Wednesday, May 30, 2012

A smidgeon more

More Chesterton:
If, for instance, a reformer proposed to resist the concentration of capital in combines and corners, the dear old gentleman would declare that nothing could stop the growth of monopolies and money-rings, because we could not alter human nature. This only serves to prove that he was himself singularly ignorant of human nature, if only because he was singularly ignorant of human history .... His moral theory was entirely modern, and was in flat contradiction to the moral theory that is really ancient. Most of his ancestors regarded making a corner simply as a crime, like that of cutting a throat or picking a pocket. Forestallers, as our fathers called them, were often put in a pillory, or even hanged on a gallows, to stop them from doing what he declares they cannot be stopped from doing.

Biodiversity at the Triassic beginning

An article on biodiversity after the Permian collapse (summarized at SciTechDaily if you don't have an institutional subscription) describes the possible reasons for a very slow recovery from whatever it was that ended the Permian era so dramatically. Volcanic eruptions through coal beds or massive sulfur releases that turned the oceans anoxic are just a few of the options. There was a largish decrease (4%) in C^13 at the extinction time (the attribute this to dumping ancient carbon into the air), but much larger (10%) and suspiciously linear swings later were associated with increase in biodiversity. The Siberian Traps eruptions starting kicking in about the time of the Permian extinction and kept going for 6 million years.

There's some evidence for serious anoxia in the deep seas that started well before and got worse well past the extinction boundary--with the same sort of thing playing out on a smaller scale in the shallow seas.

Maybe. Figure 1 shows that the record is pretty spotty about that time.

And before I start believing that it took 10 million years to recover biodiversity, may I respectfully request some error bars on the numbers? Numbers for numbers of species, numbers for estimated vulcanism, anoxia rates, etc are all presented as 100% gospel with no effort at error estimates.

But if you can, by all means look at that Figure 1. Even allowing for a great deal of uncertainty, there's a lot of strange and dramatic history laid out there.

Update: You can't--the image at Nature is too small. This is easier to view: but I'm not going to include the captions--I don't think they'll stretch fair use that far.

Monday, May 28, 2012

As I was Saying by Chesterton

Chesterton again: As I was Saying

Read it. Some samples from chapters

About Germany; relevant to the recent crisis in Greece and Spain where a recent report says the Greeks do work hard

As a matter of fact, that industrial type is not generally any more industrious, if so much, as what we used to call the idle and lounging peasant of the South, who works hours before any of us dream of waking up, and sometimes hours after we go to bed; but rests in the heat of the middle of the day, not being a born fool. But, anyhow, in so far as it is true that the Germans are very industrious, did you ever hear of anybody loving anybody merely because he was industrious?

About pacifism

Men who have no intention of abandoning their country's wealth, not to mention their own, men who rightly insist on comfort for their countrymen and not infrequently for themselves, still seem to have formed a strange idea that they can keep all these things in all conceivable circumstances, solely and entirely by refusing to defend them. They seem to fancy they could bring the whole reign of violence and pride to an end, instantly and entirely, merely by doing nothing. Now it is not easy to do anything by doing nothing.

About traffic congestion (the two seasons of Wisconsin are Winter and Road Construction)

In the middle of a prolonged block in the Uxbridge Road, I have been known to exhibit a gaiety and radiant levity which has made me loathed and detested for miles round. I always feel a faint hope, after a few hours of it, that the vehicles may never move on at all; but may sink slowly into the road and take on the more rooted character of a large and prosperous village. Perhaps, after all, it is thus that our culture may return to the stability and sanity of the earth, which is now its only hope. I have sometimes felt inclined to get out of the car and make a little garden just outside it, staking out a claim and symbolically renouncing all hope of any further advance.

About censorship and culture

I do not believe in ignoring the Pagan morals all around us: it does not diminish the Paganism; and it only deprives us of the pleasure and advantage of denouncing it as Pagan. The assumption that tradition, and even convention, that virtue and even Victorian virtue, is still the rule, and anything else an exception, is all on the side of the sophists who defend vice. It is a rule by which we carry all the unpopular emblems of power, while they enjoy all the practical fruits of victory. They can flout us, because they profess that there is nothing to conceal; and we cannot fight them, because we pretend that there is nothing to fight. But, above all, from the point of the honest orthodox, the present one-sided truce has this enormous disadvantage: it prevents us from pointing out the one solid, staring, stupendous fact which is before all our eyes. It is the fact that we
have not only seen a modern materialist civilization rise, but we have seen it fall. We have seen industrial imperialism and individualism a _practical_ failure. It is no longer a question of using the modern machinery; but of cutting loose from the wreck of it.

About shamelessness

Do not be proud of the fact that your grandmother was shocked at something which you are accustomed to seeing or hearing without being shocked. There are two meanings of the word "nervous," and it is not even a physical superiority to be actually without nerves. It may mean that your grandmother was an extremely lively and vital animal, and that you are a paralytic.


But for about fifty or sixty years nearly all our culture and controversial trend has been conducted on the assumption that, as long as we could get used to any sort of caddishness, we could be perfectly contented in being cads.

About Voltaire and Frederich the Great; with considerable insight into domestic politics

All Christian history began with that great social occasion when Pilate and Herod shook hands. Hitherto, as everybody knew in Society circles, they had hardly been on speaking terms. Something led them to seek each other's support, a vague sense of social crisis, though very little was happening except the execution of an ordinary batch of criminals. The two rulers were reconciled on the very day when one of these convicts was crucified. That is what many people mean by Peace, and the substitution of a reign of Love for one of Hatred. Whether or no there is honour among thieves, there is always a certain social interdependence and solidarity among murderers; and those sixteenth-century ruffians who conspired to assassinate Rizzio or Darnley were always very careful to put their names, and especially each other's names, to what they called a "band," so that at the worst they might all hang together. Many political friendships--nay, even broad democratic comradeships, are of this nature; and their representatives are really distressed when we decline to identify this form of Love with the original mystical idea of Charity.

Why the armor?

We’ve all seen pictures of Dunkleosteus, but refresh your memory with this:

Vicious-looking fellow, isn’t he? Those face bones would allow strong muscles to attach. I’m not expert enough to spot the tell-tale attachment points and estimate the bite strength, but others estimate it at over 1000 pounds of force. Apparently a lot of the prey was armored too. They could be about 20 feet (6m) long. Some varieties had bony plates all around the body, and some of them were 2 inches (6cm) thick.

But the thing that always seemed odd to me is the bones around the eye. Why? It isn’t as though you need huge strength to turn an eyeball in its socket.

On the principle that there’s a reason for the way things work, I assume that there’s some good reason for putting bones over part of your eyeball and pretty much all of your face.

If strength isn’t needed, it seems most likely to be for protection. The protection is pretty far-reaching, so the threat must have been extensive also.

  • We’ve no evidence of anything larger in the sea with it—no bones, at any rate. That doesn’t mean there may not have been giant jellyfish or boneless tentacled beasts with a taste for fish, though how they might contrive to eat anything so large, even unprotected by bones, without at least a beak to gnaw with, is hard to design. A jellyfish couldn’t digest quickly enough to avoid being a target itself for scavengers. A keratin beak might disintegrate but still be fairly useful. So… maybe there was a boneless fish eater that had tough but non-bony gnawing parts and was big enough to be a threat to mid-sized Dunkleosteii.
  • Maybe they ate each other too. There are scar marks on some Dunkelosteus fossil plates that match Dunkleosteus "teeth." That doesn’t quite explain the eyebones, though, unless they used fins or something to blind or distract each other. Which doesn’t seem very plausible given the way fish usually seem to fight.
  • Maybe some of their prey objected to being eaten and lashed out with some kind of limbs or stingers. There’d be very good reason to armor up at least the head and eyes to avoid having chunks taken out by irate dinner prospects. We haven’t found such tentacles, but we very well might not. I like this option myself, not because it is obviously more likely but because it is more dramatic. Set the scene—a mollusky thing with an alien fur of tentacles around its opening is minding its own business when the water stirs and a fish appears trying to get a good grip to crush its shell. Mollusk goes into "full-flay" mode lashing with stingers to chase it off before it gets a good purchase.
  • Maybe there were vicious parasites. I remember seeing a grim picture (National Geographic) of a whale’s eye with some small (1cm or so) feathery arthropods stuck in it, and being very grateful for having fingers. We’d not find any relics of such things, any more than we’re apt to find fossil mosquitoes (except in amber), so they might have been rare or ubiquitous.
  • UPDATE: Or the bones of the face flexed so much when chomping through shells that the strain on the eye would have been painful unless it was buffered by connecting the eyes to a "floating" ring of bone. This would mean that appearances are deceiving and the eye was much smaller than it appeared.

That’s a fairly wide selection of options, with no obvious way to decide. And I can’t.

What brings up this meditation on possibilities is a set of the "Walking with" books Youngest Daughter brought home from the library. (Walking with Dinosaurs, Beasts, Cavemen, ...) The modeling is excellent, the settings very well done, the story arcs catch your interest, but speculation is too kind a term for the result.

They describe one of the species of "cavemen" as having no whites in the eyes, so they could not read each other’s emotions easily, so they did not have complex societies or interactions. (And they know this how, exactly?) That’s just the most egregious of the fantasies they indulge.

They pick the most dramatic possibilities (I have a liking for those myself, see above), but there’s no suggestion that this might be utter nonsense.

But of course it is BBC, and it looks realistic, so it has to be real, right?

Sunday, May 27, 2012

Wisdom again

Age is supposed to bring wisdom. I’m not notably wise now, but I’m a bit more so than I was when I was young, so perhaps it’s true.

But what is the difference? Most of what I know now I heard long ago—I just didn’t think much of it.

Faithfulness, for example: I heard sermons on it when I was a babe in arms, though, now that I think of it, not much on how faithfulness played out in everyday life. Hmm.

Woody Allen said "90% of life is just showing up." Just showing up, day after day, is one aspect of faithfulness. When you persevere you make the choice day after day after day to ratify the original decision—a choice that spreads out over years and flavors all of your life. We’re not eternal, but faithfulness makes our lives reflect a little eternity.

I didn’t think much of that when I was young—I didn’t think of it at all. Since then I’ve had some experience of having vinegar on the teeth and smoke in the eyes—and realized that I’ve been that to other people. And I’ve seen what’s happened to the iffy. All this I sort-of knew before, but I hadn’t experienced it. Or I hadn’t internalized it.

What does it mean to have "internalized" some knowledge? We know the difference between rote and internal, but it is hard to articulate exactly how the knowledge has been made integral to our way of thinking and acting instead of just being something we can take or leave. Language is frail sometimes.

The song I’m glad there is you describes "this world of overrated pleasures and underrated treasures," and somewhere along the line you start to realize how underrated some of those treasures were—even by you, who gave lip service to knowing better.

"With knowledge you pat yourself on the back for being so smart. With wisdom you kick yourself in the butt for having been so stupid." Peter Sinclair

Thursday, May 24, 2012

The first time is eerie, the N'th time banal

Haggard can be a fun read. He's the father of the mysterious lost land genre of adventure story, and although he's very far from PC he always tried to make sure that the natives were human--some vile and some more noble than the hero.

King Solomon's Mines introduced Alan Quatermain, who proved popular enough that Haggard wrote over a dozen more books about him. Haggard killed him off in one book, but wound up writing more anyway. And She had several sequels as well. Magic and reincarnation and Isis pop up all the time. (And I'm afraid a sampling of sequels suggests that they suffer from the usual deterioration.)

In The Ivory Child he sets the character for one lady by having her make mysterious pronouncements about her and the hero's lives having been connected before they were born, with a "mystic look come into her face."

That may have worked well enough a hundred years ago to outline a fey woman of mystery, but today it conjures images of a bored woman whose life hasn't been exotic enough and who has read a book of pop mysticism. Jarring. I'm not sure I'll finish it.

Didn't think of pictures in time

The past few years Google has found for me the records of many helpful people who found a problem and documented them for the aid of others.

The driver's window switch in my V6 Saturn (the gas mileage of a V6 combined with the get-up-and-go of a 4-cylinder) failed, so that the window would roll down but not up. This is unacceptable in winter. The shop said it would be O($200) to repair it. I found a few step-by-step fix-its on-line; none concerning the model I had.

I popped the switch console out and fiddled with it, bought a dental-probe affair and disassembled the last bits of the switch, cleaned out the melted plastic and swapped switches and it works.

And, after snapping it all back in place, I realized that I had missed the chance for a public service. I forgot to take pictures.

Without pictures... I'll try anyway.

Fixing the power window switches in a Saturn V6 console.

A switch console pries out easily from its location between the front seats, and if you can't figure out how to get the cable unconnected yourself just push it back in and don't bother reading any further.

The switch console has two rocker switches (and a lock switch as well if it's the driver's side console). It is made of an electronics board sandwiched inside three snap-together plastic housings, each held to the next with 6 tabs. A thin flat screwdriver and a little twisting will detach the housings from each other. Pop off the "lock" switch tab and the rocker switch heads.

Standing up high on the electronics board are the business parts of the two rocker switches, each held on with 4 tabs that you cannot pry away without using something like a dental probe. I bought some probes (not sharp) and found the plastic quite soft. I was able to pry the upper part of the switch loose from the tabs on one side easily, after which it came off quickly. I've never seen switches quite like it--the bottom (soldered to the board) part held two channels each with a metal strip (3 contacts underneath it) and the upper part of the switch had two short stubs that pressed against these metal strips--except that one stub was melted and the corresponding strip was covered above and below with melted crud. I scraped the strip and the contacts under it clean using a tiny flathead screwdriver, replaced the bad upper switch with the identical one on the same console, and reassembled all the bits.

Figuring it out took the longest time, and cleaning it came second. Reassembly was about 3 minutes. Strictly speaking this isn't a fix but a hack that swaps a heavily used switch for a lightly used one. But the window works now.

Heroin "statistics"

I went to a school session on heroin use in the area. The bulk of the time was taken up with a movie in honor of a DeForest youth who OD'd last year. The movie wasn't very illuminating--I already know heroin was bad--but his mother (who made the movie) was there and I decided against complaining.

Heroin overdoses are rising, and so are associated deaths--the majority aged between 16-30 and most of those out of high school. They advertise a medication that, delivered promptly, cures an overdose, but there aren't many doctors qualified to "prescribe" it to a heroin buddy. Overdoses are pretty easy to measure: you count the hospital/morgue visits and add a factor from repeat requests for the Narcan. The ratio of overdoses to deaths was about 10:1 in Madison and Sun Prairie.

Not all their statistics were quite as trustworthy. They had numbers on Oxycontin abuse and other drug abuse in middle and high school that are apparently self-reported. Umm, I think I want a little more info.

Apparently Oxycontin is easy enough to get hold of that it makes a good gateway drug.

One other little number was dramatic by its absence: the ratio of users to overdosers. I trotted back and forth a bit online to try to get a ballpark estimate and it looks like of order 50-25:1. (Don't use that number anywhere, OK?) Which would make heroin use remarkably dangerous, so maybe I made a mistake. But if I didn't then that means 250-500 addicts in a town of 16,000. Apparently there are two aspects to habituation: the brain's changes to lead it to expect opiates, and the rest of the body's changes, in particular the respiratory system. The brain takes a year or more to clear itself, but the body recovers in a few days. Which means that after being on the wagon for a week, the same dose you took before could kill you. At least so they say.

The story arc was that middle and upper middle class youth were starting to abuse Oxycontin et al in school, but finding the supply limited, were migrating to heroin. The supply of heroin is greater, but it winds up costing a great deal. Upper middle class kids typically were preying on family and friends, and turning to dealing.

The numbers weren't quite consistent from slide to talk. At one point they were talking about doses for \$25, and another for \$150. A factor of 6 seems pretty large to be a habituation effect.

The state is into "harm reduction" and provides paper lunch bags with heroin kits, and encourages addicts to never shoot up alone. I wonder what the overall effect is. The harm for any given addict is presumably less, if he isn't sharing infected materials, but if the addiction rate rises it isn't instantly clear that we've reduce the harm. (I'd not expect the rate of addiction to rise immediately. Instead I'd expect a period in which the bags slowly grow more normal in the background, and then a rise.)

Simple googling didn't find me the sorts of n-tuples I like to work with, so I'll let this be a back-burner mystery for a while longer.

Monday, May 21, 2012

General Philosophy by David Elton Trueblood

One nice thing about vacations is you can catch up on reading. I don’t like to think how long ago I first started this book.

This was part of my father’s library, and has some of his underlining in a couple of chapters.

As the title suggests, it is an overview of philosophy. Trueblood emphasized the back and forth questioning of issues, and filled it with illustrative quotations with appeals to read the original author. "The best way toward greatness is to mix with the great." Dialog is one of the best ways to sharpen your own thoughts, and he tries to approximate that in the book.


A.N. Whitehead, The Function of Reason "Scientists animated by the purpose of proving that they are purposeless constitute an interesting subject for study."

Or his own words in the penultimate chapter, on Society

The ultimate enemy is not any of these things or events to which we usually refer. The most terrible enemy is triviality. No society will be a good one, no matter how adequately people are fed and clothed, if it is not a society in which men and women can be made to feel, without deception, that their lives are important.

The ideal social order, then, must include many things, but three are preeminent. It must include freedom; it must include order; it must include a sense of meaning.

The book was an interesting journey through the to and fro of debate about the limits of proof and determinism and chance, and what can be built on the foundation of the certainty of error. (If you know that at least somebody in a debate has to be wrong, what can you infer about the existence of an objective order?)

He spends more time than warranted on the Oxford group that followed Wittgenstein, but their approach (philosophy is merely the study of the grammar and never tried to figure out if propositions are true or false) isn’t so much of a live issue these days. Which was inevitable.

It was fun to see where I fit in the spectrum, but I have a lousy memory for names and I'd have to re-read it to make sure I had the labels right.

Read it. Though perhaps Trueblood would say "Read Plato and Temple and..."

State and Church and Family

So 43 Catholic groups are suing the feds over the health insurance mandate rules.

Good for them. The mandate was unwarranted, and unConstitutional on the face of it. Together with the earlier (unanimously defeated by the Supreme Court) effort to make the government the arbiter of who is and who is not a religious employee, these are strong indicators that the movers and shakers in DC want full sovereignty, with no other pesky institutions competing for loyalty.

I understand why statists (setting aside the power-hungry; bureaucrats and politicians tend to become addicted to power) want a society with a clear power center and no pesky independent institutions: there's no confusion, and the beneficent rule of law can be applied everywhere. But that presupposes that that "rule of law" is always benign, and that the state has the right to override other institutions at will, and not just in emergency.

For example, consider a religion that believes the sun won't rise unless there are human sacrifices. That's an emergency. Consider another that believes that white people are rebellious androids made by a black genius, and that won't let white people join. That's not an emergency by any stretch. Consider another that teaches that you have been invaded by alien parasites and need to spend tens of thousands of dollars on therapies. OK, that's pretty dubious (France and Germany agree, btw).

I don't think the statists are all that concerned with the smaller religions, or even the big ones (like Southern Baptists) that aren't that organized. But the Catholic church really gets up their nose.

Family loyalties aren't spelled out in statutes or canon laws, but what do you do without them? They can go too far, as when jurors refuse to convict someone of their tribe. But if you allow that the state has the right to spell out those limits, the state likewise has the right to modify them at will. And if the culture discourages them substantial fractions of our children are ill-raised. (Easily observed, btw)

The relationships between the various institutions in our society are not clearly defined. Though historically the state and religion and tribe were often united, they were not always, and certainly are not united in the United States now. I wonder how many of us remember what a terrifying horror an utterly sovereign state can be. Or the chaos of multiple tribes in conflict. We hear plenty about the dangers of religion(*)--disproportionate to the real situation here, though it is serious enough in places like Saudi controlled Arabia.

I wonder how united we can remain without some sort of philosophical consensus on what the limits to power are. The old agreement is still available if we want to try it again.

(*) I've argued before that statism is a variety of religion, but I leave that out here for brevity.

Need to know

Kodak had an enriched uranium reactor in a basement for decades. It was a neutron source for research; safe with no danger of melt-downs, and no leaks. "Small plates of highly enriched uranium multiplied the neutron flow from a tiny californium core." Even (perhaps especially) when it was decommissioned there was no whisper of what was going on. After 2001 the rules changed, but even before then the US was very careful about information about the location of enriched uranium. The most dangerous time was after decommissioning and during transportation, when the 3 1/2 pounds of enriched uranium was loose for the first time.

Of course the neighbors never knew, nor did most of the employees, nor, I gather, did the city emergency officials. Nor, of course, did the Center for Nonproliferation Studies.

Hat tip to the Cranky Professor

Friday, May 18, 2012

Forbidden fruit?

I know I should have reported on this before, but I went for an apple.

High fructose diet hampers memory! Well, sort of. It was a combination study looking at the effects of deficiencies in n-3 fatty acids (omega-3?) and added fructose. This changed insulin levels and other chemicals I’m not familiar with, which isn’t a big surprise and not the point of the headlines. The authors posit that excess insulin monkeys with brain membrane fluidity (at minimum—have a look for yourself).

I skipped the chemical level details to go to the good stuff. They divided the rats into 4 samples, with 6 rats in each sample. (Alarm bells) Some got normal diets and some got low omega-3 diets. Each of these groups was further divided into sets that got extra fructose and not. The difference in weight gain was minor. Before the diet, the groups were tested on a maze. Before you ask:

All surfaces were routinely cleaned before and after each trial to eliminate possible olfactory cues from preceding animals.

That’s good, but I wonder how sensitive rat noses are. (Maybe high sugar levels decreases sensitivity?)

They measured "latency" time, which I suppose this is how long it took the rats to run the maze. They trained the rats on the maze for 5 days, gave them 6 weeks of the special diet, and then tested them again one day. Their result was that the high omega-3 diet without extra fructose rats ran at pretty much the same rate as the last time, those with low omega-3 or high fructose didn’t run the maze quite as fast, and those with both low omega-3 and high fructose even less rapidly. The lights over the maze were bright, to encourage the rats to find a way out quickly. (Maybe high-fructose diets leave rats feeling more laid-back and not so irritated by light?)

Figure 1 is very odd. A shows that all rats tended to run the maze in about the same time by day 5: about 40 seconds if I read the graph correctly. B shows that the good diet rats ran it in 20 seconds on the average 6 weeks later. Something doesn’t add up.

Also notice how huge the errors are in A until they converge by day 4. One sample averaged twice as fast on day 4 as on day 5. Those large errors are consistent with the authors’ interpretation of forgetfulness, since one sees large errors on the 6-week later test samples also.

Figure 2 shows the correlation between serum triglyceride levels (and insulin resistance) and the "latency". Except for one flier (is it the same animal both times?) the high and low omega-3 diet (both low fructose) rats seem to have the same speed running the maze. It is interesting that the insulin resistance level seems to have a cleaner correlation with "latency" than the serum levels do. I suppose that’s consistent with the authors’ interpretation that excess insulin changes the brain, so the overall resistance, a proxy for the history of insulin exposure, would be more important than the triglyceride levels of the moment.

This is flawed, and does not justify the headlines. It needs more work, and to address some of the alternative interpretations I mentioned.

I hope it doesn’t pan out. The folks who know what’s good for you (FWKWGFY) already say we shouldn’t eat meat (canine teeth must have no reason for existence), and now fruit is supposed to be bad too. Next year it will be vegetables.

Thursday, May 17, 2012

Inexact computers

Let's see if I can do this one without looking at the report. Researchers found that with pruning computers could be made much more energy efficient at the price of some accuracy.
The concept is deceptively simple: Slash power use by allowing processing components — like hardware for adding and multiplying numbers — to make a few mistakes. By cleverly managing the probability of errors and limiting which calculations produce errors, the designers have found they can simultaneously cut energy demands and dramatically boost performance.

Let me guess.

Digital devices like computers use electrical pulses to represent bits. These are typically square wave pulses. When the clock tick comes, if the voltage is beyond some threshold (WAG 1.6V), that's a "1", otherwise it's a "0". There will always be a little distortion, and even a perfect square wave at one end of a bit of wire will be a bit blurry by the time it reaches the end. Over a long enough wire the blurriness might drop the voltage at the clock tick below the threshold, and a "1" could turn into a "0". So you want the voltage enough higher than the threshold value and the distances short enough to keep things clean and accurate. And that clock tick needs to be clean and sharp too.

But, the more jumping up and down the voltage does the more energy turns into heat.

You can cut down on the heat by lowering the voltage. If you lower it too much you start to get occasional errors. Tradeoff.

I think you can also cut down on the heat by giving up on square waves. A perfect square wave is a sum of the sine wave with the same frequency and an infinite sum of higher frequency waves--and those higher frequency components have to contribute to the heat. So if you use a blurrier wave you could be more efficient too--but again at the cost of some accuracy.

Sometimes accuracy isn't that critical--you can get "good enough" results, as the article shows. For those kinds of applications, this can be very useful--but don't try it for your orbit calculations.

So. Let me go to Rice and have a look...

One example of the inexact design approach is "pruning," or trimming away some of the rarely used portions of digital circuits on a microchip. Another innovation, "confined voltage scaling," trades some performance gains by taking advantage of improvements in processing speed to further cut power demands.

Ok, I'm not a computing guru. I completely forgot about error correction circuits. And I'm not sure if they mean the same thing by "confined voltage scaling" as I suggested above. I can't find their paper at the conference web site :-(


Neutrinos are so hard to detect and so low mass that one question that you'd think was easy hasn't been answered in the 80 years since Majorana proposed it: does a neutrino have a distinct anti-particle? Electrons do, no question, likewise protons, muons, and so on. Photons don't. The "anti-particle" for a photon would be another photon exactly out of phase with it. (Which is to say that the photon would never exist at all, since it would be cancelled at every point by the other photon.) You could say similarly that a π+ and π- are each others' antiparticles; they are composite particles and their quark content is the mirror image of the other (up and anti-down quark vs anti-up and down).

But the neutrino is so hard to pin down that we can't tell. What would you get if two did annihilate? They're neutral, so photons don't couple to them very easily. (Photons can couple to "W" particles which in turn couple to neutrinos, but adding more links in the chain makes it less likely to happen.)

You can try a trick.

But first...

When a nucleus has a radioactive decay that releases an electron (or a positron which is just an anti-electron), it also releases a neutrino. Since momentum and energy are conserved, the recoil of the nucleus and the energy of the electron tell you about the energy of the neutrino (which... ooops... you didn't see, so we call it "missing energy" Creative names, right?). Put the atom in a crystal lattice so it can't recoil easily (so its kinetic energy will be trivially small). That makes the problem much easier; it is more like a 2-body problem. Now measure the energy of the electron. Do it again and again and again and look at the distribution. If the neutrino's mass were a little larger you could see the edge of the distribution of electron energies drop to zero a little faster than if it were massless. That's a tough experiment, and hasn't been successful yet. We know the neutrino has mass, but it is so small that this method hasn't spotted it yet.

Now pick something even rarer: nuclear decays that toss out 2 electrons at once (not one after another!) Only a handful of nuclei do that. Most of the time you expect 2 neutrinos as well, but if the neutrino is its own antiparticle sometimes the neutrinos will annihilate each other and you only get 2 electrons, with no "missing energy". These experiments have been tried for years without success, but there's a new one that's a little more ambitious. It is still a prototype, but the final system (if it works well enough to build) will have a ton of germanium-76 crystals inside multiple layers of shielding and detectors. Natural radioactivity in the rock will add a background of random energy in the system unless you shield against it carefully, and cosmic rays dump in energy and sometimes catalyze nuclear reactions to boot--so you want to detect them on the way in and veto against them: "A cosmic ray just went through the system so don't look at anything for a few microseconds."

If Majorana was right the Standard Model of particles will need some revision. We know the model isn't complete, but it works very well anyhow. Any revision has a hard row to hoe to do as well.


A mathematical model says 9% of mammals in the Western hemisphere won't outrun climate change. I can't find the original article--I don't think it has been released yet. They especially worry about monkeys. There've been a few substantial cold snaps in the not-too-distant past, and some of the transitions are supposed to have only been a thousand years or less. The missing bit here is "how long are these researchers assuming the changes will take?" 100 years? 1000? If 1000 years, I wouldn't believe them--the fact that the animals are there at all says they can weather changes like that.

Wednesday, May 16, 2012

Outdoor services

A knot of about 80 people were on the west corner of the capitol square singing songs about recalling governor Scott Walker. The second one I heard was based on "Down by the riverside" and the first was also based on an old spiritual.

At first I thought it was a bit odd, but on reflection realized that they were still religious songs, rewritten for a different religion. No doubt there was a sermon too.

Monday, May 14, 2012

How sharper than a spider's tooth

How does a spider bite work?
Although their armour consists of the same material as their predator’s fangs, flies, grasshoppers and other insects that are the usual prey of spiders have little to offer by way of defence against the spider’s bite.

The answer, according to some Max Plank researchers, is that the fang's layered and reinforced structure beats ordinary layers of chiton. I've never seen a spider fang close up before: go look.

There are still lots of obvious questions lying around loose for people to study. A few years ago someone asked: why do fingernails tear across rather than randomly (in general), and found that they're made of layers of oriented proteins. Long before the advent of nail clippers or scissors, we needed to have fairly reliable nails... And I, like probably vast numbers before me, remember staring semi-crosseyed at bathroom tiles to get a 3-d effect--but never went the extra step to try to make 3-d images (Magic Eye).

So I ought to have suggestions--they'd be obvious, after all...

In the dirt at last

The tomato and pepper seedlings have been popping in and out of the house for the last month. "What's the forecast? Bring them in." But the time has finally come to give them a home, and the garden has some new denizens: "One for the cutworm, and one for the crow, one for the taxman, and one to grow." Or something like that.

The calendulas returned so prolifically that there were enough to edge a whole other section of the garden. I'm not sure what calendulas are, but I think I got the right patch.

The maple roots are strangling one garden so we're moving tulips out before turning it and adding some decent dirt again. I hope tulips don't need a lot of depth--I landed some above a buried stump. (The other buried stump is easily spotted by the luxuriant crop of mushrooms.)

My better half says that children should learn to garden: "There's no better way to make the connection between hard work and eating."

Saturday, May 12, 2012

Survivalist Spectrum

Just for laughs I googled "survivalist spectrum" and wouldn’t you know it, the spectrum that topped the list was declared to be sort-of political (the sustainable vs the well-armed).

What sparked the query was a nagging question that finally surfaced: "Why do people buy gold shares?" I’ve seen them advertised for several years and didn’t pay much attention. The ads are getting more pervasive, and I finally spent a minute thinking about them. Gold is one thing, gold shares another. You can hide gold in the backyard if you like and it’ll have some value in almost all scenarios, but gold shares have no value unless there’s still a functioning banking system and the government hasn’t gone all Order 6102 on us.

Governments suffer the besetting temptation to seize things that don’t belong to them, and bullion stock in gold shares firms beckons. Thus gold shares are only really valuable until the government starts hurting for income or starts soak-the-rich theater (only against the non-cronies, of course).

Even plain old gold isn’t always as useful as advertised. Imagine—TSHTF and you’ve got yourself a pantry of canned goods and trading booze, and a couple of Krugerrands. You need a new axe handle and some nails, so you go to what’s left of the lumberyard to buy some. What are you going to get in change? Do they even carry that much change? You need a bank to get full use of your gold, and some sort of currency system.

The spectrum I had in mind was more along the lines of the degree of preparedness and how major an upheaval you expect.

For the moment ignore the folks who don’t prepare.

At the near end you have those of us who keep a week’s worth of canned goods on hand and have an emergency kit available. This is "dual use" preparedness: good for when a blizzard shuts down the roads and also good for camping trips. We have a lot more of the latter than the former, fortunately. And no gold, so don’t bother burglarizing the place, OK?

At the far end you have the fortress farm with a jack of all trades who has ammo for a hundred years. You hope he never gets appendicitis.

Scenarios abound, of course. What sort of assumptions do you make? Will there still be a flow of fertilizer and pesticides at prices farmers can afford, and fuel for food transport? I can’t easily find a listing of fertilizer plants, but it isn’t hard to come up with scenarios in which the number is regulated back and the price turns unaffordable; and in a major downturn there might not be any way to build new ones in time. That would be very bad.

If there are jobs but not good ones, with just enough food to keep body and soul together, then knowing how to make do and having neighbors you can trust is the largest part of preparation. And being adaptable. We’ve been there before—most of the world lives like that.

If you just think this will be a 10-year downturn and we’ll be back on our feet again afterwards, then it makes sense to invest in business property and congressmen, and try to hang on. If you think cities are headed for Detroit-ization, business property is the last thing you need.

I’ve already written what I suspect about the zombie apocalypse.

Thursday, May 10, 2012

Lensing gamma rays

Gamma ray lenses? A team at Ludwig Maximilian University of Munich created a silicon prism that slightly deflected gamma rays with an effective refractive index of 1.000000001. Using gold instead of silicon should increase it quite a bit: If this is Delbruck scattering as they think, then since the scattering goes as Z to the fourth power (Z is the atomic number of the nucleus), the refractive index would go to 1.000001. Not big, but it could be useful.

X-rays can be focused, apparently using multiple very small lenses, or using "zone plates" (or see this). I think you could get better luminosity by using x-ray channeling in slices of crystal tilted at angles that increase with distance from the center of the lens--but I never figured out how to fabricate one easily/cheaply, and the best focus you'd get would be of the order of the size of the crystal slices. (I was thinking of uses in dental x-ray systems, where you could increase the intensity in the interesting direction without increasing the power and total radiation dose.)

Update: FWIW, I didn't remember anything about Delbruck scattering when I read this; all I know I learned from looking it up for the occasion. And the "SciTechDaily" site made hash of the story, so I didn't link them. Gamma ray wavelengths are smaller than the size of atoms, so it isn't surprising that any sort of bending would be small.

Wednesday, May 09, 2012

When a mammoth isn't

The story seemed surprising: I'd heard years ago that Cretan mammoths were tiny. But it is alleged that the original story was based on flawed DNA analysis. The detailed study of teeth and fore"arms" confirms that they were mammoths after all. 3.7 feet high and 680 pounds isn't quite terrarium size, but that's about 1/30 the size of its northern brothers.

I remember the old article claiming that on one of the islands the remains were only about 5000 years old, and another (much older) article suggesting that the cyclops legends rose from elephant skulls (which look sort of like there's a big hole in front which might be for an eye).

Tidal heating

An astrobiologist at UW-Seattle claims that the "habitable zone" (where water can be liquid) around red dwarfs isn't quite as large as you'd guess from naive luminosity estimates, because the planets have to orbit so close to the cool star that tidal forces will heat the planet's core. As the planet rotates different parts get squeezed (the popular article fails to mention that, btw), and friction heats the rock; at least until the planet's rotation decreases to the point where it keeps one face at the star all the time (tidal locking). Then the only squeezing comes from the different forces at different times in a non-elliptical orbit (which the article does describe).

It seems plausible. But the devil is in the details. Some back-of-the-envelope with how long tidal locking would take with an earth-sized planet at half Mercury's orbit and a star 1/10 the Sun's mass gives about 6E10 years. If that's the timescale for heating I don't think the temperature would rise much. But I may be leaving something out, and there's a fair bit of tweaking room. For example, at 1/10 of Mercury's orbit, the time scale is only 4 million years. And my model of an iron planet isn't very realistic (I only get 100 degrees C rise). It would take a few days work to research a decent model, and it is getting late at night...

Tuesday, May 08, 2012

Growing electronics

Magnetic bacteria may help build future bio-computers! If that were so, "your computer has a virus" might be an even more disastrous diagnosis than it is now. Instead the BBC story suggests that one can use bacteria to grow electronic components. After all, if the bacteria can grow little crystals of magnetite, can't they be induced to grow larger structures? Think slime molds and go nuts.

Of course further down in the story you read about bacteria making tiny parts of disks-- presumably tiny read/write heads--without details about how you persuade them to add coils and electrodes.

Details, details.

For the English teachers out there

The stages of grading:
Stage II presents with mild but steady localized pain, mostly along the GI tract, and an inability to concentrate. Despair is still contained, but it’s eyeing the lymphatic system’s freedom train. Women are "co-modified." Men are "discluded." Role models are "immolated." Passages are "taken out of context due to objective reality." "Often times" is everywhere.

Pull your forelock

I was going to marvel at the thoughts of our betters in the Massachusetts school board, but Texan99 beat me to it and said it better.

Bake sales are so middle-class...

Sunday, May 06, 2012

Football explained

This is the only collection I could find, of Peter Sinclair's earlier work. He got better with time.

I guess the poor guy's world-view didn't have room for threats from outside: he descended into Trutherism after 911 and I haven't heard of him since. A pity--he did some nice work.

Friday, May 04, 2012

Don't eat me, CD47

The "New drug could potentially shrink and cure all tumors" headline is a bit excessive, though the report is very interesting. Possible side effects will immediately occur to you.
The treatment uses an antibody that blocks the 'do not eat' signal that's usually displayed on tumor cells and coaxes the immune system to destroy cancer cells.

Leukemia cells produce higher levels of the CD47 protein than healthy cells. It's a marker that will block the immune system from destroying healthy blood cells. Cancers take advantage of this by using it to trick the immune system into ignoring them.

CD47 is found on every human primary tumor that the team has tested. The researchers transplanted human tumors into mice. Once the rodents were treated with anti-CD47, the tumors shrank and did not spread.

I don't understand their Figure 2A/B on survival rates with high and low CD47 tumors, which don't seem consistent, and not all the approaches gave statistically significant results.

And (drum roll please) the tumors were human tumors in mice (at least if I understand what "xenotransplantation" means). So once the "don't eat me" proteins were gone the mouse phages should be pretty good at getting rid of alien cells. The next test should be with mouse cancers, though that will be a longer study. And then long term studies to see what gets accidentally eaten during the treatment....

Zombie ants

The pictures of zombie ants are gruesome enough for the M rating ("not before Meals"), but the report is encouraging. Ant colonies are defended against the parasitic brain-controlling fungus by a regime of thorough grooming and by another parasitic fungus in ants that attacks the brain-controlling one. "Because the hyperparasitic fungi prevents the infected zombie-ant fungus from spreading spores, fewer of the ants will become zombies."

When did you last wash your hands?

Wednesday, May 02, 2012

International Law

There seems to be one fundamental principle in international law: encapsulation. You limit the number and complexity of interactions with other groups by channeling them through a recognized interface--the sovereign government of the region where those people live.

If a foreign ne'er do well steals from you, you don't ask the Berwyn police to find him, you hand over that exercise to the police there, and ask the judge there for compensation. It simplifies life considerably, and avoids a lot of rough edges and anger at interfering strangers.

Of course it doesn't always work right. Sometimes you find egregious evils that you can't do much about without violating encapsulation. And sometimes one or more countries (it only takes one) figure a war is worth the risk. That also is supposed to be channeled through recognized interfaces, where the armies do the fighting until some recognized authority gives up or arranges a treaty.

All bets are off if a country loses a war and those "interfering strangers" get to do what they like, including adjust the loser's laws. Sometimes this makes the losers slaves, sometimes it (e.g. Japan after WWII) makes them a better country.

The encapsulation (sovereignty) model doesn't seem to work very well with groupings like Pakistan or Somalia. If they were peaceful nobody would worry; and if they were wholly occupied in molesting themselves and didn't bother outsiders we could look the other way and pretend it wasn't happening. But they often do attack outsiders, and are effectively sanctuaries for "international criminals."

That leads to several questions.

How do you identify such a situation in a non-arbitrary way? How do you select the representatives of the group that is causing mischief? And what restrictions are you under when you intervene?

Somalia doesn't have a government; it has a low-level civil war with some outside powers taking a hand here and there. It is a collection of closely related peoples who don't like each other: a "nation" in the old sense of a tribe and a language, but not in a sense useful to themselves or others.

One warlord can decide that he's going to attack Americans. We can complain to the paper central government all we please; nothing will happen. We can collect forces to go do battle--but who do we fight? We've no beef with the rest of the Somalis, just the one warlord and his minions. We can, thanks to superior technology, send a drone in to blow him up--or blow up somebody, anyway. Our humint is often very bad in rotten corners of the world where we don't speak the language and everybody lies to outsiders.

Where are you able to make the distinction between that and an Iranian government, weary of dealing with the futile atheists in British government, sending a drone to kill Salmon Rushdie?

The distinction seems obvious, but I'm not at all sure how one can make the distinction precise--as precise as "work through government channels." If there were an international way to decide that nobody recognizes the existence of a government in area Z, that might be a method, but all we have is the UN. Big joke. (To play devil's advocate: Iran could point out that Britain fails to punish criminals but afflicts citizens who try to defend themselves, is failing to teach its youth knowledge manners or morals, and fails completely to enforce laws found around the world against blasphemy. It sounds like a failed state when you put it that way, doesn't it?)

Who regulates contracts in the tribal areas of Pakistan? The local jirga or Islamabad?

In a region of mixed tribes it isn't clear who to deal with or how. The local warlords threaten families because it works. Ought we, in wild areas of the world, pay assassins and threaten families the same way, to focus the retaliation on the enemy and not the other bystander tribes in the area? The traditional alternative is sending in troops that shoot up the area and make the bystanders mad too--especially if we need troops to hang around to keep tribe Y under control. Afghanistan comes to mind.

Take Mexico as a nearby case. Sections of it are allegedly no longer under the control of the central state, or even of the local government, but instead under the sway (but not detailed control--they don't run schools or courts: see Britain above :-) ) of what are universally recognized as criminal groups. If US citizen are regularly attacked, when does the US have the right to send in forces? (When in hot pursuit?) Maybe peel off a section of the country and administer it ourselves to put things right?

This seems a can of worms. I can see why people keep trying to shoehorn every grouping of other people into something resembling a government.

I started this as a comment, but moved it higher to expand it.


The Anchoress has a warm post around George Will's column about 40 years with his son Jon. She says it well, and so does he.
This year Jon will spend his birthday where every year he spends 81 spring, summer and autumn days and evenings, at Nationals Park, in his seat behind the home team’s dugout. The Phillies will be in town, and Jon will be wishing them ruination, just another man, beer in hand, among equals in the republic of baseball.

Reporting around a hole

The Wisconsin State Journal reported that the local bishop had sent out a letter warning of penalties for those who spread rumors and gossip. I gather this has to do with a dispute about some priests the locals don't like (the priests insisted on rules like boys-only for altar service and rubbed some people the wrong way). If you read the article you glean all sorts of details--except what exactly the bishop meant by "rumors" and why he came down hard on the subject.

I wonder why they ran with a story that never got around to getting a statement from the bishop. I can easily imagine him saying that he didn't want to give the rumors any further publicity by quoting them, which would probably be translated into "no comment" by the editor, but to have nothing at all is very odd.

I'm protestant, without a dog in the hunt about "lay ministers" in Catholic services. I don't think there are any other kind of ministers (alternatively, all ministers are holy), but I find the spokes-Catholics who think the same way to be uncongenial company. Maybe this is a spokes-critter bias (the noisy shout down the kind), or maybe it is a Madison-esque clustering of attitudes.


I know an on-the-scene witness to this story about a police standoff. Other little details not in the story include: 4 families having regular violent arguments, attempts to intimidate the police, police and sheriffs and child protective service officers all around, police patrolling all around all the buildings past 1am, police still on the scene into Wednesday afternoon, and a general sigh of relief when the violent pit bull ate lead. (Yes, it was one of those trained to be violent--sorry, pit fans.) This iceberg has a lot below the water, and we'll probably never hear the whole story. But we can guess the shape pretty well.

Evolution and reporters

The headline reads "Human Evolution Still Tied to Darwinian Selection". The study used the records of "6,000 Finnish people born between 1760-1849". If you are going "? What systematic variation can they measure in only 4 generations?!?" you're with me.

The abstract concludes a little differently:

Our results emphasize that the demographic, cultural, and technological changes of the last 10,000 y did not preclude the potential for natural and sexual selection in our species.

Not the same thing at all. What they actually measured was

Individual differences in early survival and fertility (natural selection) were responsible for most variation in fitness, even among wealthier individuals. Variance in mating success explained most of the higher variance in reproductive success in males compared with females, but mating success also influenced reproductive success in females, allowing for sexual selection to operate in both sexes. The detected opportunity for selection is in line with measurements for other species but higher than most previous reports for human samples.

This team includes some of the same researchers that claimed that having boys reduced the mother's lifespan, even if the boy died in infancy--which was tentatively blamed (by Lummaaa) on testosterone's effect on the mother. Although she admitted it was clearly more complicated than that.

Do they have solid rings?

Astronomers at Vanderbilt claim to have identified 675 rogue stars they claim look as though they are inner galaxy stars (more metal because of recycling in the denser center) that were ejected by the black hole. They're red giants. If they came from the core they took about 10 million years to get here, which they suppose means they were yellow stars when they headed out, since apparently red giants don't last more than a few million years. The theory is that a binary system (of which there are plenty) falling into a black hole stands a chance of splitting the kick so that one falls in faster and the other is kicked away fast.

That's quite an addition (if accurate) to the 16 candidates astronomers already have. I can't get at the article yet. I'd like to know why they looked for red giants particularly. If the high "metalicity" ratio (of metals to H and He) they're looking for shows up best in red giants, there are probably a whale of a lot of other slingshot stars around. Of course it could be that they have their estimates for the tail of the distribution of metalicity wrong and these are ordinary stars. And (I wish I could find the report) the popular article didn't mention any velocity measurements. In fact, it said they were checking to see if these were "unusually red brown dwarfs instead of red giants. Because brown dwarfs produce a lot less light than red giants, they would have to be much closer to appear equally bright." Which could mess up velocity measurements...

UPDATE (3-May) to clarify what I mean by the tails of the distribution. The ratio of heavier elements to hydrogen/helium is going to vary star by star. Those that form in areas full of dust from vanished stars will pick up more. A few years scanning stars will give you a distribution of ratios. There'll be some average and a distribution about that average that typically looks a little like a bell curve (if you squint or only look at the part near the average). Far away from the average you typically get tails that don't look like a bell curve's: they drop off too fast or wayy too slow.

The point of their exercise is that these distributions aren't the same in the core of the galaxy; the average is higher. So if they underestimate the number of ordinary out-in-the-boondocks stars that have high "metalicity" they'll interpret them as "they don't belong there" type stars.

There are other possibilities as well. For example, if a red giant is in the process of absorbing a few planets, that could drive up the ratio. How much? That depends on how fast planets evaporate and how fast convection is in the outer layer of the star.

The team looked in a relatively clean area between us and Andromeda. Astronomers suspect that the Milky Way has indulged in collisions with other galaxies and clusters. These stars could be relics of some dispersed dust cloud from the center of a star cluster that collided and went away.