A dispassionate timeline of events and responses is needed.

  • The mandatory evacuation probably saved tens of thousands of lives. Thousands more could have been saved by a determined effort to evacuate everyone. (It has been suggested that the city should have been evacuated much sooner, but that smells of hindsight.)

  • Barring the Red Cross from New Orleans while thousands of people were still desperate for food and water was unnecessarily cruel and likely cost lives. (More broadly, turning back individuals who went there to help seems like a bad idea to me, but I am apparently the only one.)

  • Every government account of anything related to the hurricane has been appallingly politicized, from Mayor Nagin's sleep-deprived rant to Lt. Cmdr. Sean Kelly's revised account of the Bataan's response. The President's observation that there will be time for criticism later hasn't prevented him or anyone else from conspicuously passing the buck.

  • The key people in the government didn't grasp the magnitude of what was happening. Neither did the media, which reported Katrina Monday and Tuesday with about the same tone as they reported Andrew (which killed about 65 people; Katrina probably killed thousands). Undoubtedly if the right people in government had been smart enough, prepared enough, and confident enough to ignore the media and proceed with the certain understanding that things were soon to get much worse, many lives could have been saved and much suffering avoided. Whether this is too much to expect from a government agency is an open question. More on this later.

30 August 2005

A prize is awarded

Ichabod's has a trivia question every day. The first three people to answer it get free coffee. Yesterday, the question was: What does the term zugzwang describe in the game of chess?

Zugzwang is one of my favorite words. The coffee was only so-so.

29 August 2005

How to win in Iraq

How to win in Iraq. This article is amazingly good. If all it did was explain what the hell is going on over there, it'd be worth reading. But that's just page 2 of 7.

Cliffs-notes version. Don't let the New York Times masthead fool you; David Brooks is a senior editor at the staunchly conservative Weekly Standard.

26 August 2005

Exchange rates

Can someone explain to me how governments can “pin” currency exchange rates?

China's currency, the yuan, apparently remained pegged to the U.S. dollar for years. How did they manage this? It seems like it would be easy for a third party to arbitrage this arrangement to smithereens. Here's how.

Let's assume for the sake of argument that the "correct" exchange rate is 5 yuan = 1 usd (U.S. dollar). That is, on average, a nonperishable good that goes for 1 usd in America can be had for 5 yuan in China. Now let's say the Chinese government has, by government fiat, set 10 yuan = 1 usd. Here's my brilliant plan:

  1. Buy 500 yuan worth of stuff in China.
  2. Sell it for 100 usd in America.
  3. Pocket 50 usd.
  4. Take the other 50 usd and buy 500 yuan. You can do this at any bank that accepts the mandated exchange rate.
  5. Repeat until there's no more stuff left in China, or no more consumer demand in America, or no more room for dollars in your pockets, or (as seems most likely) the banks wise up and stop giving away yuan.

Net result, you get infinite dollars. Who loses? The manufacturer in China got a fair price, as did the consumer in America. Only the bank got ripped off. So why do banks go along with exchange rate controls?

There are only a few explanations I can see. First, this brilliant plan might be impossible because the friction is simply too much. That is, the bank charges a fee for changing your money; and it costs something to ship the stuff from China to America; and there are taxes and tariffs to be paid; and occasionally you'll buy something that you can't sell for whatever reason; and you have to employ people to do the work. Maybe it's just too expensive. But I don't believe it. If this were the case, there would be no international trade.

Second, maybe the Chinese government steps in to prevent people from running this scam. They could detect that you're spending too many yuan and revoke your business permits and such. I hear the rules involving bringing currency into China are incredibly involved, so maybe this is the case.

Third, the scam is self-limiting. The more it runs, the more it drives down U.S. prices. The lower U.S. prices get, the less attractive the scam is. Likewise, the scam drives prices up in China. That TV you're buying for 500 yuan and selling for 100 usd might cost 600 yuan next month (due to greater demand) and might only bring 75 usd in America (due to greater supply). These prices are much closer to the 10-to-1 exchange rate the Chinese government desired, and they make the scam much less profitable. But even with this explanation, the scam will still run; it just doesn't "go infinite". Undervalued yuan are still being pumped out of banks in exchange for dollars worth half as much.

Fourth, maybe supporting the mandated exchange rate is just a cost of doing business for banks in China. The government requires the banks to do it, at their own expense. In exchange, they're allowed do business in China. I don't see how this would be possible. The price tag is just ridiculous.

Fifth, maybe the Chinese government somehow taps into this cycle so that every time the scam runs, the government ends up with enough in its coffers to pay for all the money it's throwing away. In that case, the higher your tax rates, the more power you have to control exchange rates.

Sixth.... maybe this is what's actually happening. It would explain the ridiculous trade deficit between China and the U.S. But if so, the Chinese central bank is paying for it—right? So in that case, how is the trade deficit a bad thing for the U.S.? Sure, it hurts domestic manufacturing, but it's not like unemployment is out of control over here. And aren't we getting cheap stuff at China's expense?

25 August 2005


Seen on Lambda the Ultimate: Parallel Analysis with Sawzall. Sawzall is a highly parallelizable query language invented at Google. The paper gives several examples of ad hoc queries Google might run internally.

The foolishness of The Wisdom of Crowds

Two competing memes predict what happens when a choice is made by a group of non-geniuses: “the wisdom of crowds”; “design by committee”. Same scenario; two opposite predictions. What gives?

This book review spills the goods. Only some crowds are wise. The author thinks he knows which ones: they are diverse, their members are independent, control is decentralized, and preferences are properly aggregated.

I have a cynical theory about this book. It takes an idea everyone on the right already believes. It repackages the idea for center-left consumption, scrubs it of all math references, adds lots of stories (for interest) and studies (to pass it off as novel), and phrases the thesis such that you can swallow it and retain every political notion you came in with. That is, the thesis has no real consequences. It's unuseful. I guess the book deserves praise for having a thesis at all; cf. the blob of factoids that is Freakonomics.

I have a less cynical thought about this book, too. It reminds me of control systems.

A control system is something like a thermostat. There's some variable it's trying to control (the temperature); and to do that it has sensors to measure that variable (thermocouples) and some kind of machinery that can affect it (your air conditioner). One of its important features is that it deals with random external factors that threaten to screw up the system (it was 110° outside today until noon, when it suddenly became rainy and cold).

Another example: in a robot arm, a control system drives the motors to move the arm to the desired location without overshooting it or taking all day to get there. (This is harder to design than you'd think.)

A typical control system can be described by a system of differential equations. The equations describe not only how the control itself works, but also how the world responds to the things the control can do (for example, how fast your house cools when the air conditioner is on). If the control system is good, the equations converge to a desirable state, in the absence of external disturbances. If the control system is bad, the equations result in ringing (unwanted fluctuations) or don't converge at all (your house freezes over). These equations are a precise mathematical model of the system; you can actually solve them to determine the real-world outcome. A qualitative description of the properties of a good control system might appear on page iv of a nine-hundred-page textbook. The equations tell the real story.

Now the private-sector economy could be seen as a huge system of differential equations. In the absence of external disturbances (weather, new technology, natural resources being exhausted, war, Congress, and life generally), how would it behave? Conservatives think it would converge to a balmy 72°F. Liberals think your house would freeze over.

In comes The Wisdom of Crowds with the brilliant resolution to this hairy* math problem: it depends on four unmeasurable variables, and whether they fall above or below certain unspecified thresholds.


*To English majors, hard problems are thorny; to geeks, they're hairy.

08 August 2005

The kids get little uncles

I called Dad the day Abby was born to break the news, but he trumped me. He and C. are expecting twins. Two boys, to arrive in November.

They'll be James's half-uncles, but two years younger. He'll have little uncles. They'll have a big nephew and a big niece.

Fewer people are dying

Prior to 1995, on average, 145 Americans under 50 died every year from chickenpox complications.

In 1995, the chickenpox vaccine was introduced, and now that number has dropped from 145 to 66.

The vaccine probably isn't economically worth it. That is, for the cost, you could probably save a lot more lives doing something else, like overhauling dangerous intersections.

06 August 2005

James thinks it over

Of course, we tried to prepare him.

Months ago, we told James about the baby in Mommy's belly. We told him what the baby would do, where it would stay, and what it would look like. But there's really no way to prepare a first child for the arrival of the second.

When James first saw Abby, it was clear from his expression that he had to reevaluate the whole proposition. He reserved judgment for several days. He became silent and thoughtful when O. or I brought up the subject.

Now he has gotten used to her. Sometimes he's curious. Sometimes he's sweet. He's still generally crankier than usual.

Abby is born

Olivia woke up at 3:30 last Sunday morning with contractions. At 9:10 AM on 31 July 2005, she gave birth to a perfect baby girl, Abigail Parks Orendorff.

She's sleeping now.

28 July 2005

Processing dazzles

Complexification.net is an online art gallery. On display: visual applets programmed with Processing and Flash MX. Pretentious, striking, unquestionably cool, occasionally beautiful.

22 July 2005

Recent history is weird

It was illegal to sell contraceptives to unmarried women in Massachusetts until 1972, when the U.S. Supreme Court ruled the law unconstitutional (Eisenstadt v. Baird).

Further reading: History of the Pill; Sex-related U.S. court cases.

27 June 2005

I get lost

Twice I have printed out directions from Google Maps, and twice it has gotten me off track. I think Google must be using an inferior database of roads to generate their directions.

On the way to Salem, my printed directions omitted the fact that I-95 North is the same road as Mass. Rte. 128 for several miles; completely failed to mention that when they diverge, you want to take the left fork and stay on 128; and got the exit number for Peabody wrong.

It's a shame, because Google Maps is pretty good otherwise. Its user interface puts it way ahead of its competitors.

But all these sites have a long way to go. For example, none of the free map sites show highway exit numbers on their maps. And none show public transit lines. Or pay parking lots. Driving directions could also be improved dramatically without a great deal of technical effort. The most obvious thing is landmarks. Good directions periodically reassure the traveler that he's on the right path. "You'll pass Boston Common on your left." Or, "Pass through Joe Bloggs Tunnel." Or, "Turn right on Walnut St. at a large stone church."

The other thing that sent me astray this weekend was the sign on I-93 South (into Boston) that says, "Commuter lane - no exits until Storrow Drive". Actually, the Storrow Drive exit is one of the ones bypassed. I think this has been the case ever since the new bridge opened almost a year ago. It's crazy that the sign hasn't been changed.

22 June 2005

M. plays Blank White

My sister M. was in town a week ago. We did the tourist thing and played 1000 Blank White Cards. I guess we should have used the rejects the gang sent back with us from Florida, but we decided to start from scratch.

Normally this game is something of a free-for-all, but with the card set we ended up with, it was surprisingly stable. For me, this makes the game even more fun. After the last hand, we ended up with:

  • six plain point cards
  • six point cards that actually do something
  • eight action cards
  • six counterspells

This is a good mix. I've been thinking about doing the counterspell thing for a long time. My impression is that it really helps the game. It also helped that very few of the cards had you doing something every turn. I think the game works better when it can move along pretty quickly.


Evil Thieving Monkeys - Steal another player's card in play and play it on yourself. (inspiration)

Turn to the Dark Side - Flip any card in play face-down. On the back (if it's blank), draw a cruel, evil version of the front.

Let it all hang out - All new cards must have a PIG theme! +3 pts. (There was a Star Wars theme card, too. Never managed to get both in play at the same time, alas.)

In Triplicate - +3 pts. Whenever you play a non-point card, triple its effects.

I'd better stop here. James wants to look at motorcycles.

19 June 2005

I'm caught in the crossfire

I bought Ben Folds's new album, Songs for Silverman, but I had to return it because it wouldn't play on my computers at home or at work, making it mostly worthless to me. It wouldn't play because it doesn't conform to the Compact Disc specifications. It is a DualDisc.

Supposedly DualDiscs exist to add DVD-like bonus features to CDs. Videos, interviews, that kind of thing. In particular, Silverman comes with a copy of the whole album in 5.1 SurroundSound—a phenomenally cool feature, if I had a SurroundSound home theater, which I don't. But really DualDiscs exist to fight music piracy. I know this because Songs for Silverman isn't sold on CD. This really drives me crazy, since I don't pirate music and haven't for years. I can't even buy an album I like and listen to it at home anymore. Technology sucks.

I'm particularly irritated at Sony because the fact that the disc was not a CD appeared in print about 1/8-inch high on the back of the CD. Barnes and Noble graciously gave me store credit when I returned the DualDisc, even though I had already opened it. I hear some record stores don't do that. They all should; it's borderline fraud to put these non-CD things into CD cases and sell them alongside CDs without warning your customers.

James wants to look

James's first four-syllable word was avocado, but only because he long ago started saying “cycle” for motorcycle. Yesterday he said, “momocycle”—a conscious effort on his part.

I showed James some pictures of motorcycles on Google Image Search and it's now one of his favorite things to do. “Look—cycle?” he'll say. Whenever he's in the computer room he demands to see pictures. It's becoming quite difficult.

Some of his favorite searches are purple fish, red motorcycle, and blue bus. (James is into two-word phrases these days.)

I had forgotten—you do find some strange and beautiful things when you spend time aimlessly wandering on the Web.

A rift in the Ross Ice Shelf. “At the narrow end, [these rifts] are covered by a snow-bridge and are nearly invisible. As a result, they can be quite treacherous, as some covered portions of the cracks are large enough to swallow tracked vehicles.” (more)

A child's toy filled with gasoline - Not for sale in the land of the free, for obvious reasons. But what a cool toy. You can take it apart and rearrange the pieces to make this. The same page has a picture of the most beautiful motorcycle I've ever seen. This motorcycle could star in its own movie.

A search for big rock turns up lots of nice pictures, including this one.

26 April 2005

James barks

Since the snow melted, James wants to spend every possible moment outdoors. Olivia and I have no problem with this. We go for walks. We regularly hit two, three parks a day.

On Sunday night there was a dog in the second-story window of the house across the street. He barked at us, and James said, “Arf, arf, arf,” and pointed at him.

He woke up at 4:00 Monday morning and crawled around the bed saying, “Arf, arf, arf.” We don't have a dog, but our son wakes us up in the middle of the night barking.

I took him downstairs and carried him around for about half an hour until he went back to sleep. I hope that doesn't become a routine; he woke up early this morning too. While I was carrying him around he barked at the birds chirping outside.


Disconcerting day at work.

I woke up at around 4:45, spent about 45 minutes getting James back to sleep, took a shower, and drove in. I got a large cup of black coffee and spent a few hours at my desk squinting. Coffee wakes me up if I'm sleepy, but it also makes me shake and puts a strange edge on the world. Today wasn't the best day to be under the influence, as I had a 10:00 meeting to meet the new VP of Engineering.

The new VP. My boss's new boss, responsible for about 30 people in all. It's his second week, and he's making an effort to sit down and talk with everyone for half an hour or so. He seems like an interesting guy, so I was looking forward to it. I had jotted down a few notes so I would have something to talk about. I got them organized and wandered over to his office.

Then something weird happened. The VP listened very carefully to everything I had to say, took notes, remarked that I had just brought up all the major things he had learned about the situation in Engineering since he'd started, and that he could have saved a week and a half by talking to me on the first day; and added that unless I kicked and screamed, I was inevitably going to end up in management. (I'm sure this was intended as a compliment.) He had other nice things to say, too. It was a strange and very flattering meeting.

Maybe the coffee was making me paranoid, but I spent most of the morning wondering what to make of it.

In the afternoon the VP dropped by my office (God knows how he found it) and mentioned that everybody says I'm brilliant. “You probably don't usually get that kind of feedback,” he said. He's right, I don't. From what I hear, I'm an idiot.

I have to wonder what is going on.

What to do, except make some kind of crazed, doomed effort to live up to a favorable first impression?

15 April 2005

Compression and Surrealism

Come to think of it, the thoughts below suggest a fun new compression scheme.

First, some background. JPEG is a common lossy image format. It has a problem: JPEG artifacts, noticeable image distortion as a result of having thrown away some details. All lossy compression schemes lose information, and since a compression scheme never really knows what information is safe to discard and what isn't, they all have artifacts. But it seems that if a compression scheme had an idea of what chaos is, and what it looks like, then chaotic details could be discarded without losing anything a human would ever notice.

It would amount to including a macroscopic charactarization of some parts of the image. JPEG, for example, has a particular problem with tree branches. But if the file format included a way to encode, “ all right, the whole background of this image is tree branches,” and a few parameters such as the colors and thickness of the branches, then JPEG wouldn't even need to bother trying to encode all those sharp lines and angles. The JPEG viewer would fill in the background with exquisitely rendered random branches.

It's fun to consider how this idea could be taken to ridiculous extremes. The brightly colored background of a sports photograph could be filled in with randomly generated fans. As JPEG has trouble rendering corporate logos (too many sharp edges and fields of solid color), you could randomly generate those too. They all look more or less the same anyway. A dias could be filled with randomly generated politicians; a movie poster with explosions; a gallery with randomly generated modern art.

Anyway, this would result in much better compression: either fewer artifacts for a given file size; or much smaller image files; or a combination. But the resulting image would be part authentic and partly fabricated based on a high-level description of the original.

Arguably this is the direction our media culture is going anyway.

Entropy and compression

A computer scientist might observe that entropy is information content. To describe the insides of a hot air balloon exhaustively (setting aside the Uncertainty Principle for a moment) would require notes about the position and velocity of every air molecule. The crystalline structure of ice would save us a lot of writing, as the possible variation in position and velocity of the individual molecules is much less. Pound for pound, steam contains more information than ice. And so it is with digital information. A GIF or PNG file of a highly chaotic image is much larger than an image that consists of just a few colors and shapes.

Yet somehow it doesn't seem that way to us humans. We have one-syllable words for both ice and steam. Steam doesn't seem so much more complex.

This is because we generalize. We throw away low-level variations in favor of high-level order.

In a high-entropy situation, a lot of the details are, almost by definition, unobservable at the macro level. This means we probably don't care much about them. Lossy compression is the computer science equivalent. If we wish to compress a large amount of chaotic data to a manageable size, we can discard information about details we consider irrelevant. This is what lossy compression schemes, like JPEG, attempt to do.

(I'm not very sure of the technical details here. Still reading about it.)

Evolution and chaos

My view of life on Earth is that it is an accumulation of order made possible by the tremendous influx of energy from the sun. What do I mean by order? I'll get to that in a minute.

Poormattie's view is that life is a random development in the rampant chaos that is the universe: an eddy in a river, or bubbles on a stormy sea.

As always it seems to me that poormattie's opinion is more correct. But the two views aren't as opposing as they seem. In part, the distinction has to do with how one defines chaos and order.

In statistical thermodynamics, entropy is a function of the amount of variation that could happen on a microscopic level without really affecting the macroscopic properties of things. To my mind, entropy is chaos; lack of entropy is order. For example, consider steam and ice. Steam is made up of water molecules careening through space at high speeds and colliding into each other. An individual water molecule in steam could basically go off in any direction it wanted, and the difference would be unnoticeable from a macroscopic perspective. But the individual water molecules in ice are locked into position within a crystal. No variation is possible. Steam has greater entropy than ice.

For a very different and probably completely inappropriate example, Utopian fashion is extremely regular. There can be no deviation from the standard on an individual level without changing that society-wide uniformity. So Utopia has low fashion entropy. It is a highly ordered society.

This definition ignores higher-level order. Entropy is only concerned with the variation allowed individuals at the smallest observeable level. So although life certainly appears chaotic from the perspective of poormattie just looking around at the state of the world, it's hard to say whether the presence of life indicates a state of higher or lower entropy.

13 April 2005

Evolution and long life

G.D., a coworker, introduced me to a radical new idea today.

I had been under the impression that evolution cares little for individuals who are too old to reproduce. Their survival is irrelevant to the propagation of their genes. So they die before long. That is, a species doesn't evolve for greater longevity, beyond child-rearing years. There's no evolutionary advantage to it.

A consequence of this, or so it seemed, would be that natural selection is not working very hard in the rich world these days. Most people in America and Western Europe live long past child-raising age. Natural selection requires inferior genes to go out of circulation through death. No death, no evolution. (Evolution is still happening all the time, but mainly through other mechanisms: sexual selection, for example.)

But this view makes the longevity of humans, modest though it may be, something of a mystery. Especially for women, who typically reach menopause in their forties. (A healthy 50-year-old American woman still has, on average, 30 years ahead of her.) Nature wastes nothing. It seems implausible that people would continue living for decades after evolution no longer had any use for them.

It turns out that women and men whose mothers are alive and nearby start having kids sooner and have more children on average than people whose mothers aren't there. And their children are more likely to survive to adulthood.

Why we die, why we live

Grandmother's footsteps - A much better article. Expensive subscription required.

Kicker: This means natural selection is hard at work even today. According to studies of 18th- and 19th-century records, women who die at the tender age of 50 have on average four fewer grandchildren than women who survive to 70. Even after your kids grow up and move out, evolution is still watching.

08 April 2005


I'm not much for breakfast food. The ideal jorendorff breakfast is a ripe juicy pear accompanied by a slice of raisin spelt bread toasted and topped with Kupel's cream cheese.

Kupel's is a Boston bagelry that makes the world's best cream cheese. I've never been there myself. I buy the cream cheese through a coworker, D.J.L., whose mom goes there every week. For a time I had the name confused. I thought it was “Kegel's Bagels”. A Kegel is actually a pelvic floor exercise. Thanks to D.J.L. for straightening me out on that one.

27 February 2005

James is sick

James woke up in the wee hours Friday morning with a temperature of 102.9°F. We called the doctor; the nurse on call said to give him some Tylenol and lots of fluids and call back in three days if the fever didn't go down. So we did nothing and the fever went down on its own. James was pretty miserable for a day or two and couldn't sleep for more than two hours at a stretch. Now the fever is gone. He has a nasty-sounding cough. He's a trooper.

When I was a kid, my parents routinely cured my childhood diseases with Jell-O and fun music tapes. Or so it seems through the lens of memory.

03 February 2005

James stomping around in his dad's shoes.

01 February 2005

They run for fun

More signs The Economist is going soft: Here's a quote from a light article on the economies of online role-playing games, some of which allow players to sell virtual in-game items to other players for real-world cash, and some of which forbid it.

Normally, this newspaper's devotion to free trade is unwavering. Yet curbing the trade of in-game items is defensible, since game economies are run to maximise fun, not efficiency. While writing his forthcoming book, “Synthetic Worlds”, Mr Castronova has been pondering whether real economies could be run for fun too. “Wouldn't that tip the economics texts on their heads?” he muses.

I'm reminded of the words of the great Dr Seuss (reviews):

From there to here, from here to there
Funny things are everywhere.
Here are some who like to run.
They run for fun in the hot, hot sun.
Oh me! Oh my! Oh me! Oh my!
What a lot of funny things go by.

31 January 2005


We had a summer day yesterday. It must have been 40°F.

We took James sledding. It's a lot of effort, somehow, to get everybody all suited up and out there. James was a little squirmy. He likes going downhill well enough, but all told that's a small fraction of the time you have to put in climbing, waiting for people to get out of the way, and so forth. It was a nice workout, though, and a lot of fun.

I'd like to go back today, but there's a lot of work ahead of me and not that much daylight.

28 January 2005

The food here stinks

In the excellent economic blog Marginal Revolution, Tyler Cowen recently wrote, “Raw ingredients in America—vegetables, butter, bread, meats, etc.—are below world standards. Even most underdeveloped countries have better raw ingredients than we do, at least if you have a U.S. income to spend there, and often even if one doesn't.”

Why is that? Today Cowen posted some guesses, all very interesting. Perhaps it's “an exogenous demand-side question. Americans have bad taste in food, just as the Chinese have bad taste in lounge music. Why, for that matter, do the Japanese like karaoke so much? Why do the Scots serve deep-fried Mars candy bars?” Why, indeed.

Or maybe it's just that canned and frozen foods are so cheap here, and we're so used to them, that the good stuff looks awfully expensive by comparison.

26 January 2005

X, redux

In Scrabble, the X tile is worth 8 points. But the expected value of drawing an X is more like 40 points, thanks to the five two-letter words AX, EX, XI, OX, and XU. (An ex is the guy you used to date. Xi is a Greek letter. The xu is a former Vietnamese monetary unit. But if you stop to read the definitions, you'll never be an elite Scrabbler.) If you know the trick, it's very easy to score up to six times the value of the tile.

So in casual Scrabble, the X is a bomb.

But in competitive Scrabble, each player in a typical two-player match will rack up around 350 points. This means the average play is already worth 30 points or so, and a good play, one where you play all your tiles, is worth 60 points at least. It's hard to play the X in a seven- or eight-letter word. So drawing the X is bad in championship Scrabble for the same reason it's so good in a casual game: it's good for 20-52 points, but beyond that it doesn't much reward skill.


We have perhaps sixty letter magnets on the refrigerator. James likes the X's. He knows exactly which ones they are. He knows the name of the letter x when he hears it, and if I ask for them, he can extract the X's from the other rabble in short order.

Last night I sat him in front of the computer, opened Notepad, set the font to thirty-six-point, and showed him where the X key was on the keyboard. He got it immediately and was pretty excited.

It's so exhiliarating to see realization dawn on your baby's face.

25 January 2005

jorendorff.com is back up

My web site, jorendorff.com, was down for a month or two. It's back up, which means it sorely needs updating.

19 January 2005

How good is your doctor?

Change is coming to medicine. Ready or not.

More and more, doctor performance is being measured. The results are at once obvious and iconoclastic: not all doctors are equally good at saving lives.

Once we acknowledge that, no matter how much we improve our average, [this inequality] isn’t going away, we’re left with all sorts of questions. Will being in the bottom half be used against doctors in lawsuits? Will we be expected to tell our patients how we score? Will our patients leave us? Will those at the bottom be paid less than those at the top? The answer to all these questions is likely yes.

And about time, except for the bit about lawsuits. Read the full essay. Fascinating.

(Also gripping. I don't normally read the New Yorker; my amygdala is reeling.)

How to save the public schools

My coworker S.N. proposes the following scheme:

Suppose you have a child in second grade. Before the school year begins, the school sends you a list of all the second-grade teachers. You rank them: first choice, second choice, and so on. The school then does its best to assign your kid to the teacher of your choice. (Not everyone can have their first choice all the time, but there are plenty of fair, simple mechanisms for making sure the preferences are followed as closely as possible.)

That's it. That's the whole idea. What would this do?

The main intended consequence would be to force parents to get more involved in the quality of their children's classes. Parents would be much more likely to ask other parents what they thought of their children's teachers. Parents, in short, would be more aware of teacher quality. S.N. doesn't think parents do this now, and I think he is probably right.

But there are some problems with this:

  1. Parents wouldn't have the power to do much with this information. If all parents are interested, they all obtain the best information possible, and they all choose the same rankings, then this has no effect. If only some parents are interested, their kids win at the expense of other kids whose parents expressed no preferences. In other words, it's a zero-sum game. The school as a whole doesn't improve.

    Perhaps the propagation of new knowledge to parents in this way would improve schools indirectly. Perhaps parents armed with this knowledge would be more of a political force. Perhaps teachers would compete to improve their standings. Perhaps school principals can do something useful with this quantitative information. But it seems to me all these are secondary effects, if not wishful thinking. Unless parents' choices actually have the power to reward good teachers (schools) and penalize bad ones, it's hard to see how those choices can significantly improve the system as a whole.

    To frame our disagreement another way, S.N. thinks the value of a free market is that people make better choices when those choices directly affect their own future well-being. Good overall decisions emerge from individual self-interested decisions. Parents don't care about teacher quality; ergo schools suck. I think the primary value of a free market is in destroying bad producers and powering the propagation of good ones. Parents can't fire bad teachers or withdraw their financial support from bad public schools; ergo schools suck.

  2. This presumes that choosing good teachers over bad teachers is the way to improve education in the first place. Many important factors are not in the hands of teachers: class sizes, curriculum, interruptions, discipline, funding, and so forth. I think choosing good schools over bad schools is the right level for parents to express preferences.

  3. I suspect the idea is doomed anyway, because teachers would hate it. People don't like being judged.

Interesting proposal, though. S.N. often comes up with original approaches to problems. I wonder how he does it.

17 January 2005


The book on Olivia's bedside table begins, “The place was called Gorgantum.” If you don't know why this struck me as funny, either you've never read a fantasy novel or you've read too many. The Prologue continues in a similarly hilarious vein after that, but the first sentence can't be topped.

Or not by much. Curious, I flipped through the book to a page at random and read this: “Tanaros gazed over a sea of Fjel, with thick hides of smooth grey, of a pebbled greenish-brown, or black with bristles. His troops, his men.” Tolkien meets Dr. Seuss.

14 January 2005

“The Love Charm”

Eugene Volokh is a law professor and blogger. Today he posted a link to a short story he wrote, “The Love Charm”, and—bonus feature!—its original ending which I think I prefer to the published one.

Although “The Love Charm” is interesting in its own right, I'm posting this because the story reminds me of poormattie. It is a story about love, cruelty, and ethical dilemmas. But its tone is abstract and casual, as though the narrator made a secret decision not to venture too close to the story in the retelling, for fear of reliving the pain.

It might be interesting to rewrite this story as a collection of highly emotional scenes and see how much punch could be got out of it.

12 January 2005

Book Review: Passage

Passage by Connie Willis, 2001.

If, as I recommend, you choose to pick up this book, don't read the blurb on the back. It may reveal too much.

Two scientists study near-death experiences. The book does a lot right. The attitudes toward death ring true. The characters ring true, mostly. The plot is interesting, the writing thrilling, the portrait of humanity immensely gratifying, the emotional range fantastic.

A certain extended metaphor, which I won't reveal much about, figures centrally in Passage. But only the symbolic side of the metaphor is revealed at first, and its real-world analogue is hidden. The meaning of the metaphor is the central mystery of the book, and it's a clever one. So far, so good. But there are a few problems with the book.

One is that the metaphor Ms Willis chose is, through no fault of hers, rather overexposed as metaphors go.

But the main problem is that Connie Willis is too much a character in this book. Death is a marvelous mirror; in it we each see our own beliefs about the fundamental nature of the universe. Ms Willis's own views come through quite clearly in Passage, and her portrayal of other views unfortunately doesn't bear the same verisimilitude as the rest of the book. It would have been possible for the protagonist, a scientist, to despise her unscientific counterpart without Ms Willis piling on. Similarly, too much of the suspense in reading the book is due to the question of where the author is going. One wonders whether the book will maintain its hard-nosed scientific attitude to the bitter end; whether the author will ultimately try to give some kind of answer to the question of the experience of death, which hangs over the entire story; and so on.

Still, a hell of an effort by the reigning champ of science fiction novelry.


11 January 2005

A web usability rant

You know who has a surprisingly bad front page on their web site? The NFL.

By my count the front page has three hundred trillion links on it. Some of the links aren't even accessible to me because I don't have a paid subscription; these are conveniently marked with the green-F, white-P icon (this stands for "fmembers-only pcontent".) There are 8 separate boxes for: the top story, "Headlines", "NFL Network", "Today in the NFL", "My Team News and Sites", "Analysis and Opinion", "Video", and "NFL on Sportsline.com".

Suppose you want to see the league standings. Viewing standings is easy from nfl.com. All you have to do is click on "Standings". If you can find it.

How do people actually use this site? When I go to nfl.com on a weekday (when there isn't a game), I'm interested in one of three possible things:

  • last week's games
  • next week's games (what's going to be on TV)
  • league standings
When I go to the site on a game day, I'm interested in one thing only: which games are on TV in my town.

Not only should this stuff be easy to find, it really ought to be just about the only thing on the page. All the NFL news-style content should be clumped in the center, blog-style. Each story should have a headline, photo, teaser, and "read more" link. The newest news should be listed first. That way, it's easy to find new content.

The frustrating thing about all this is that the content on nfl.com is quite good. It has rich statistics on every team and every player in the league, and is updated regularly with fresh news and commentary. But too many people will never know, because the front page is so alienating.

07 January 2005


A typical tsunami is about 1 meter high in open water—and travels at 700 kph.

06 January 2005

05 January 2005

What do you believe

Scientists, mathematicians, and writers were asked, “What do you believe is true even though you cannot prove it?”

Their answers. Moderately interesting.

03 January 2005

Math lessons, here and there

I was blown away by this document: How Culture Shapes Math Instruction in Japan, Germany, and the United States. (1997.)

Researchers videotaped a few hundred real 8th-grade math lessons in three countries. The results are fascinating:

  • “We measured how many times the lesson was interrupted by someone coming into the classroom or an announcement coming over the public address system. This happened during 31% of the American lessons, 13% of the German lessons and none of the Japanese lessons.”
  • American classes spent 62% of their time on homework (either going over the previous day's homework or starting on the next assignment). This number was 40% for Germany and 10% for Japan.
  • In America, 8th-graders learn pre-algebra arithmetic. In Germany and Japan they learn algebra and geometry.
  • “Fifty-four percent of the Japanese lessons included proofs. None of the American lessons included proofs.”

And on and on. But I'll stop here, because this last topic is of particular interest. A proof is simply a step-by-step, defensible logical explanation of why a particular mathematical statement is true. All of higher mathematics is entirely about proofs. Isn't it a little weird that we never teach this to schoolkids?

As I mentioned in a previous post, the first proof I can remember encountering showed that the area of a circle is πr2. I found it in my third-grade textbook. (Here is a rough version of this proof online.) I was fascinated. The teacher never tried to use it in class.

Why do you think American and Japanese classes end up being so different? The document answers that question, too. There are two apparent reasons. First: Japanese teachers care about and labor incessantly to increase lesson quality. They intentionally teach large classes (40 students or more) so that they'll have time to critique and improve their lessons, which they do constantly. In the U.S., individual teachers make their own lesson plans or get them out of textbooks. Classes are small. Teachers spend most of their day teaching classes and, typically, none of it looking back. Second: the goal of a Japanese lesson is to teach the kids math concepts and how to think about math. In the U.S., the goal is to teach the kids how to solve problems. I think this translates to “how to solve the homework problems”, all of which are similar for any given lesson and can be solved without much thought. (If you read the paper, you'll get a much clearer idea of how profound all these differences are.)

The paper argues that the underlying cause of these differences is cultural, which might be true. Or it might be economic or political.

(Found on Isabel's math blog.)