Never apologize

The Japanese, at least in theory, had a culture of responsibility, where senior leaders apologized for errors in judgment. Sometimes the apologies were gruesome.

I don't think Charles Krauthammer should commit Seppuku, but an apology would be nice. Kudoes to Crooked Timber:

Crooked Timber: Credibility problems

... But even better to my mind, was Krauthammer’s confident judgement on Iraq WMDs back in April 2003.

...Hans Blix had five months to find weapons. He found nothing. We’ve had five weeks. Come back to me in five months. If we haven’t found any, we will have a credibility problem.

Krauthammer is now advocating military action against Iran.

Bics and bump keys: innovation diffusion

Innovation diffusion studies were popular in the 1980s. How do new techniques, new methods, move through a population? I don't hear much about it any more, perhaps because diffusion has accelerated substantially.

Crooks have quickly learned how to steal "theftproof" cars (manufacturers left "back doors" for mechanics and for internal use). A few years ago the internet had demonstrations of opening bike locks with a Bic pen -- a vast number of locks had to be replaced.

Now it's the household lock. The newest innovation in breaking and entry is the use of the (Future Feeder) � Bump key. A set of 9 or so specially cut keys, combined with a light hammer blow and a quick twist, will open the vast majority of household locks. The methods leaves no sign of entry, so you may simply discover your diamond tiara is gone -- but nothing else is touched. Might be hard to get a claim paid ...

New locks will come along. The interesting bit to me is the speed of diffusion. Car theft, lock picking, blowing up planes -- new techniques are deployed quickly.

Leiberman lost on the merits

He wasn’t defeated by a conspiracy of loons. He deserved to lose.

Brad DeLong's Semi-Daily Journal: Was David Broder Always *This* Stupid?

... Not true. A switch of two percent of the voting electorate would have given Joe Lieberman a victory. If Joe Lieberman had announced that he was going to live or die by the result of the Democratic primary, spent his $2 million in the bank on the primary, and gotten all the extra Democratic politicians who would have been happy to campaign with him under those conditions into the state--he would have won. If Lieberman had made up his mind whether Lamont was really a Republican from Greenwich or a tool of the far left that believes Osama bin Laden is not a threat, he would have won. If Lieberman had been willing to talk this spring about how Bush could handle the situation in the Middle East better--he would have won. If Lieberman had been willing to call for mass resignations from the Bush administration for failed implementation--he would have won.

Lieberman had to work very hard to lose this one. In claiming that "it was clear" that Lamont would prevail, David Broder is simply being a very rare species of idiot. ...

One of the reasons I don’t subscribe to the NYT Select is so I don’t have to worry about accidental exposure to Broder, who’s far more irritating than Ann Coulter. Ann is a self-referential parody and loon, but Broder is a brainy moron. Ok, so the Times Select also costs $50, which is a lot to pay for Value = (Krugman) – (Broder+Friedman).

Talent, terrorism and the shoulders of giants

Yesterday I wrote (Gordon's Notes: 8/09: How talented is this group?) that I thought the talent level of al Qaeda’s team has been pretty low the past five years.

I'm looking forward to learning more about the current crop of suspects (some of whom will likely turn out to be innocent); particularly how many have engineering or science degrees from serious institutions, etc. That will tell us if George Bush's al Qaeda recruitment strategy is bearing fruit. If it turns out that this is a talented bunch then we have to give even more credit to Pakistani, UK and US counter-terrorism efforts. Given the size of the conspiracy, however, I suspect they’re dolts. No-one with any brains would launch an effort that big from the United Kingdom.

Today I give credit to two recent public radio shows for emphasizing another part of the talent puzzle. One show was part of a superb series on supermax prisons and solitary confinement, the other was an interview this morning with a remarkable counter-terrorism expert. I wish the latter kept a blog ...

First the supermax. The inmates at supermax prisons are not generally known for their creativity, insight, or intelligence. So I was surprised by the range of ingenious and lethal techniques they deployed to support violence, commerce, and recreation. A handful of innovators, combined with modern and traditional communication channels, intense motivation and ample time to scheme, and memetic selection, have delivered advanced techniques to the average prisoner. These men and women stand on "the shoulder of (nasty) giants".

Then the counter-terrorism specialist. He pointed out that a large range of modern terrorist methodologies seem to have been developed by a few unfortunately talented IRA engineers and specialists. Their techniques and technologies, developed during the 1950s to 1980s (perhaps with some KGB help as well?) have been widely disseminated though print and spoken methods (I’m sure the net helps too). Apparently some of this team continues to work gainfully in South America (guess where?), though it’s not clear how inventive they are nowadays. Terrorists too have shoulders to stand on.

So the British terrorists (they were British citizens and terrorists, so they’re British terrorists) may have been executing a derivative attack, and they may have been dolts, but a lot of knowledge has been packaged to a point that even dolts can execute it. Process improvement, knowledge management, and the falling costs of lethal weapons — it’s all a part of the falling cost of havoc.

Quicken: I do loathe thee

Quicken 2004 is up to its old bugs again -- auto classifying every transaction incorrectly, creating myriads of nonsensical transaction splits. I clean it up and then it does it again. It also threw a memory exception; I had to 'copy' (rebuild) the database again.

I really only use Quicken now to download my credit card transactions and check transactions. I've gone back to spreadsheets for overall finance work and tax reporting, and given up on balancing the checkbook.

I've used Quicken since DOS version 2. It was a great piece of software then, though some DOS versions had nasty database corruption bugs. After about DOS version 4 it started downhill, and it's been sleazing about for years. One day I'll abandon it completely.

Progress in software is not guaranteed. The personal web building toolkits have never returned, that market is gone (iWeb doesn't qualify). Finance software is pretty meager on the PC, and non-existent on the Mac. Word processors aren't as good as they were 10 years ago. Photo management software is, of course, far better. Overall though, it's a mixed picture.

8/09: How talented is this group?

The initial impression, obviously uncertain, is that the 8/09 multi-plane bombing plan was (is?) a serious operation of an al Qaeda like group. I'm most interested in the talent level of the group.

The current hypothesis about why we haven't had major attacks post 9/11 in the US (exempting the mysterious anthrax bio-attack) is:

1. No robust command and control function left post Afghanistan.
2. A desire for a 'big' attack -- no wish for smaller actions.
3. An unexpected (by many) level of opposition, or at least non-support, among US muslims for any attacks.
4. Some minimal hardening of the most obvious targets. (Except for the cockpit doors, however, this is largely discounted.)
   

My personal hypothesis is different. I think al Qaeda has had a very weak bench. I think that the intellectual capabilities, operational effectiveness, and creativity of their "team" has been miserable.

I began to think this after the Richard Reid "shoe bomber" episode. Sending a cognitively disabled paranoid schizophrenic on an attack only made sense if it was either a "feint" -- or if that was the best they could do. In retrospect the Reid attack was not a feint, so the implication is that they had almost nobody left with any talent.

Why is this? Genius is everywhere. There must be tens of thousands of Muslims who are smarter and more creative than anyone I know -- albeit with, on average, fewer educational opportunities. If they wanted to destroy western civilization we'd be on our backs now. As it is, despite Bush's incompetence, we're still breathing.

For some reason, al Qaeda hasn't been able to recruit really talented attackers. I think Bush has been doing an outstanding job of helping al Qaeda improve their recruiting pool. Once we get a look at this bunch, we'll see how successful he's been.

Update 8/16/06: Early returns suggest a very thin bench here. I'm still waiting to learn if any went to college, and particularly if there were any scientists, technologists, geeks, or engineers.

Cancers can become parasitic life forms

A canine tumor turns out to be a parasitic form of a cancer cell from a dog that died 300-500 years ago. Now the hunt is on for additional forms of parasitic cancer:

The Loom : A Dead Dog Lives On (Inside New Dogs)

...One question the scientists do raise is how common such cancer parasites may be. Scientists have reported tumors that spread from transplanted organs, but these don't have a way to sustain their spread for centuries. Still, between mating, biting, and otherwise making contact, vertebrates provide plenty of opportunities for cancers to spread. And it is striking that our immune systems--and the immune systems of other jawed vertebrates--are equipped to battle so strongly against foreign tissues. What for? It's not as if Devonian-era sharks were giving each other liver transplants 400 million years ago. Perhaps, the scientists suggest, our ancestors had to fight against a different sort of tissue donation: cancer parasites.

In the dogs the parasite is relatively benign, in Tasmanian Devils a similar sort of parasite is lethal. This will be interesting ...

Microsoft in crisis: it's more than Vista

To put it mildly, I've never been a Microsoft fan. I think MS Word alone has knocked billions of our GNP. Their path to power was littered with the corpses of better solutions. On the other hand, I've always given them credit for Microsoft Excel, I think XP was a very good OS until the security crises of the past few years (not as good overall as OS X, but good), and I've always thought they made superb hardware. Really, they've been a good hardware company.

Then I bought the Microsoft LifeCam VX-6000. It takes real genius to produce an XP STOP screen on startup -- the rarely seen XP Blue Screen of Death (not on my home machine, on my pure work laptop). Only Microsoft would have the knowledge and cojones to hack into the ring 0 driver code of XP and do something like this (my guess is that the LifeCam software was written for Vista).

Microsoft is in deeper trouble than a Vista product delay. They increasingly feel like a company coming apart at the seams. This won't be pretty, but maybe it's the best alternative we have to a stifling monopoly. With luck they'll be back in 10 years -- without the monopoly ...

Ta Mok: An obituary

What was he like, as a child? Was he loved once? Do his children mourn him?

The Economist: Ta Mok

Aug 3rd 2006

Ung Choeun (“Ta Mok”), the last surviving leader of the Khmers Rouges, died on July 21st, probably aged 80...

... His name, “Ta Mok”, meant “respected grandfather”. Villagers mourned him because they said he had brought prosperity and work to their poor forests. He claimed the same. Under his authority, especially in the south-west of Cambodia where he had been zone secretary in the communist Khmer Rouge days, roads, dams and bridges sprang up everywhere, and bright green rice fields stretched to the horizon.

Yes, Ta Mok built dams. They were erected in the late 1960s and 1970s by thousands of slaves. These people had never done hard labour before. They were doctors, teachers, writers, scientists, forcibly evacuated from the cities with whatever they could carry, made to live in barracks and worked for 12-14 hours a day. Their food was rice, which at one point fell to 150 grams a day, or rice gruel, or watery soup of banana stalks. If they did not die of disease, starvation or exhaustion, they might be killed for reluctance or dissent, or for wearing glasses. Between executions and deaths from aggravated causes, almost a quarter of Cambodia's population—1.7m people out of 7m—died between 1975 and 1979, when the Khmers Rouges were in power.

Ta Mok denied that there was blood on his hands. Towards the end of his life, as he awaited a long-delayed trial before a UN-Cambodian court on charges of genocide, he asked his lawyer to tell the world that he had never killed anyone. Technically, this may have been true. The killings he ordered in the south-western zone were done in the deep jungle, where he never saw them. Those he master-minded for Pol Pot, the Khmer Rouge leader, when he was one of his senior advisers and later his army chief, were more like mob massacres in which no general needed to intrude...

... As early as the 1950s his cruelty was notable, and noted. So was his distrust. He saw traitors everywhere, and more as the years went on. Not only his party rivals, but ordinary people too, became CIA agents, lackeys of the Thai government, agents of the Vietnamese: those neighbour-countries, especially, being threats to the Khmer Rouge regime. He made the peasants in his zone wear black clothes, the better to control them.

... Ta Mok, born a peasant and with no fancy French education, talked of himself as one of the “lower brothers” at the grassroots, doing the hard revolutionary work. But he enjoyed the material rewards. He put his family—brothers, sons, daughters, in-laws—into party jobs, and built himself a fine red-brick headquarters in the middle of a lake.

In power, his basic, crude advice often cut through the jargon of the other cadres. Phnom Penh was cleared out and sacked in 1975, with the deaths of 20,000 people, largely because Ta Mok condemned his colleagues as “layabouts”. The year before, in Oudong, he had cleansed the old royal city of its 30,000 residents and burned it to the ground. The people were marched away to an uninhabited region in which they could build the new Cambodia.

After the collapse of the Khmer Rouge regime in 1979, Ta Mok kept on fighting for two decades. By the end of that time, though crippled by a landmine, he was the movement's leader. He had argued with Pol Pot, and seized power. The former leader died in 1998, in his custody, as both men fled farther into the northern mountains...

The writing is cold and restrained. The deeds must tell the tale, curses have no power here. I do not know how Ta Mok was made, but in a world of billions there must be hundreds such. Men mostly, with minds of malice and hate -- but for power they would be Ta Mok.

See also: Robespierre.

Paradigm: banned by Nature

Marvelous. Nature (the journal) gets many taste points:

Rapped on the Head by Creationists | Cosmic Variance:

... Did you know that Nature has an editorial policy forbidding the use of the words “scenario” and “paradigm”? Neither did I, but it’s true ...

Did Colonel Petrov save our world?

It's an old story (2003), and I've heard variations on it before, but the story of Colonel Petrov is worth a read. The allegation is that a Soviet computer error triggered a false alert pm 9/26/83, but rather than notify superiors (who might have launched the Soviet arsenal -- or decided the alert was an error), Colonel Petrov decided on his own initiative that the alarm was false. It was.

If you want to look for supernatural intervention, forget about intelligent design. Look to our having survived the 1980s. It seems unlikely, even in retrospect ...

Could ultrasound predispose vulnerable fetuses to neuronal migration disorders?

Be the Best You can Be: Ultrasound, neuronal migration, and autism

The day the software died - my worm and the end of the second golden age of the PC

Within a day of downloading two sharware webcam products [1] for testing with my new LifeCam I found my XP box running hot and minimally responsive. Bad feeling time.

Since Norton Antivirus, which does a full disk scan of my system every night, hadn't alarmed I tried the very latest version Microsoft's free utility for worm detection. It found nothing on the standard scan, but a deep scan and clean reported "partial removal" of W32/Mytob-W or some novel variant thereof. A follow-up NAV scan found nothing, but I will repeat with updated viral definitions and then migrate to Microsoft's new security product. This is my 3rd infection in about 17 years; one was with Win98, one with Mac OS 9 (yes!), and then this.

I will now, with less vanity than it appears, declare the end of the 2nd era of personal computing -- The Day the Music (Software) Died. It's dead. Put a fork in it.

The first Golden Age, from 1982 to 1998 was a time of amazingly creative software built on minimal resources. The first Golden Age died with the Microsoft monopoly.

The second Golden Age began with Gopher in the early 1990s and probably ended around 1999. The networked power of the second era, combined with the monoculture of the Microsoft monopoly, carried the seeds of its own destruction. When uber-geeks like me [3] are afflicted by worms and massive quantities of spam, the end is in the rear view mirror.

What will the third era look like? I think there are two paths, and we'll take both of them. One is the 'trusted computing' path of locked down software and a complex chain of control, combined with a move to software leasing and web services. Microsoft will lead this path.

The second is the trusted community path together with strong reputation management and coupled with open source and public file formats.

These paths are the Authoritarian state and the Citizen state. Sparta and Athens [3]. We've walked these roads before.

In the meantime what new measures will I take?

1. I'll abandon Norton and switch to Microsoft's security package. I think that if one is going to run XP there's really no alternative to Microsoft for antivirus and security protection. They have the incentive, the resources, and the expertise to provide the best possible service. [4]
   
   
2. I'll accelerate the migration of my last remaining XP machine to an OS X replacement [5].
   
   
3. I'll install a hardware router/firewall with traffic analysis, automated shutdown and alerting to support early detection and termination of problems.
   
   
4. I'll no longer download software from the websites of owner/authors. Instead I'll download from versiontracker and other community sites where risk is broadly shared and feedback identifies problems.
   

Other measures to be added here ...

[1] No, I'm not doing icky stuff, it's for the most boring and dull work you can imagine. It's a good thing I tested this at home however. I downloaded the products, one of which I'd registerd for an earlier failed project, from the author's sites. I thought that and Norton Antivirus was a reasonable combination, but I was probably wrong. I didn't bother to prove the sites were the source however, so I won't name them. The evidence is circumstantial.

It's even a bit hard to say that this really is a new event. I've gotten worms and spam from "myself" for years, but when I've investigated I found spammers were forging my address. So I wouldn't necessarily spot a worm infection ...

[2] Sure there are more XP security measures one could take. Internal and external firewall and traffic analysis, better antiviral solutions, etc. My XP machine already strains under the weight of its current security infrastructure.

[3] Ok, it's a cliche and a gross simplification. Athens was hardly pure.

[4] I know Microsoft has been performing lately like a corporation on Crack. I didn't say they'd do well.

[5] OS X is more secure than XP because it allows users to work successfully in non-admin accounts. It's also more secure because the OS X community is far more interactive and reputation based than the XP community, the difference between a village and a city. Lastly, and most obviously, OS X machines are very expensive to acquire and have limited use in business; kids and crooks will target XP first. Of these distinctions only the first is likely to persist, and if Apple doesn't publish its own security solution soon OS X will be in deep trouble.

Obesity, nutrition facts and corrupted government

Act One: Obese senator decries the sorry state of American's children.

Act Two: Nutrition Facts on Tortilla Chips package

Serving Size: 11 chips
Servings per container: 2 3/4
Calories per serving: 150

Background:

1. It's likely that fewer than 1/10 Americans can multiply 2.75 * 150 to figure out how many calories this bag contains (about 415).
   
2. This bag is likely most often consumed by one person at one sitting -- with cheese dip.
   
3. In the US "Nutrition Facts" are a federal mandate.

Ergo one may deduce:

1. The federal government is exceedingly corrupt, as well as incidentally hypocritical.
   

I probably don't need to explain how this inferential chain to my eccentric readership, but for the record the trick of displaying calories per serving on a label, and then declaring that a single user package has a non-integer serving size > 1, is an obviously evil attempt to hide how caloric content. That's fine, the market demands evil. It's the obvious political corruption that's more noteworthy.

Why corruption? The proof of corruption is that the fix for this fraud is simply to require that the "Nutrition Facts" also display the Calories per container. This fix has never been made. It's the absence of the obvious fix that proves the corruption.

We are getting the government we deserve, and evidently we've been really, really, bad.

Old skate parts, Google, and the walk to school

When I was a boy we had to fight the wolves and the blizzards on our 19 mile walk to school. That was before global warming, before the Twin Cities became the Riviera of the midwest.

Also when I was a boy we couldn't just Google on "Salomon TR parts" and find a source for a missing bolt from an inline skate that was last manufactured in 1998.

Sigh. I thought I'd get a new pair of skates out of that lost bolt. (BTW, the next time I buy skates, I'll also buy a set of parts. Skates are nowhere near as standard as they used to be.)

PS. Skates.com offered Google Checkout. The first time I used Checkout with Firefox there was a minor glitch, but this try was delightful. Checkout keeps my email address confidential, giving the vendor a disposable address that routes to my gmail account. Checkout also keeps all my receipts and I can view the order status on that page as well. Very much as expected -- a direct assault on Amazon.com.