Monday, October 02, 2006

I have been reading, still, but I've been blocking on writing for some reason. The first of the recent crop was Vikas Swarup's Q&A. It's a story of a poor Indian waiter who manages to win a "Who Wants to be a Millionaire" style game show. Most of the story is an explanation of how he managed to win by explaining where he learned the answers to each of the questions in a series of flashbacks. I can't say it was a bad book, but it was lacking in a certain necessary something that a good reviewer would be able to articulate. It certainly didn't sugarcoat the brutality of poor India, but there was nonetheless an inappropriate dreamy naïeté about it. It was fairy-tale-ish in a vaguely "Forest Gump" sort of way that no doubt charmed some readers, but to me blunted its impact.

Under the name Iain M. Banks, he writes science fiction (including the excellent "Culture" novels). Under plain old Iain Banks, he writes more literary fiction, including the creepy and excellent The Wasp Factory. I read one of his newer books, The Business. Like Q&A, it wasn't bad, but it wasn't good. It's again a simple story, that of a contemporary career woman at a bit of a crossroads in an ancient, secret merchantile business dating back to the Roman Empire. In that, it has much in common with Charles Stross's The Family Trade. Banks shows off some of his wit, and generally does a good job as a writer. It's as a plotter of stories that he falters. What happens over the course of the book just isn't very interesting.

That is not a comment that one could reasonably make about Felaheen, Jon Courtenay Grimwood's conclusion to his "Arabesk" trilogy, following Pashazade and Effendi. Again, we rejoin ZeeZee/Ashraf in an alternate future North Africa. For the first time, the story leaves Alexandria. Where the other two stories followed Ashraf uncovering the histories of others, in this final volume he discovers where he himself came from, mysteries that were hinted at but not elaborated on in the earlier books. Again, the story is interesting, the world fascinating, and the writing as excellent as before. Now that I've finished the trilogy, I can give it a strong endorsement.

Fans of Neil Gaiman's American Gods will find themselves at home with his more recent Anansi Boys. Fat Charlie's always had things a little rough in his life, with an unpleasant job, a future mother-in-law-from-hell, and a painfully embarassing father. Things get worse when he meets the brother he never knew he had and learns his father was actually an incarnation of the African god Anansi. It's a good enough story, entertaining in many ways. It's not great, though, for reasons that I am again unable to articulate well. It's too breezy, in a lot of ways, which was probably intentional, but in my mind makes it less than the weightier American Gods.

The final book on my recent reading list is Judas Unchained, Peter F. Hamilton's sequel to his Pandora Star. Ignore the goofy titles and the silly cover art: these two books are great science fiction. Sure, Hamilton cheats by adding faster-than-light travel and force fields and other elements "hard SF" writers shun, but it doesn't matter. He manages to stitch together the disparate stories of many players in mankind's discovery of and war against a strange and alien enemy. In lesser hands (*cough* Robert Jordan *cough*), so many viewpoints would be confusing and annoying, but Hamilton is equal to it. His world is imaginative and interesting; he's clearly put a lot of thought into what things would be like in such a setting. That's not to say the books are without flaws, of which I'll mention a few. First, he has a bit of the horny teenaged boy in him. It's not overwhelming, but it's a little much. Another is his use of phrases like "he instructed his e-butler 1 to tell the car's drive array to take him home," instead of simply saying "he told the car to take him home." Once you've established that the character has such a software agent and cars drive themselves, skip the techno-babble. It would be like saying "he used his feet to walk up the stairs," or "he extended his arm, opened his hand, grasped the glass, and retraced his arm to bring the water to his mouth, where he ingested the liquid." Finally, I found the climax to be, well, anti-climactic. The story fizzled out a little towards the end. These flaws are minor, however. The books are excellent, though, and quite a deal, as Hamilton splits 1600+ pages of story over just 2 volumes where he could have easily turned out 5. I highly recommend reading them.

1 An awful term in and of itself.

( books )

I miss spinach.

( food )

Tuesday, October 03, 2006

The torture bill that just passed is one of the most reprehensible actions Congress has ever taken. There is so much wrong with it at every level. I'll just cover a few highlights of why torture is such a terrible thing, with assistance from this surprisingly intelligent Something Awful thread.

Bush & Co. argue that nobody can define precisely the boundaries of humane treatment, so there's no point in such guidelines, but does anyone doubt that ripping off people's nipples with pliers is torture? Somehow we have a firm enough definition to know that. They misleadingly say that since there is some grey area, it's all grey area. They also claim that what they're doing "isn't that bad." If that was true, why would they do it? It's a waste of energy and time to try things that don't work, not to mention just plain mean. If is effective, then it really is that bad. What would make you reveal information that could result in the capture and death of friends and relatives? Just because what they're doing sounds mild does not mean it actually is mild. It's likely that the methods they have disclosed were chosen particularly because they sound less innocuous than pliers and hot irons while still being cruel and devastatingly effective.

It's also not just about any particular method in isolation, but the totality of circumstances. There's a huge difference between reading a description of waterboarding or even having it demonstrated on you once in a safe, controlled environment, and having it inflicted on you over and over again by people who wouldn't mind if you died in a gulag from which you have no chance of escape. Certainly that provides a motivation for telling the truth.

But then, how do the torturers know you're telling the truth? If they have enough information to know what is fact and what is fiction, then the torture is pointless in all but the most extreme and contrived cases. If they don't have enough information to distinguish, they could stop before they get the real truth or continue long after. What if their corroborating information is wrong, and the victim is telling a truth that disagrees with it? The goal of torture is to break the victim, but once the victim is broken, their attachment to their cause severed, and their will made subservient to the torturers, it's ludicrous to think they could have any attachment to the truth. Torture doesn't work any better than slower, less cruel methods. More effective and less cruel methods focus on psychological manipulation that builds trust and makes the subject want to help you. And no, "make the pain stop!" is not the same thing as wanting to help you. That is the key practical issue with torture. It provides you with little that is useful while destroying something of great value.

What torture does is turn a human into a savage beast. When a torture victim is "broken," you reduce them to a sub-rational animal state of helpless terror, where the world reduces to a simple binary choice: say this and the pain continues, or say that and the pain stops. Whether there is any truth in it back from when you were still a rational human being is immaterial. Whether the interrogators promised to keep you from suffering permanent injury or dying is irrelevant. Even if you even remember what you were before or what promises you heard, you're not likely to believe it or be able to think rationally when your world has shrunk to hold only agony. That is the fundamental immorality of torture, that its effect, indeed, its goal, is to take a human being and reduce him to a cowering, fearful, pathetic animal.

The effect isn't to make you tell the truth, but to make you do whatever it takes to make it stop. There is effectively no difference between breaking your will to keep secrets and just plain breaking you and turning you into the torturer's eager slave. As a means of obtaining information, torture is almost always ineffective in comparison to the gentler methods, and you can never know until after the fact whether it was the rare case where it was more effective. As a means of obtaining reliable information, it's useless.

That doesn't mean that torture is useless in general, though. It's great for getting people to say what you want and to confess to whatever you want them to. That's why Stalin and Pol Pot and an endless list of murderous sadists liked it so much; they weren't after information or justice but instead wanted a confession to justify their actions, no matter how valid or true the confession might be. It's also effective as an intimidation tactic aimed at anyone who might cross you. In theory, that's not us, but bills like this recent one and the apparent inability of our government to listen to people with experience in these matters tells me that we're not so much better after all. Among a host of tragedies, that is a final one.

Using torture robs us of the most essential weapon we have in our battles against Al Qaeda and the like: the moral high ground. America has made plenty of mistakes, but we've always been distinguished from other nations in that our defining essence is based on principle rather than geography, ethnicity, or other accidents of history. What makes Al Qaeda and the Iraqi insurgency and the Chechens and all of those terrorists so very wrong isn't what they want, but how they're willing to hurt and kill innocents to achieve those ends. We bind ourselves with limitations because we understand that is what separates right causes from wrong ones, not the goals. What is fundamentally American is that a nation ought to be founded on freedom and justice. Torture ought to be beneath us, because we should be both too smart and too good to do it. And yet, here we are.

( issues )

Wednesday, October 04, 2006

They say everyone breaks; it's just a matter of time. I don't know if that's true, but I'm willing to accept it. On the face of it, that suggests the best you can do is delay, but you can never win. It turns out, however, that you can use that against your captors.

Let's suppose you are captured by those who would torture you to get information from you. You must assume you will break. Once you know that, you can use it against your captors. Well before you actually break, you start telling them things. You tell them things that are completely true, things that are mostly true, and things that are fabrications. You tell isolated facts and whole stories and everything in between. You spew a barrage of information, within which you bury the truth. They'll be able to verify some of it and disprove some of it, but most of it will be ambiguous. After all, if they knew that much there would be no need to torture you.

Your goal is to prevent them from learning the truth so they can defeat your cause. What if they hear the truth, but consider it a falsehood, or if they hear a lie and call it truth? In that case, you have succeeded, even if at some point the truth came out of your mouth. This is even something you can train for, in much the same way an improvisational comedy troupe practices. If I can think of this, I'm sure Al Qaeda can too. In fact, knowing that your victim is using this strategy would help you not one bit. By overloading them with false information, you make yourself useless as an intelligence source, even if you know lots of true and useful information, even if you tell it to them, and even if they break you.

Furthermore, you can do more than just make your interrogation pointless. If the information you give them sounds reliable, but clashes with other intelligence, you introduce uncertainty and destroy the value of those other sources. This will happen regardless of whether you're telling the truth or a lie. Your torturers will know well the limits of torture, but their bosses might not. You might not be able to save yourself, but you can protect your cause.

( issues )

My idea for an anti-cheating business is so ground-breaking and visionary that someone else has been doing it for years. I even was dimly aware of them because they have very badly-behaved web spiders that I've banned.

( ideas )

A week or so back, The Christian Science Monitor reported:

Today's conventional hybrids command a premium price - $2,000 to $4,000 more than their nonhybrid counterparts - and their owners will recover that extra cost in about three years, assuming $3-a-gallon gasoline and 12,000 miles a year of driving, the report found.
Those numbers sounded suspicious to me, especially after I did the math myself, so I obtained the original report (bug ye not). What it actually says is (on page 15 of the PDF, page 9 of the report):
With a gasoline price of $3 per gallon, fuel for a 30-mile-per-gallon conventional vehicle driven 12,000 miles costs $1,200. A hybrid achieving 50 miles per gallon and driven the same amount uses $720 per year on gasoline.... the ... payback period for the hybrid relative to the conventional vehicle is just over seven years.
Seven years now? Wow. That's a big disparity. How did they screw that up? Simple. Reading further down the page, it says:
With battry costs at the long-term levels in Table 1, however, the picture is quite different. Assume that ... incremental costs for ... a hybrid relative to a conventional vehicle are $1,500, declining to $1,000 ion the long term... which assumes high-volume battery costs of $400 for the hybrid.... results in [a] payback period of 2.9 years.
The "long-term levels" they refer to are the cost of batteries if they are produced in much higher quantities than today. The CSM reporter took completely hypothetical cost and rate-of-return estimates and presented them as facts.

My point here isn't so much about the efficiency of hybrids as it is about bad reporting. The mistake was instantly obvious to me, and it took barely any time to prove it (most of the time was spent creating a fake account for the ACEEE site). And yet, a key fact presented in the article was still wrong and made press not only in the CSM, but as a reprint in Yahoo News and possibly other publications.

The moral is to find the primary sources when you can. Of course, the media enjoy too much their role as mediators to make that easy. Given how often they make mistakes, though, it's kind of necessary. I think this is part of why people feel disillusioned with scientists. A paper will appear in a journal describing how daily injections of a particular substance into a genetically-modified strain of mice caused tumors of the spleen to spontaneously reduce in size 38% of the time, which will turn into the headline "Cancer Cure Discovered!" This will happen in politics, too, where a bill that grants the President the power to arbitrarily detain and torture anyone he wants is called a "compromise." But I've posted enough about that subject for now, so I'll stop.

( oil | media | science! )

Thursday, October 05, 2006

I think a lot of people choose to have kids too late. I know many people who had their first child in their early to mid 30s. That does not seem optimal to me, and not just because Uma was born 4 days before my 27th birthday. Consider how old I will be when Uma is at the normal college graduation age: 49. That's pretty young. Suppose Uma gets married at roughly the standard age. I'll be in my 50s. That's good, too. Think ahead to the far future where prospective grandchildren do the same things, assuming it's at roughly the same ages. I'll be 76 when my first grandchild graduates college, and about 80 at the wedding. If, instead of 27, the first child happened at 32, I'd be 86 and 90 respectively when those things happened, which calls into question whether I'd even be alive. And that's just the first child. Most people wait a couple of years at least between children. Then consider how it can take time to conceive as well as the scarily-high miscarriage rate. To the extent that one can even guess things like that, I have a pretty good chance of being around and being healthy until Uma is well-established on her own 1 .

The other major factor is the age of the mother. Pregnancy and health issues of various sorts are much more frequent once the mother is 35. If you want two kids, you have to have the first one in plenty of time for the second one to be born before Mom is 35. What if you decide you like having two kids so much that you want a third? If you started at 32, you have some tough choices to make. There's also the issue of women who stay home while their kids are young. You can certainly get further along in your career before having kids by delaying, but that means you'll be even older when it comes to returning to the work force, which can be rather intimidating.

I'm not saying that everyone should do this. I'm just suggesting that people who have a choice examine the consequences more closely. People who aren't financially stable ought to wait. So are people who aren't in a very stable relationship 2 . Ditto for people who haven't finished their education. There are any number of really good reasons for delaying, but you should make sure you've thought through the consequences, including the options it forecloses. Having a child is easier at a younger age, and when you have your children has ripple effects through all of your life and all of theirs. Delaying has clear advantages for financial stability and getting out your last ya-yas as a footloose and fancy-free youngster, but it comes at no small cost.

1 Not that we're ever going to let her leave.
2 I laugh (sadly) at people who think they should have a child to stabilize their relationship. Few things pull at you like caring for a child. It pulls you up, certainly, but it can drag you down, too. If your relationship is already on the edge, a few rough nights could easily push it over.

( deep thoughts )

Monday, October 09, 2006

It's no fun denying water to your child when she hasn't had anything to drink (that she's kept down) for 6 hours. She doesn't understand pacing yourself; she just knows that she's thirsty and Daddy won't let her have any water.

To the list of reasons why you shouldn't let your young child watch television, add that it gives you something to break out when you get absolutely desperate, and, since it's completely novel to her, it doesn't really matter what you watch. Uma does know how to identify the important parts of Mystery: the baby ("baow"), the dog ("woof woof"), and the car ("vroom").

( us )

Thursday, October 12, 2006
New Cuyama, California, would like you to know certain facts about their town, especially the total.

( fyi | funny )

I came upon this a long, long time ago, and just recently rediscovered it. If you haven't read it before, read it now: things my girlfriend and I have argued about.

( funny )