Friday, January 14, 2005

adversarial press

Donald Crankshaw on scandals:
At the least, I am glad to see someone defending Mr. Williams. You can expect liberals to go after any conservative they dig up some dirt on, but when one of their own is criticized, even when he does something appalling, they circle the wagons and defend him to the death. With most of the mainstream media on their side, they can get away with it. Conservatives are among the first to turn on one of their own. This can be both good and bad--there needs to be a bit of internal discipline and self-policing in a movement--but mostly its just a response to their relationship with the media. A liberal can defend a fellow liberal who's done wrong without being tarred with the wrongdoing, but a conservative defending another conservative always suffers for it. (Incidentally, the roles are reversed when it's a situation of ideas rather than ethics. Conservatives are much more tolerant with heretical ideas, while liberals can be quite dogmatic.)
I had never thought about it before, but he's right about the other side of the liberal press bias. I've always said that the only reason conservatives are less nasty than liberals is because the liberal-controlled press would crucify any conservative who even once gets as over-the-top as Ted Kennedy does on a regular basis. I guess the same thing can be said for why conservatives are so good at policing their own.

It goes to show how valuable it can be to have a genuinely adversarial press. Too bad the real press is only adversarial to one side...

Cokey goes flying

This story is a tribute to my favorite childhood author, Jim Kjelgaard and a little dog named Freddie who may have lived through something like this, and a little dog named Cokey who I loved as a child (the name is pronounced "coke" + "ee").

Cokey lay under the piano bench in the perfect semblance of sleep listening to the sounds of a busy morning. He was hardly more than ball of black curly hair. The kids were shouting as they searched for their coats and mittens and lunch boxes. Mom was sternly trying to marshal them all together. Dad was rustling the paper at the breakfast table. From his lair beneath the piano bench, the little Pekinese/terrier mix did not have to open his eyes to know what was going on. He knew this routine by heart.

This morning was especially busy. He could smell the breakfast. Sugary cereal and milk. Some toast. No bacon or sausage. Nothing worth begging for.

So he waited. He waited patiently, as his wild ancestors had waited for prey. Because today was especially busy. And because busy people created opportunities. Opportunities a clever little dog could take advantage of. If he was ready. If he was out of sight/out of mind. If no one thought of him. So Cokey waited. And pretended to sleep.

After a few minutes the cacophony raised in volume. It was the leaving time. Cokey didn't know where Dad and the kids went during the day, but he knew they all left in the morning. And it wasn't a simple exodus: open door, leave, close door. Not on especially busy days. Not when the door was already open and Ronnie was still shouting about his mitts. Not when Tracy stepped out the door and then her voice raised in sudden alarm as she realized she had forgotten something. Not when James mocked her for her forgetfulness and she pushed him in retaliation. Not when mom shouted at them to quit fighting. Not when Tracy ran back inside and left the door wide open. That's when Cokey made his move.

In a flash, the little ball of hair transformed into a fully awake sprinter. As he went from the carpet onto the linoleum he had to make a 90-degree turn. His little legs slipped and churned like a cartoon dog as he searched for purchase on the waxed surface. A moment later he got traction and shot through the door and into the garage. There were obstacles on both sides of the station wagon --obstacles that would be Family in other circumstances, but right now they were obstacles-- so he went under the car.


He heard James shout, but he was already out from under the car and sprinting out of the garage.

"Cokey got out again!" shouted James.

"Cokey! It's too cold!" shouted Ronnie.

But Cokey ignored them all. He was free! He was Outdoors!

Cokey ran for a half a block in case they might try to chase him but there was no pursuit. They were late for school.

"He'll be back before dark." Dad assured them. "Until then, if he gets cold, it's his own fault."

Dad hustled them into the station wagon and off they went.

Cokey was panting with exertion and excitement. The air was chill air but exhilarating. It had been weeks since he'd been out front because the Family seldom took him for walks. He had a big fenced-in back yard, but that was almost more frustrating that pleasant --all those sounds and smells that he could detect but not investigate. He wanted to go, not just experience.

The yards were covered in snow, some of it too deep for the little dog's short legs to handle easily, but the road was clear. And all the fun stuff was along the road anyway. That's where other people took their dogs for walks. That's where the marks would be.

The first tree Cokey investigated had been recently marked by Jed, the foxhound mix next door. Not interesting, but Cokey marked over it anyway. Can't let Jed forget who's boss. The nearby mailbox was more entertaining. Fred the basset hound from down the street, and another dog that he had smelled before but never met. Cokey sniffed enthusiastically to see if there was anything else. Nothing he could detect. Maybe this would be a good opportunity to track down that new dog and intimidate him. This was Cokey's turf, after all --when he could get out. Cokey was one of the smaller dogs in the neighborhood, but with his aggressiveness and indomitable spirit he ruled the other dogs --when he could get out.

Over the next few hours Cokey enjoyed himself wandering the neighborhood and marking his territory. He hadn't yet found other dogs to bully, but there would be other days. Now, he was tired and cold and thirsty. All that marking takes it out of you. The pads of his feet were hurting from the snow and the cold, hard street wear. Cokey just wasn't used to this much activity. He decided to head home where he knew Mom would let him in and give him some fresh water. As Cokey trotted along the sidewalk, he didn't notice the sinister shadow floating over his own.

The eagle had been watching the little dog for several minutes now. Normally the old eagle wouldn't have seriously considered Cokey. The little dog was the wrong color for one thing. And he didn't behave like the eagle's normal food animals. Cokey walked with confidence, stopping often to satisfy his curiosity. When a rabbit sees something odd, it will stop and examine from a distance, back on its toes, ready to flee.

Cokey was not like that. The little dog's ancestors were hunters who survived by investigating movement, by seeking out the curious and eating it. Cokey would head right towards anything unusual. Not the sort of animal that the old eagle would normally have hunted.

But the eagle was tired and hungry. It had seen only one rabbit all day, and that rabbit had bolted into its hole before the eagle even got close. So the eagle made its decision. It circled until it was behind the unsuspecting little dog. It stooped.

Cokey didn't know what hit him. One moment he was stopping to sniff a fire hydrant and the next moment he was flat on his belly, stunned, with powerful fingers clutching him. Powerful and sharp fingers. He might have been able to fight, but in the back of his mind he thought that maybe Dad had tracked him down and caught him. Sometimes Dad got angry when he got out. So Cokey made the mistake of lying there submissively as the eagle gathered itself and took flight --the little dog clutched in its cruel talons.

Once he got airborne, Cokey had no intention of struggling. Maybe it was some saving instinct, or maybe just the experience of being carried by humans, but Cokey knew that struggling might get him dropped and that he really didn't want to be dropped. This time the passivity saved his life, not only because the eagle didn't drop him, but because it made the eagle think he was dead.

The eagle flapped laboriously to an area where the ground became too rough to build. There was a hill here with abrupt rock crags. The eagle landed atop one of the crags to devour its prey. Since Cokey hadn't struggled the eagle didn't pierced his spine with its powerful beak immediately. Instead, the old bird stepped off the little dog with one foot to consider where to start eating.

Cokey took advantage of the relaxed grip to erupt like a little hairy explosion. With a vicious snarl he snapped at the eagle's leg. His jaws closed like a steel trap and he jerked his head savagely as if he were worrying a rat. The eagle was taken completely by surprise and it leapt backward with wings flapping violently. The buffeting confused Cokey enough to make him let go.

The eagle jumped and half-flew a few feet away to consider its prey at leisure. But the prey didn't cooperate. Cokey charged the eagle as fast as his little legs could carry him and the eagle flew away again, this time to a nearby crag.

That didn't stop Cokey's revenge. The little dog quickly scrambled, slid and jumped down from the rock where the eagle had left him and charged over to the eagle's new perch as the old bird stood on one foot, watching haughtily. The little ground creature had surprised it, no doubt. The eagle's leg was severely injured. But the creature was much smaller than the eagle and the bird knew it could overwhelm the little dog with a back-breaking stoop, so it decided to wait until the prey had turned its back.

The prey didn't turn its back though. It was trying to climb the rock to where the eagle was. Impossible! But the little dog had already leapt half-way up. It was scrabbling for purchase, it's little legs were whirling at a furious rate, but it still had breath to snarl and bark continuously. It found a small ledge, it leapt upward again. The eagle, startled, spread its wings and opened its mouth to look even more fearsome, but the little dog only barked more furiously.

Cokey's ancient ancestors may have been hunters, but Cokey's recent ancestors had been guardians. Hunters play the odds. They only go after another animal if their chance of eating is much, much better than their chance of being eaten. After all, to reproduce they have to eat many times, but to fail to reproduce, they only have to be eaten once.

But Cokey's ancestors were bred from prehistoric times to be guardians. Not for their fighting so much as for their aggressiveness, their willingness to go after anyone. They didn't have to win the fight, they only had to give the Family time to wake up and get their weapons. All those centuries of breeding were focused right now in the Cokey's determination to get at the eagle. The invader. The Threat.

The eagle didn't know all of this of course. All it knew was that this obnoxious little animal had chased it away once and was now defying the laws of land-bound creatures to chase it away again. As Cokey came within range of the eagle's fearsome break, the eagle struck viciously at the little dog's eye. But Cokey's more recent ancestors had been bred for fighting rats. Cokey had hard bony ridges over his eyes and an instinct to protect eyes from sudden attacks. He ducked. The eagle tore a great gob of flesh from the dog's eyebrow and made him yelp in pain, but it damaged his eye only a little.

The eagle thought that ought to settle matters. Cokey didn't. The little dog finally managed to get traction with his back legs again, and then he was leaping for the giant bird. The eagle panicked and took off, half falling from the crag as it tried to escape. It left a wad of tail feathers in the little dog's mouth.

The eagle laboriously gained altitude as the impudent little dog stood on his crag and barked at him. Hurling challenges and insults. The eagle was a hunter, not a guardian. It needed to have a much higher probability of being eater than eaten and the little dog convinced it that in this case the odds were not so much in its favor. It flew off, seeking safer prey.

Cokey barked the receding eagle out of sight. Since the little dog had poor eyesight, that was not very far. Once he could no longer see the eagle he stayed a little while longer, staring after it and huffing occasionally. Eventually he gave up on it and snorted contemptuously. He'd fought the toughest dogs in the neighborhood, and that overgrown pigeon thought it could frighten him. He only wished it would come back for another round.

The little dog began to look for a way down from the crag. Now that the excitement was over, it looked a lot higher. And the blood from his eyebrow was leaking into one eye. That ruined his depth perception and made the trip down even harder.

He scrabbled down a ways and then was blocked and tried to back up, but he couldn't because it was too steep. He tried to turn around but couldn't do that either. He was truly stuck and it was to high to jump. The little dog considered his options. He sniffed the rock because sniffing is always good, but it didn't really help. Then he crouched to jump from the high shelf, but decided not to. He tried to back up again. No way. Then with a sudden leap he was airborne.

Cokey hid the ground with a painful jar. The deep snow broke his fall, so he had no broken legs, but his abused feet hurt even more now. He huffed in defiance and started leaping through the snow toward home. One good thing about they eye, it would get him a ride to the vet. He didn't especially like the vet, but he loved car rides.

Thursday, January 13, 2005

does HIV cause AIDS?

On Dean's World there is a long series of posts bout AIDS and HIV. You can reach them all from the first one.

I've read some of this stuff before. It's one of those painful issues where your logic takes you one way and your common sense takes you the other way. There is an enormously compelling case that HIV does not cause AIDS. Yet I'm not a physician and I don't have the expertise to go against the entire medical establishment. Yet some of the people who make the case are physicians and do have impeccable credentials. Yet they are a tiny minority. Can the entire medical establishment be so wrong and so stubborn for so long? It seems beyond credibility. Yet the evidence is very strong. What should you believe in these sorts of cases?

I was just remembering when the HIV virus was first discovered and how odd I thought the events were at the time. Some guy announced that he had discovered a virus that might be related to AIDS. There were lots of published cautions about how much research had to be done to confirm it. Then someone else came out and said he had discovered the virus too. Then there were various articles about how the AIDS virus had been discovered, interspersed with a few cautions that the link had not been confirmed. Quite quickly, the caution articles went away and everyone agreed that HIV causes AIDS, but I never saw an article about the successful confirmation.

At the time, I assumed that I must have missed the articles about the confirmation or that I didn't understand what a confirmation would take. But it seems that some people who really know their stuff are saying that there never was a confirmation.

Wednesday, January 12, 2005

Doc Rampage -- the most metaheroic blog?

I've kicked off a marketing campaign to have Doc Rampage considered the most metaheroic blog in the world. The funding for the campaign is a bit, ah, sparse, but I've already managed to get out the first installment over at Back of the Envelope.

Donald Crankshaw did a great job and he works cheap. I mean really, really cheap. He actually dredged up an old post from my first week or so of blogging where I defined what a metahero is. He also defended my title to the term (and on unimpeachable grounds). Overall, an impressive effort of marketing on a shoestring. Thanks, Donald.

If anyone else would like to be a paid blogger and is willing to work really, really cheap, send me an email and we can work something out. Just think of the advantages you would have in blog disputes. You would get in an argument with some other blogger and you could say, "What do you know, you amateur? I'm a pro! People pay for my blogging!"

who links here?

I just found a great new (well, new to me) blog tool over at Mostly Cajun. It's called "Who Links Here", and it finds lots of people who link to you.

I used it to update my blogroll. Not just people who have blogrolled me, but people that I've been meaning to blogroll and haven't yet.

I've never been entirely comfortable emailing people to let them know I've blogrolled them. On the one hand, they probably want to know (unless they are huge blogs), on the other hand, it kind of sounds silly to just write, "Uh, you don't know who I am, but I've added you to my blogroll." Now they can use this tool to find out for themselves.

dog vs. eagle

Hmm. A Pekinese is snatched by an eagle then gets away and finds his way home in a deadly cold winter. Sounds a little farfetched. I'd guess the the vet who came up with the theory was a Jim Kjelgaard fan.

Monday, January 10, 2005

vacation photos

I know everyone is dying to see my vacation photos. Here are a few of them. All are taken on Kawaii. I'd give specific locations but that would require work. The rocks by the side of the road are the ones I dodged on my way to the cancelled boat. I took the picture on the way back. The chickens are just some random wild chickens.

what if they had been good forgeries?

The Rathergate report has restarted discussion of the 60 Minutes segment where Dan Rather exploited forged documents to accuse Bush of youthful indiscretions. Even though these documents have no chain of custody, there is no reference to any of these documents anywhere, the documents are full of physical and textual errors that show they were not produced at the time and by the people that they say they were, yet the report still refuses to say that the documents are forgeries. What does it take to get a document declared a forgery?

Dan Rather has even had the audacity to say that no one has proven that the documents are forgeries. Hey, Dan, since when does someone have to prove that documents are forgeries when there is no evidence at all to suggest that they are genuine? Does anyone believe that Dan Rather or the others who are so uncertain about the status of these documents would have any doubt about them if they helped George Bush? Or would Dan be on the air every day in his best outraged tone, accusing George Bush personally trying to effect an election through fraudulent means?

But what if the documents had been good forgeries? CBS demonstrated that if you want to prove something bad about a Republican, all you need is a word processor with a typewriter font and a little bit of text research. You don't need to find old or authentic paper --they'll accept copies. You don't need an actual typewriter --they'll accept documents that have been faxed and copied so much that typographical analysis is impossible. You don't need access to the location where the documents would normally be stored --they'll accept unsubstantiated stories about "personal files". You don't need to make up a legitimate-looking set of related files --CBS doesn't need to see anything except the key documents. You don't need to have any involvement in the organization that the documents came from --they'll accept that you got the documents from a dead man that you can't prove you ever even met. CBS, in short, is begging to be scammed, as long as the scam tells them what they want to hear.

Ultimately, this isn't about evidence or truth, it's about power. The evidence is just a means to an end, and if it doesn't serve that end, it is discarded. That is what makes this incident so frightening: CBS and their many defenders have shown that when it comes to attacking a Republican, they don't care to apply even the most minimal standards of evidence. An accusation is as good as proof, and if the accusation is repeated in the mainstream media, that's as good as a conviction. If those documents had been even moderately well-forged, it would be settled fact today that George Bush refused an order to get a physical and was given special treatment in the Guard. As it is, those ideas are considered settled fact by half the country, even though there is no evidence of it except for some fraudulent documents and a few self-serving accounts by Bush's political enemies.

Memogate isn't the only example of this in the past few years. Democrats still claim Bush was lying about African uranium, even though the man who made the accusations has been thoroughly discredited. Democrats claim that Bush was behind the Swiftboat Vets even though there was never any evidence of this. Democrats still claim that Bush was behind some sleazy tactics against McCain in the 2000 primaries even though there is actual evidence that someone else did it. Democrats still claim that Bush arranged to have a plastic turkey at the Thanksgiving in Iraq, even though there was never any evidence for it and it has been thoroughly debunked. Democrats claim that Cheney intervenes to get jobs for Halliburton even though there has never been any evidence of this.

I could go on, but you get the picture. Without Fox News, talk radio and blogs, I don't doubt that this barrage of lies would have done the president in. The mainstream media came very near to overthrowing a sitting American president with lies. How frightening is that? And how much does it make you re-evaluate the media stories of the past during times when there was no check on the national media?

Sunday, January 09, 2005

probabilty and the universe

Patterico and Dean Esmay are doing probability paradoxes. Lots of this post is inspired by those posts and the comments on them.

Esmay offers the venerable Monty Hall paradox
You find yourself on a game show called "Let's Make A Deal." The game is very simple. There are three doors: door #1, door #2, and door #3. Behind one door is a million dollars. The other two doors contain worthless joke prizes. All you have to do is pick which door you want to open, and you get whatever is behind it. But you only get to open one door. By simple math, then, you obviously have a 1 in 3 chance of picking the correct door and becoming an instant millionaire.
You pick a door. As soon as you tell Monty (the gameshow host) what door you want to open, he stops and says, "Now I'm going to do what we always do in this game. Now that you've made your choice, I'm going to open one of the other two doors for you, one with a booby prize." Then he does so. Then he asks, "Okay, now, would you like to stay with your original guess, or would you like to switch to the other door that's still closed? You only get one shot, so do you want to stay with your original choice, or switch?"
So here's the brainteaser: is there any compelling reason to switch doors?
He is impressed with the fact that smart people can be so certain about the answer and be wrong. The correct answer is that you should switch. I'll get to why in a minute, but I recommend that you think about it for yourself. Most people say that there is no reason to switch. And they are quite certain about that. They will argue vehemently and call you stupid for not seeing things their way. And they will be quite wrong. How does this happen?

A lot of it is the perfectly rational rejection of telekinesis --mind over matter. Most intellectuals don't really believe that anyone can bend spoons or move playing cards just with the power of their mind. Yet there is an impression that if Monty, by letting us know where the prize is not can change the probabilities in this way, then he has effected the universe in a strange way. The change in our knowledge has effected the external universe, much like telekinesis.

This impression rests on a fundamental misunderstanding of probability. Probability isn't about the outside universe at all, it is fundamentally about knowledge. A numeric probability is a concise summary of what is known (or believed) about a situation. It isn't strictly about it's apparent subject at all.

There would be no mystery about it if after you chose a door, Monty would open another door that showed the prize and ask you if you wanted to switch doors. In this case, Monty has given you positive knowledge that changes the probability of the new door to 1. There is nothing mysterious about that.

Compare probability with physical properties such as mass. When you give the mass of an object, you are telling me something about the universe. The mass is a measurement that concisely summarizes an enormous amount of information about how that object will behave in interactions with other objects. You can't change the mass of an object by giving me more information --you can only change my estimate of the mass.

A probability is more like your personal estimate of the mass than the actual mass. If I tell you that a coin has 1/2 probability of landing heads, I'm not telling you something directly about the coin, I'm giving you an estimate of the coin's future behavior. That's why knowledge can change probabilities whereas it cannot change physical properties.

Consider the following problem. I have two cans and two fair coins (coin1 and coin2). I put a coin in each can and shake it. I look at coin1 and tell you that it is heads. What is the probability that coin2 is heads? If you said 1/2, then you are assuming that your knowledge of coin1 does not effect your knowledge of coin2. You're right.

Now try this one: I shake the cans and instead of just looking at coin1 I look in both cans and tell you that at least one of the coins came up heads. What is the probability that the other one came up heads? If you said 1/2, you are confused about the question. Which coin am I asking you about, coin1 or coin2? You don't know! I gave you some information about both coins and I'm asking a question that you can only answer by considering both coins together. I haven't changed anything about the physical universe, but I've changed the character of the information available to you. And that requires a different method of calculating the probabilities.

To see how this works, let's try a third experiment. I shake the cans and without giving you any information I ask what is the probability that both coins came up heads. You should be able to figure out that it is 1/4. Just count all the possibilities: coin1=heads/coin2=heads, coin1=heads/coin2=tails, coin1=tails/coin2=heads, and coin1=tails/coin2=tails. There are four equally likely possibilities and only one of them has two heads.

Now, let's go back to the one where I told you that one of the coins was heads. When I tell you that one coin is heads and ask for the probability that the other one is heads, what I am actually asking is whether both coins are heads, given that one of them is.
Does the probability of both being heads suddenly go up to 1/2 just because one of the coins is heads? Doesn't seem likely. After all, there are three ways to get at least one heads, and only one of them has two heads. The probability that both of them are heads is 1/3.

Let's go on to Monty. Let's suppose that after you pick one door, Monty opens another door at random. If this random door has the prize, then the game is over and you lose. Otherwise you are given the chance to switch. In this case, you have no reason to switch because by opening a random door, Monty has given you unbiased information --that is, it doesn't tell you anything different about the door you chose vs. the other unopened door. He has removed one of three possibilities, leaving you with two, and you still have no information to distinguish one of the two from the other. After he opens one door, each of the other doors has a 1/2 chance of hiding the prize and there is no reason to chose one over the other.

But that's not how the game is played. Monty uses his knowledge and a set of rules to decide which door to open, and if you know what rules he is following, you can extract information from what he tells you. The rules go like this (assuming you chose door A):
if the prize is behind door A then choose one of the other doors randomly. If the prize is behind door B then chose door C. If the prize is behind door C then choose door B.
From this you can deduce some conditional probabilities.
probability that Monty opens B given that the prize is behind A = 1/2
probability that Monty opens B given that the prize is behind A = 1
probability that Monty opens B given that the prize is behind B = 0
Given these two conditional probabilities and the initial probabilities, you can calculate the inverse conditional probabilities
the probability that the prize is behind A given that Monty opens B =1/3
probability that the door is behind C given that Monty opens B = 2/3
(email me if you want to see the work)

Monty's rules create these conditional probabilities and the conditional probabilities are what lead to the unexpected result. By contrast, when Monty was opening a random door, the conditional probabilities were not useful:
probability that Monty opens door B given that the prize is behind A = 1/2
probability that Monty opens door B given that the prize is behind C = 1/2
probability that Monty opens door B given that the prize is behind door B = 1/2
This post is an attempt to explain one of the oddities of Monty Hall's paradox: why people are so hostile to the correct answer. I think in part it is because the answer smacks of the occult to people who don't really understand the metaphysics of probability. The other oddity, how people can be so sure they are right when they are wrong is related. I'll discuss that in a later post.

UPDATE: oops. I forgot the links. Fixed now.

a small proposal to help

Against all hopes, the 21st Century is shaping up just as bleakly as the brutal century that preceded it with devastating hurricanes, a catastrophic tsunami, terrorism, wars and genocides.

One of the small tragedies in these larger cataclysms is that families get separated. Also, when someone dies in these circumstances, it can takes weeks or months for the family to find out. The national and international organizations probably don't have time for dealing with such small things while they are saving lives, but it's something that a few individuals can do. Or at least help to do.

Here is my proposal: I'd like to write an analysis tool that takes data from aid workers and matches people up. The aid workers would be able to install a simple program on any laptop or palmtop (a field device). This program would let them take down data from people who have become separated and to record deaths. It would record identifying information, current location, and future planned moves. In more advanced versions, it might record GIS information and take photos or even videos. It would keep the data encrypted so that in war or genocide conditions, captured field devices could not provide the aggressors with intelligence.

The data would be uploaded to a central database where it would be decrypted and analyzed. When a field device coordinates with the central database, the database would download information to help the aid worker reunite families or notify of deaths.

While looking for something that I could do personally to help, this application suddenly hit me. It's something that would help these people and something that I could actually do better than most (the technical aspects anyway, I have considerable experience in writing applications like this). I don't have a great talent for dealing with large organizations, though, and that would be a huge part of the project. I would also need help with languages and internationalization.

And before I even got that far, I would need to find out what facilities are already available for doing this kind of thing. It could be that my proposed system is already out there and in use (I made an effort to find out what the Red Cross does about this problem and got nowhere, possibly because I called just a few days after the tsunami).

Does anyone have any comments on the project? Any knowledge of the current state of technology in this area? Any interest in helping? You can email me or leave a comment.