Sunday, October 30, 2005

What is inappropriate about the truth?

A curious controversy cropped up last week in sporting circles when Fisher DeBerry, the coach of Air Force Academy, blamed the lack of African-American athletes in his program for his team's inferior speed. To quote DeBerry:
It just seems to be that way. African-American kids can run very well. That doesn't mean Caucasian kids and other descents can't run, but it's very obvious to me that they run extremely well.
News organizations immediately picked up on the remarks and deemed it newsworthy enough to make it a national headline like this one here. The reports were all careful to not criticize DeBerry directly but the very fact of the story being newsworthy spoke volumes about what they thought. And then, rather predictably, the official reprimand came through, with athletic director Hans Mueh denouncing DeBerry's remarks as "seriously, seriously inappropriate".

Mueh didn't care to elaborate on exactly why the remarks were inappropriate nor who exactly found the comments offensive. Were African-Americans insulted at being praised for their superior athletic prowess, or were white Americans upset at being called less athletic?

Did it matter that the allegedly controversial statements made by DeBerry are actually true and plain obvious to anyone who has watched American sports? Did it matter that barely anyone in the country can name a Caucasian wide receiver in the NFL? Did it matter that no one remembers the last time a white American ran a sprint at the Olympics? Dit it matter that "White men can't jump" is a cliche?

Or did it only matter that the Air Force doesn't like being portrayed as being unfriendly to the minorities?

Sunday, October 23, 2005

The Losing Battle Waged by Law Enforcement

The FCC has recently issued an order requiring ISPs as well as universities, libraries and airport wi-fi providers to ensure, paying out of their own pocket, that their networks can be tapped by the federal government to monitor subjects' e-mail and web access (New York Times story here). The thought process behind the order is pretty clear -- the CALEA act of 1994 effectively enforced the same requirement on telephone service providers and the government figures that what applies to phones should apply equally to VoIP and e-mail. According to the US government, expanding the scope of this act to include the internet is designed to help catch terrorists.

One small problem, though. The terrorists have to be stupid enough to use unencrypted e-mail. Moreover, the implementation of such a tapping system would definitely have the consequence of making encrypted e-mail and VoIP sessions much more pervasive than they are today, especially when it comes to nefarious activities. I wonder if anyone has considered the possibility that the government is better off with today's system, where people don't feel the need to encrypt everything, and the government can still get at many people's e-mail by subpoena-ing ISPs to look at the content in their mail servers.

The root of the problem lies in the fundamental differences between telephone and internet-based systems. Telephone systems were designed for an end-to-end application terminated by humans on both ends -- carrying voice traffic across a wire. The only way to obfuscate communication was for the humans to speak in an invented foreign language that no one else understood, and this was very difficult. (You could also do some basic scrambling with automated devices that could be screwed on to the phone, but they were pretty complicated as well.) On the other hand, the internet is designed for applications terminated by programmable computers at both ends. Individuals can easily protect their communication by the simple expedient of using custom software that runs on both ends to encrypt and decrypt what goes out the wire. There isn't much use in tapping the communication channel if the channels are terminated by modifiable software!

Monday, October 17, 2005

Cricket: More Luddite sentiments

Rudi Koertzen thinks that extended use of technology to make umpiring decisions is a bad idea. In his own words:
We all make mistakes and I think the players actually make more mistakes than the umpires do. So they should leave it up to us to make the mistakes. We've got to live with that.
This argument is not too far away from the other nonsensical arguments one often hears about technology eliminating the "charm" of the game, or robbing it of its "glorious uncertainties". There are some basic facts that appears to elude such luddites:
  • Cricket is a game that involves players scrapping with each other. Umpires are not participants; they are external arbiters whose role it is to make sure that the game is played according to the rules. That umpires are human is a necessary evil created by the lack of technology to automatically adjudicate games.
  • Mistakes made by players are part of the game and is exactly what people pay money to watch; people don't pay money to watch umpires make mistakes. People don't pay money to watch the umpires make mistakes; they do pay money to watch the players make mistakes.
  • Umpiring errors may add to the uncertainties of the game, but there is nothing glorious about it. Cricket is a game of skill involving a combat between batsmen and bowlers/fielders. Umpiring errors are just as extraneous to cricket as the idea that there will be a coin toss after every ball to determine -- by chance -- whether the batsman is out. If the latter idea looks silly, so should the idea that umpiring errors enhance the game.
  • Officials often like to quote some statistic like "The umpires get it right 94% of the time.", as if everyone must stare at the 94% number in awe. The right question to ask is not what fraction of umpiring decisions are correct, but what fraction of umpiring errors are of a game-changing, or career-changing, nature and whether the use of technology to improve accuracy leads to a significant improvement in enabling fair outcomes.
The only valid objection I have found for the use of technology is the amount of time that is wasted in referring decisions to the third umpire. I don't understand why more effort has not been expended in speeding up this process -- giving the third umpire access to replays immediately -- so that the decision can be made in a matter of seconds rather than in minutes.

Thursday, October 13, 2005

The New York Times and Judith Miller

If your sole source of news is the New York Times, you might be justifiably surprised at the news that has had the rest of the world abuzz over the past week. Judith Miller has suddenly gotten a whole lot less self-righteous and a whole lot more cooperative in the Valerie Plame case; she even discovered some rather intriguing notes about a previously unpublicized conversation with Scooter Libby. You could do worse than go over to Mark Kleiman's blog to read about the whole intriguing story. And today's update on the Mousetrap theory.

Of course, if you are a poor NY Times reader, you would be confronted by drivel such as this (registration reqd.), an entire article about Rove and the White House that somehow fails to even mention anything about Judy Miller's latest escapades. Where is the public editor?


Wednesday, October 12, 2005

The blog in defence of grassroots democracy

The biggest recent event in the Indian blogosphere revolves around the IIPM controversy. Amit Varma provides a detailed storyline but the big picture is pretty straightforward: magazine publishes expose on the so-called Indian Institute of Planning and Management, blogger links to said expose, blogger gets threatened by a hilarious ransom note demanding that the link be removed, IIPM threatens blogger's employer with bad publicity, blogger resigns to defend his principles. All very familiar, and all very depressing.

But there was one difference this time: hundreds of reputable blogs with high PageRank that carry the truth and cannot be suppressed. Grassroots democracy has finally arrived.

Tuesday, October 11, 2005

Cricket: Why Ganguly can no longer be one-day captain

I'd been watching the new one-day cricket rules at work over the last few months when something suddenly clicked into my head: the rules effectively imply that Sourav Ganguly cannot stay on as India's captain, even if he were to regain his form as a batsman!

To understand how I arrive at this conclusion, we need to start with the new substitution rule: each team is allowed to replace one of its eleven players with a pre-designated "super-sub" anytime during the match. I have already commented in the past about the flaws introduced by this rule and how they need to be fixed (delay announcement of the eleven until after the toss) but those turn out not to be relevant for what I am discussing today.

The best way for a team to exploit the substitution rule is to use different batting and fielding XIs, replacing a batsman with a bowler between innings when batting first, and vice versa when batting second. In either case, the key observation is that a player from the batting XI should not be present in the fielding XI.

Now, for the sake of argument, let's assume Ganguly is a fantastic batsman and therefore certainly deserves a place in the batting XI. Who else is going to be in this XI?

First, we've got to have at least four specialist bowlers on it since the whole point of substitution would be to have five specialist bowlers in the fielding XI. [ One could argue that the batting XI should just have three specialist bowlers and the fielding XI can make do with four, but this is a bad strategy for three reasons: (a) the fifth bowler has always been a major point of weakness for India and becomes crucial to have when one bowler has a bad day; (b) eight batsmen in the batting XI is way too many; (c) if the toss goes awry and the team finds itself having to bowl first when having a bowling substitute, they are forced to use the substitute up front and are still left with a poor bowling line-up, instead of having a choice between 4 and 5 bowlers that they can exercise depending on how well the first inning is progressing. ]

Throw in a wicketkeeper (Dhoni) and that leaves five batting spots open besides Ganguly. These spots are taken up by automatic selections Tendulkar, Sehwag, Dravid, Kaif and Yuvraj Singh.

Now, which of these eleven can stay out of the fielding line-up? Obviously not the four bowlers or the keeper. Not Tendulkar or Sehwag who are both valuable part-time bowlers; moreover, Tendulkar is India's best boundary fielder while Sehwag is also a competent outfielder as well as a second-slip specialist. Not Dravid, who is a masterful slip fielder and is probably the best strategist on the field as well. Not Yuvraj or Kaif who are the two best fielders in the side. And who does that leave us with? Sourav Ganguly, the worst fielder in the side by a mile.

So, we conclude Ganguly needs to be substituted out when India is bowling. (Or subbed in when batting, depending on which scenario we are looking at.) But what's the point of having a captain who is never on the field? Therefore, it follows logically that Ganguly can no longer be captain even if he deserves his spot in the batting eleven!

Now let's see if the selectors understand this. One can always hope.

Monday, October 03, 2005

Wherefrom the liberal agenda

Nicholas Kristof begins his New York Times op-ed today by stating:
The main mode for seeking a more liberal agenda should be the democratic process, not the undemocratic courts.
Unfortunately, the rest of the article is on Times Select and I haven't found the content worthy enough to pay for. But that shall not hold me back from taking issue with that one flawed statement.

I do not believe a majority vote can ever be relied upon to create fundamental human rights. If the constitution of nations had been drawn up solely by majority vote, I'm willing to bet that slavery would be omnipresent. As recently as fifty years ago, a majority of Americans in many states thought that desegregation was a terrible idea as well. And, of course, children in the USA would be required to learn creationism in science classes, since 55% of Americans want it that way.

The whole point of liberalism, or at least my idea of it, is to endow each individual with the right to pursue their own lives in any manner of their choice, so long as it does not interfere unduly with anyone else's fundamental rights. The democratic process equips the majority with the ability to mandate away the rights of the minority, thus conflicting with the goals of a liberal society. The only way to pursue the liberal agenda is to enshrine these fundamental human rights in the constitution and control the courts interpreting this constitution, until that far-away day when they can rely on an enlightened public. This is also the reason that constitutional amendments are designed to require a significant majority to be approved -- it should not be easy for a partisan majority to rob the minority of these fundamental rights.

Saturday, October 01, 2005

More Movie Reviews

Quick capsule reviews of movies I've seen in the last few weeks:

Fresh: A surprisingly good adventure/drama from the '90s that deserves to be known much better. "Fresh" is the right word, for it is a unique blend of the authentic Brooklyn settings of Spike Lee with the clever double-crosses of Kurosawa's Yojimbo with some characters right out of Scarface thrown in for good measure. The story revolves around Fresh, a smart, twelve-year-old schoolboy in a Brooklyn neighborhood who strives for a normal life while learning chess from his hustler dad (Samuel L. Jackson), squeezing in drug runs into his daily schedule while becoming the right-hand boy of the local kingpin, and protecting his elder sister from the clutches of the big bad men.

The Constant Gardener: Based on the John le Carre novel, this stylish and strongly scripted drama from director Fernando Meirelles is the most intelligent film to have emerged from Hollywood this year. Meirelles combines the soulful melancholy of the le Carre novel with the vibrant earthy portrayal of the third world that characterized his earlier, and superior, effort City of God. The movie eventually runs on too long and its artiness wears thin at times but it feels unfair to quibble in a season full of unworthy trash.

Short Cuts: Yet another of Robert Altman's large-canvas productions, telling the intertwined tales of nearly two dozen people over a few days in L.A. While not all of the stories are compelling, Altman does a fine job of sketching three-dimensional characters and keeping us engrossed. I would rate the movie a notch below his best work in M*A*S*H and The Player but your mileage may vary, depending on how much you care about plot versus characters.

I'll Sleep When I'm Dead: Disappointing British crime drama from Mike Hodges and Clive Owen, the same tandem that brought us the far superior Croupier. There are some good scenes here and there and the story builds a fair bit of atmosphere and portent but fails to go anywhere and collapses in an air of muddled confusion.

Ed Wood: Tim Burton and Johnny Depp team up for yet another home run in this hugely enjoyable biopic of the infamous Edward D. Wood, Jr., acclaimed as the worst director of all time -- and with good reason. Right from the inventive opening credits, the film blurs the line between author and subject by mixing in a surreal atmosphere and overwrought acting that might almost belong in an Ed Wood-directed movie. Almost being the operative word, for the movie also boasts of some excellent performances, most notably by Martin Landau as the fading star Bela Lugosi. Vincent D'Onofrio is also quite good in a cameo as Orson Welles.

Million Dollar Baby: Overrated. Clint Eastwood reprises his usual role as the strong silent type, and Morgan Freeman does his by-now-familiar voiceover commentary that lends an air of majestic solemnity to proceedings, and I must say that the duo do quite a good job of holding up the picture -- with help from the always impressive Hilary Swank -- until about two-thirds into the movie. There were just too many shades of "Unforgiven", too many silly boxing scenes and poor caricatures passing for characters that robbed the film of its authenticity, and it didn't help that the story was robbed of all its suspense by that lunatic Michael Medved. Not that it would have changed things much even if the twist had been a surprise, as I failed to empathize with the emotional complications involved in the story.