Some Charity

One of the most transparently self-serving assumptions often made by laissez-faire advocates is that private charity can do a better job than government of providing assistance to the less fortunate members of society, and that such charity will be unleashed if only government would get out of the way. I offer the following example of what really tends to happen, from the first chapter of Laurie Garrett’s Betrayal of Trust where she describes a 1994 plague outbreak.

What drew industry to Surat was precisely the weakness of its government, lack of health and pollution enforcement, eager unskilled labor force, and a virtual tax-free environment.

Sounds pretty much like what some would call perfect conditions for success, doesn’t it? Not according to the very next paragraph.

“Perhaps the greatest irony,” wrote the conservative Business Standard of Bombay, “is that the epidemic has hit one of the economically most active areas of the country in a state which is considered to be the most business friendly…What is more, the Gujarat government has gone out of its way to be more accommodating to business than most and has in turn been able to reap the benefits of a rapid industrialization which is not the case with the rest of the country. But somehow down the line, the need for good municipal services was forgotten. Businessmen who were busy making money cared little about minimum civic services or the basic quality of life that says no filth, mosquitoes, flies, fleas, and rats. And when the epidemic hit, they were the first to pack their Maruti 1000s and run. India today has clearly got its priorities wrong.”

Now here’s the kicker.

GOP Priorities

This cartoon (also available locally for when the previous link inevitably goes stale) pretty neatly wraps up two current political issues.

How False Memories Form

For a while now, the issue of “false memories” – i.e. those that are the result of suggestion by more or less well-meaning therapists or investigators rather than of actual experience – and the witch-hunts they cause have been an issue for me. I highly recommend Ofshe and Watters’s Making Monsters: False Memories, Psychotherapy, and Sexual Hysteria for background on how the whole thing works both scientifically and as a revenue generator for psycho therapists (that space was put there intentionally). Now there seems to be an explanation at Science Blog that sheds a little more light on what’s going on inside people’s brains when they’re remembering things that didn’t happen.

”We think parts of the brain used to actually perceive an object and to imagine an object overlap,” said Paller. ”Thus, a vividly imagined event can leave a memory trace in the brain that’s very similar to that of an experienced event. When memories are stored for perceived or imagined objects, some of the same brain areas are involved.”

This fits quite well with the earlier observation (mostly credited to Elizabeth Loftus) that the vividness of a memory is not well correlated with its accuracy. In other words, under hypnosis you might become more certain that you can remember every digit of that license plate, but in fact the number you give the police is no more likely to be correct than it was before. The new finding also fits in well with what I’ve come to think of as the most convincing theory about deja vu – that the brain “tags” information according to whether it’s sensory input or recall, and deja vu is simply a matter of the former being mis-tagged as the latter.

One interesting implication of all this is whether memories related to senses other than sight are similarly affected. Are memories that involve sound or smell less likely to be false? Are sightless people less susceptible to false memories? Maybe the researchers will consider some of those questions in their next study.

Moved Over

Yes, folks, this is the new WordPress version of the site. I had to do some hacking to get a couple of things the way I wanted them (notably the comment display on the right) and I’ve recategorized a bunch of posts, but the content is pretty much the same and the layout is very similar so you shouldn’t see too much of a difference. The old version of the site is still available, but I won’t be updating it.

Colorful Amy

It has been a while since I posted new Amy pictures. Here are a couple that Cindy took, which we’ve used a lot for printing out and giving to people.

Amy in Chair I Amy in Chair II

In other news, Amy has discovered her thumb. For a while now she has been sucking on her (and other people’s) fingers, but in the last few days she seems to have figured out how to get her thumb out from the rest of her fingers and suck on that instead.

Killer Query

I’ve discovered what must be a bug in MySQL. A certain kind of query which should be relatively innocuous can cause the database to spin off into oblivion, ignoring query limits and client disconnections and generally sucking up all CPU time on the system until it’s killed. I happened to notice this while trying to hack pMachine to allow multiple categories for a single post. Here’s how I know it’s a bug.

  • The query was not malformed. It’s well known that joining two tables in certain ways without a relational constraint can cause the result to contain as many rows as the product of the two tables’ row counts, which could add up to a lot of rows, but this query did in fact have a relational constraint and variants that differ in no essential details do return the expected (small) number of rows.
  • Even if there had been no relational constraint, the tables in question only had about 800 rows apiece. That would have resulted in “only” about 640,000 result rows, which should tie up the server for no more than a minute or two.
  • MySQL implements query limits to address exactly the “dot product” problem mentioned above. The limits are deliberately set, both by default and by any sane administrator, to values that would also limit the query to a couple of minutes, but they apparently didn’t kick in.
  • Both PHP and Apache also have request timeouts. If those timeouts fire, the database connection the request is using should be dropped, which in turn should cause MySQL to abort any queries issued on that connection within a couple of minutes. Again, this didn’t seem to happen.

What we have, then, is a query that gets MySQL into a state where it hogs the CPU, ignoring both query limits and loss of client connections. I’ve been able to reproduce this scenario several times on my machine at home, but whether it can also happen at a commercial web-hosting service is still a matter for speculation. I will neither confirm nor deny rumors that the bug in question might have something to do with this site being unavailable for much of the day.

Bush’s Bulge

I can’t quite make up my mind about whether Bush was wired during the debates but, after seeing these and reading this, it’s clear that the president was wearing some sort of device. What? The only semi-credible alternative theory I’ve heard is that it’s some sort of medical device, but that raises a couple of new questions. What kind of medical device? Don’t the American people have a right to know about any medical condition the president has that would require such a device? After all, the health of the president does have a direct relationship to how well he can perform his job, and there’s ample precedent for it being treated as a legitimate issue. Also, the medical-device scenario doesn’t explain the president’s odd behavior – sudden pauses in the middle of sentences, responses that seem ill matched to what Kerry or the moderator was saying, and so on – nearly as well as the wireless-prompter scenario. The theory that he was getting help from off-stage during the debates is far from proven, but it’s the theory that best fits observable facts.

Take a Hint

I just read, and confirmed via multiple sources, that the guy who played Christ in Mel Gibson’s Passion was hit by lightning while the film was being shot. Twice. Now, what kind of person would not consider that as possible evidence of divine displeasure?

Malthus at Work

Apparently many people have been noticing that the squirrel population in Massachusetts and possibly throughout New England has been particularly high this year, often having noticed an abundance of squirrel roadkill or a lack of acorns first. I’ve certainly noticed – and welcome – the latter myself. It looks like we might be headed for a repeat of the 1968 squirrel migration. In that case, there was a bumper crop of acorns and other squirrel food one year, followed by a lean year. The squirrels from the first year responded by producing more young than usual, and then those young found themselves having to forage further than usual for a much smaller per-squirrel food supply. This led them to cross roads etc. more frequently, leading to the increase in roadkill. It’s a classic biological boom/bust cycle, taking place – literally – right in our back yard. We should be able to predict time-delayed booms and busts in the populations of squirrel predators (e.g. foxes) and scavengers such as crows in the near future.

New Champion

For years, “Tit for Tat” (or close relatives such as “Tit for Two Tats”) has been the most successful strategy in the game-theoretic competition known as the Iterated Prisoners’ Dilemma. Now, apparently, a group at Southampton in the UK has devised a group-oriented strategy that can defeat TfT in a standard sort of simulated competition. This might seem like mind-numbingly boring news of interest only to math geeks unless you happen to know that study of this problem was a critical part of developing the Mutually Assured Destruction strategy of nuclear deterrence and also has significant implications for economics. Robert Axelrod even wrote a couple of books about the IPD, covering these and other implications.

The Southampton success actually seems a little bit questionable to me, though. It’s not that the programs maintained history about past interactions, because that’s what the IPD is all about as opposed to the single-shot PD. It’s not that the programs “colluded” because I actually think it’s rather clever. My concern is with the fact that the Southampton strategy’s success might not be generalizable. While nobody has yet succeeded in devising an anti-TfT strategy that would not only win itself but also cause TfT’s performance to degrade relative to the overall average, it’s easy to imagine an anti-Southampton strategy that would have such an effect. What makes TfT interesting is not that it succeeds against naive opponents who know nothing about it, but that it can still succeed (i.e. perform better than average) even in the presence of determined opponents like Southampton that know everything about it. In other words, much as Deeper Blue was (IMO) created to be a dedicated Kasparov-beater without actually managing to unseat him as the world’s best chess player, Southampton seems to have succeeded in becoming a TfT-beater without actually being an optimal or even robust IPD strategy. It’s an interesting result, but it remains an anomaly and doesn’t really affect the various lessons that have been drawn from studying TfT’s success.