THE VIEW FROM CHAOS MANOR
View 576 June 22 - 28, 2009
This site looks better if you set your default font to Georgia.
Highlights this week:
For boiler plate, search engine, and notes on what in the world this place is, see below.
For Previous Weeks of the View, SEE VIEW HOME PAGE
If you intend to send MAIL to me, see the INSTRUCTIONS.
This is a Day Book. Pages are in chronological, not blogological order.
June 22, 2009
With luck this is the week I finish the draft of Mamelukes. We'll see.
Iran: a friend recently observed that this is an odd situation. It is the first time there has been any question of vote fraud in the mullah regime because there has never been any need for it: if the mullahs don't want someone to win, they simply don't let him run. One has sympathy for the demonstrators, but the outcome of this election isn't going to be a change in Iranian government principles.
The New York Times shows a former Guantanamo detainee on the beaches of Palau. He's learning to fish. I presume he's getting some kind of retirement benefit; is this paid by US taxpayers? I wonder how it compares with the retirement benefits of US veterans who served similar times on active duty rather than in detention camps? An idle thought.
There was some mail and commentary over the weekend. I'll have more mail after my walk.
The current edition of New Scientist has an article on "AIDS Denialists" which raises some questions of interest.
It pretty well summarizes the evidence for the HIV causes AIDS hypothesis, and makes the case as well as I have seen it made. I'm convinced, but then I never was an "AIDS Denialist." My concern is the overwhelming power of peer review in an era of mostly government sponsored research. In brief: many years ago, before the evidence summarized in this article (mostly the dramatic fall in death rate after introduction of the ART "cocktail") piled up, Peter Duesberg, Chief Virologist of the University of California, said he did not accept the HIV = AIDS hypothesis, and proposed an experimentum crucium which would falsify the HIV causes AIDS hypothesis. As I recall he asked for ten millions of dollars; this from a budget of over a billion earmarked for AIDS research. (AIDS research funding very quickly grew to huge numbers, larger than the amounts earmarked for breast cancer, a matter of some concern to women's organizations.) He didn't get the money, and much of the scientific community turned on him with a vengeance. He was called an unqualified quack. In those days "AIDS Denialist" hadn't been invented as an accusation, but there are few epithets that were not hurled at him.
There are several problems here. First, Duesberg was hardly a quack, and had he requested a hundred million dollars for HIV research without stating his opposition to the HIV = AIDS hypothesis he probably would have got it -- he was, after all, a key figure in the discovery of the retrovirus, and was among the leading figures in virus research. One would have thought that the simplest way to deal with him would have been to give him the ten million and see what he did with it; one would have thought that his reputation would have earned that much.
Second, given the billion dollars to be spent on AIDS research -- just about all of it to be spent on the HIV = AIDS hypothesis -- it would seem to be a simple act of prudence to invest in a crucial test of the central hypothesis driving all that research. The evidence for the HIV = AIDS hypothesis may have been pretty strong, but it wasn't overwhelming them.
And, finally, the incident raises the general question of consensus science: in an era in which funding is largely governed by the peer review process, how do dissenters fare? What if the general consensus be wrong? That wouldn't be unprecedented. nor is the use of ridicule as a major argument against those with new ideas. (Recall "Tell me, Mr. Darwin, on which side of your family are you descended from an ape?" There was, after all, a time when "Creation Denialist" was a pretty serious charge., although that phrase was not used. When I was in high school it was still illegal to teach evolution in the state of Tennessee although Brother Fidelis, having told us that it was illegal for him to teach the theory of evolution, went on to teach "what scientists say they believe" about the subject.)
It is now twenty years later. Duesberg never did get his crucial experiment, and soon separated from the University of California. He now heads a small privately funded laboratory. I have not heard of any notable results from his research efforts over the past twenty years. Meanwhile the evidence for the HIV = AIDS hypothesis has piled up to seemingly overwhelming levels. (I say seemingly, not because I continue to have any doubts on the subject -- I'm not an AIDS Denialist -- but because I have no expertise on the subject, and the only thing I've read about it in years is the New Scientist article I cite above; read that and you'll know all I do about it).
I do draw different conclusions from the New Scientist writer. New Scientist is concerned that AIDS Denialists are responsible for the enormous death rates in Africa and in particular for the belief by former Republic of South Africa president Mbecki that AIDS wasn't a contagious disease but "simply a name for the diseases of poverty that has been legion in Africa since the days of colonial rule." My question is, if twenty years ago the crucial experiments had been performed, would Mbecki have been convinced? That is, could the suppression of research that did not accept the HIV = AIDS hypothesis have contributed to the perpetuation of the AIDS is not a contagious disease beliefs? I also wonder what Duesberg, who was after all a leading scientist, the Chief Virologist of the University of California, might have found had he been funded; could his attempts to disprove the HIV = AIDS hypothesis have resulted in some useful discoveries? But that's only conjecture. I don't know enough about biochemistry to have more than a vague notion on that subject. I do presume that his experiments would have failed to falsify the HIV = AIDS hypothesis; and perhaps that might have made it more difficult for the Denialists to recruit people like Mbecki.
Most "contrarian" science is likely to be wrong. Most -- but not all. The great breakthroughs in science have often gone against consensus beliefs. We can't act as if everyone who questions a scientific consensus is right, else we'd never get anything done at all. Some contrarians deserve dismissal. Sorting out which ones deserve some attention and possibly funding is a very difficult problem. It's also one that we don't seem to be paying much attention to.
|This week:||Tuesday, June
You will find
interesting in many ways.
The Iranian Election and the Revolution Test,
Based on what we're seeing and hearing from Iran - and more importantly, what we're not seeing and hearing from Iran - I believe this analysis is more likely correct than what we're getting from the MSM:
This is pretty well my analysis. I thought of the analogy to China when it first began. Now a number of Obama critics are berating the president for not doing more in support of the uprising. I'm not sure what he could have done. The United States is not popular in Iran even among those who have smart phones. Injecting the US into Iranian politics would probably be counter productive. Encouraging an uprising we are not willing to support with blood and treasure is not often a good policy: look where that got us after the First Gulf War when we encouraged uprising in Iraq but then did nothing to support them.
The United States is not omnipotent, nor are we all wise. Iran is not Iraq, and the government there has a far different relationship to its people than Saddam Hussein and the Baathists had to the people of Iraq. Is democracy the answer to all foreign relations questions? At one time that was assumed. Democracies don't make war on each other. Liberal democracy is the end of history. So we were once taught during the heady years after the collapse of the USSR. I questioned that premise then, and it seems even less obvious now.
I believe there may be a democracy in the Middle East (other than Israel). It appears to be Gaza. It would certainly make war on Israel given the means.
The world is a dangerous place.
And if that isn't enough to think about this morning:
Moore's Law is coming to an end - because of finance:
It reminds me of the trends noted long ago about the high cost of fighters driving down the numbers bought. Makes sense.
Indicator of Success:
IQ is not the most reliable indicator of success; far from it. I believe that privilege is the single most reliable indicator of success. The higher the level of priviledge to which one is born, the greater the odds for one's success in life.
Given two prospects of equal IQ, one common born and the other born to privilege, which will be the more likely to succeed? Even if we extend that field to ten or even a hundred prospects of equal IQ, one of whom is born to privilege, the expanded field might change the odds, but not likely the outcome.
My correspondent says "indicator" rather than "predictor". When we discuss the utility of IQ, we generally mean as a predictor. There is a difference.
There are a number of knotty problems raised in this common sense observation. Most of us would agree with the general proposition that privileged people do better in life than those not privileged; so what do we mean when it is said that IQ is the best single indicator of success? In what context is this true?
The classis example is grade prediction. The University of Washington grade prediction program, funded by the Navy Electronics Lab back when funding such studies was not so politically incorrect, took measures of all the students coming into the University and gave them batteries of tests as well as gathered a great deal of information about such matters as high school grades, what school they went to, high school class standing (which turned out to be more useful than high school grade, incidentally) and other such measures. Then four years later we took the grade point average as well as the major. We also had GPA in classes in the major. All this went into a huge matrix and ended up with a multi-factor program that predicted the grades of incoming students in a number of majors. This was supplied to the student and various counselors.
The program worked pretty well. It wasn't exact, of course, but if it predicted that you'd flunk out of pre-med, the odds that you'd get grades good enough to get into medical school were way low.
Note that we had a very narrow definition of "success": what GPA did you get after four years at the UW? Note also that while there was some theory involved in selecting the predictor variables, none of that went into generating the prediction equations. The formula fell out of the matrix algebra manipulations of the data.
Similar studies elsewhere have got the same results. Most of these have been academic and "success" was generally defined by grades, but there have been other studies in other fields. The general proposition is that if you take a group of people -- bank tellers, machinists, preachers, officer candidates, gardeners -- and form a committee of experts to rank order those people on their "success" in what they are doing, then what measure might be used to predict how successful a single individual would be? And the answer to that is almost IQ or some other measure of "g". This doesn't mean that g will be a very good measure, but for most non-sports activities g will be the best single predictor. Note, though, how we have defined success here.
As to general success in life, I don't know how to define that. We could I suppose use a committee of experts (although who is an expert on life success?) to rank order a bunch of people, then look at IQ vs. "privilege" to see which does a better job of such prediction. We'd need a defined measure of "privilege", which I presume would be some measure of parental socio-economic status. I'm not sure what the result of that experiment would be (it may have been done; there have been a lot of such studies).
We certainly have done studies of parental SES and IQ as predictors of GPA in many schools and in many trades. Parental SES isn't as good a predictor of success as a plumber or welder (I'm pretty sure I recall those two studies, done a long time ago of course). IQ was better in those.
The point here is that "privilege" certainly helps. Among other things, privileged children are likely to have better educations because their parents have many more choices. If, however, you are trying to make predictions about a group of privileged kids -- say the incoming class to an exclusive prep school -- you'll find by and large that IQ does a better job than parental SES. I wouldn't care to speculate about predicting success at being, say, a stock broker, where parental SES would have a lot of influence over whom you know and whom you meet. Clearly we all know stories of successful privileged but stupid aristocrats. IQ or "g" is a powerful predictor, but one does need to qualify generalizations with the usual "other things being equal."
June 24, 2009
I have errands and medical appointments all afternoon.
There are discussions of Iran and some explication on social science and prediction in Mail.
And here's some good news:
Michael Flynn, who is a quality control engineer when he's not writing excellent science fiction, says
Bingo. When the climate modelers come up with falsifiable hypotheses and the data confirm their predictions, we must take them seriously; until then, chopping the GDP by 10% to satisfy Gore's demands seems at least questionable.
This was the most likely outcome. Alas.
We would, of course, rejoice if there were some way to get "free" energy (one presumes some kind of cold fusion of common elements would be involved). It's not likely, but e=mc^2 promises a great return for a small amount of mass, if only we knew how to do it. I have no idea of who runs Steorn, but they don't seem to be out trying to raise money. I wish them well.
Having said that, I don't expect them to bring it off. Fusion energy seems to require expensive equipment and, unsurprisingly, violent events.
June 25, 2009
Iran remains murky. There is a piece in today's
LA Times that I found interesting: it tells of machinations I had scarcely
heard of, and which weren't in Luttwak's analysis. There are echoes of the
Iron Law. Is a new monarchy forming? I have no idea. But it's worth reading,
which is more than I can say for a lot of what's in the newspapers now.
On the subject of news articles worth looking at, Roland reminds me of something I ought to have mentioned yesterday.
The story opens:
When I read that, my reaction was, why are the taxpayers paying for this farce? There may have been some students in that class who could profit from a world class university prep education. There may have been some who would profit from a general education with some emphasis on trade skills. What was happening, though, was a classroom in which the principal lesson being taught was that behaving yourself like a civilized human being doesn't earn you the right to teacher attention. There are other lessons. Read the article and consider that education consumes a very large part of what you pay in taxes; this is some of what you're getting for it.
But the schools can't just throw out the disruptive: not only are there circling legal sharks eager to defend the rights of Byron Gordon to noisily sharpen pencils and Deon Crockett to wander through the room complaining at full volume, but the school won't get any money unless the students attend the school. Other legal sharks will defend the rights of Byron Gordon and Deon Crockett to be "mainstreamed", not to be put into a classroom for the disruptive so that those who actually behave themselves and try to learn something from teacher Katy Bridger might have a shadow of a chance to get something from school attendance. Meanwhile, teacher unions somehow don't get involved in insisting that teachers have the authority to keep order and teach the students who actually want to learn, or if they do, their efforts are both unreported and generally in vain.
Most of those who read this wouldn't put up with having their kids in a fifth period class in which Byron Gordon noisily sharpens pencils and Deon Crockett wanders through the room complaining at full volume. We appear to have an extraordinarily expensive system for insuring a plentiful supply of unskilled labor a few years down the road.
When I was in fifth grade there were 30 pupils in 5th grade in that room. There were also 26 6th graders in the same room. The teacher was a two year Normal school graduate. No one noisily sharpened pencils. No one wandered through the room complaining at full volume. And despite the large class size with two grades to the room I suspect that we learned more in that deprived 5-6th school than anyone in that fifth period high school senior class described above. Perhaps I am wrong, but I wonder. I can't compare to my senior year in high school because the very notion of leaving my seat without Brother's permission still gives me chills. But that was another century.
The American education system seems geared to spending money without regard to any possible result. Perhaps it works for those of high socio-economic status in communities that actually pay attention to order and discipline in the classrooms. But I sure pity any bright kids huddling in the middle of Katy Bridger's classroom while Deon Crockett wanders the room complaining at full volume. And as a Los Angeles taxpayer I resent having to contribute to that farce.
I have seen a rash of articles explaining that nuclear power costs enormously more than coal, oil, or natural gas (but not, I think, more than wind). Dr. Petr Beckmann, when he was editor and publisher of Access To Energy, did detailed analyses of generation and operations costs of nuclear power and concluded that it was quite competitive in lifetime costs. After his death, Access to Energy became less technical and unreliable in publication date (still worth subscribing). I suppose I need to do some digging on my own, in my copious free time. The last time I looked closely at energy costs, one of the major factors in nuclear energy cost was legal fees and regulatory compliance; the nuclear industry had a great deal of data showing that if they could just get a one-stop shop and some freedom from litigation that caused construction delays, nuclear power was very competitive with just about anything except hydro.
It's pretty easy to show that nuclear is safer than most other sources. At one time the most dangerous form of power generation in terms of injury cost/ KW was rooftop solar power. The injuries were from people falling off roofs; since there weren't very many KW being generated, it didn't take many rooftop injuries to bring the injury cost/KW high. More seriously, the injury and death cost/KW of coal is high since you must count not only mining accidents, but railroad crossing accidents due to coal trains. Let me hasten to add that I have had no new data on this since the 1980's, so some of my information is probably out of date, but the health/safety concerns won't have changed much. And yes, I know about Chernobyl; that wasn't a nuclear power plant, (it was dual purpose with production of weapons grade fissionables being the primary purpose) and there will never be a positive-void nuclear reactor built in the United States. A quick Google on nuclear power fatalities indicates that things haven't changed since I last took a detailed look: nuclear power is safer than other kinds of power. This is common sense, given France's experience.
The last data I had on nuclear power showed it was the most reliable source of base power in that it had the highest percentage of on-line time. That's old data and may no longer be true, so I probably need to look into that again, but I haven't seen any indication that nuclear has become less reliable. Of course the plants are getting much older and since none are being brought on line, the average plant age grows monotonically.
As to why it is said to be so costly, I need to see the actual data and analysis; the conclusions I've seen don't square with what I concluded last time I looked at the actual numbers. I can believe that nuclear under present legal/regulatory costs is terribly expensive; lawsuits and compliance costs always have been a very high part of the cost/KW.
June 26, 2009
The climate debate has become more shrill now that a "cap and trade" bill has been brought out. This is a horrible law that will have a great effect on everyone. The good news is that the global warming hysteria has triggered some new interest in actual scientific discussion of the man-made global warming hypothesis (see mail), and this bill can be defeated. If it's allowed to become law it sets up a bureaucracy and people who will profit greatly (and therefore can afford to spend huge sums for lobbyists) and repealing it will be difficult even if the glaciers begin to advance.
Idea have consequences. This idea has a lot of consequences.
Another idea with consequences comes from Google's attempt to see that everyone gets access to every public domain work. See mail, and read the article it points to.
June 27, 2009
"Authority to accomplish the impossible implies absolute power" is an eye-opening statement. It is very good. You might want to spread it around. Much like that "power corrupts" statement.
Thank you. On reflection, perhaps the principle deserves some kind of title analogous to Pournelle's Iron Law of Bureaucracy, but I haven't thought of a suitable name yet. It does hold up to inspection. The statement was that just as in logic a false statement implies the universe class, in human affairs authority to accomplish the impossible implies absolute power, and that does seem true enough.
And now the House has passed perhaps the worst bill in its history, 1,000 pages that no one had read giving enormous regulatory power in pursuit of the impossible. The actual effect of US adoption of cap and trade on climate is essentially nil. China and India will continue to burn coal and oil as they industrialize. CO2 levels will continue to rise. Global temperatures will continue to rise. US self destruction may affect the global temperature in the year 2100, but I know of no theory that can show the effect will be greater than 1 degree C (that is, global temperature would be 1 degree C less without US contribution to CO2) and that is a very extreme limit; few of the theories show our contribution to be large enough to have that much effect. The most likely outcome is an enormous hamper to US economic recovery and no effect whatever on global temperature.
The President speaks of this as a jobs bill. The cost of each job created by this is enormous. Economic growth and energy cost have a high negative correlation and always have, and this is an energy tax; it will raise the price of energy, whatever else it does. Nearly all "green" energy produces energy at a cost great than the equivalent of $150/bbl oil.
Whom the gods would destroy, they first make mad.
June 28, 2009
I pretty well took the day off.
This is a day book. It's not all that well edited. I try to keep this up daily, but sometimes I can't. I'll keep trying. See also the weekly COMPUTING AT CHAOS MANOR column, 8,000 - 12,000 words, depending. (Older columns here.) For more on what this page is about, please go to the VIEW PAGE. If you have never read the explanatory material on that page, please do so. If you got here through a link that didn't take you to the front page of this site, click here for a better explanation of what we're trying to do here. This site is run on the "public radio" model; see below.
If you have no idea what you are doing here, see the What is this place?, which tries to make order of chaos.
For platinum subscription:
If you subscribed:
If you didn't and haven't, why not?
Strategy of Technology in pdf format:
For a PDF copy of A Step Farther Out:
For the BYTE story, click here.
Search: type in string and press return.
The freefind search remains:
Entire Site Copyright 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009 by Jerry E. Pournelle. All rights reserved.