Sunday, September 25, 2011

My take on faster than light neutrinos.

Clearly what's happening is that EM travels through some medium we never accounted for even in a vacuum (what's a real vacuum in this universe anyway?). Neutrinos are so weakly interacting that they can completely ignore all everything and go at c. It's not that c is wrong but rather that we've been wrongly measuring it by measuring photons in a pseudo-vacuum. Whereas neutrinos are so largely unaffected by things that they come much closer to the actual value of c, because vacuum-spacuum you can send them through the Earth and they'd never hit a damned thing. My guess is that our typical measurements of the speed of light do not take into account the index of refraction of spacetime due to zero point energy. Which would be experienced by photons but not by neutrinos.

That or they're just wrong.

But, heck any medium does slow down light and less interacting bits of even the EM spectrum get slowed down at different rates. This is dispersion and it's the reason we get rainbows. Because different parts of white light are slowed down by different amounts. So in air you will find that radio waves can travel faster than the speed of light in air. Really a few scientists managed to stop light completely, and before that slow it down to a crawl. Heck, I can run faster than light in such cases. But, I certainly don't run at c.

Thursday, September 22, 2011

Comments: Species Demarcation.

Kay asks:
Explain this for me if you will. The other day I saw a post from a creationist stating that the Theory of Evolution has to have some fault in it because there is no interbreeding of species. How in layman's terms did the first humans evolve from apes and then proceed to separate themselves from the apes? At what point did breeding with apes stop by humans?


That's not the theory of evolution. That's a question of taxonomy and demarcation. It's a real problem, but mostly with the way we typically think about such things, as oppose to the reality of evolution. Evolution doesn't directly deal with speciation, or how far is far enough to say that one species stops and another species starts. That's like asking where childhood ends and adulthood begins. There's just not a clear point.

Evolution says that the struggle for existence and nature selecting such that the most advantaged individuals have the most children who are similarly advantaged. This ensures that species tend to adapt better to whatever niche (way of living, food etc) they occupy. Evolution says that species adapt to changing conditions of life, and can result in amazing adaptations and solutions to environmental problems. This is all evolution says. Things go forward. But since they go forward in different ways, they also tend to drift apart.

Where the species fork, and part irrevocably is a matter of happenstance and saying how far is far enough, and what a species is, turns out to be a harder problem than one would suppose because gene pools in populations do not adhere to the sort of Platonic ideals. There is no perfect human. There's no absolute ape. There's no abstract rabbit to which all rabbitness is to be compared. That's just not the way it works.

If you go child, to mother, to grandmother, to great grandmother, and so forth and go back 350,000 generations, you'll never hit a point where a child was a different species than it's mother. It just doesn't happen. However, if you do this, with me and you do this with Oliver the Chimpanzee you'll end up at the same critter. My 350,000th grandmother is likely the same as his 350,000th grandmother. And at no point were any of my ancestors a different species than their direct offspring. And at no point were any of his ancestors a different species than their direct offspring. But, at some point, at some time, you could draw a line and say that these distant cousins are now two different species. But, that line would always be arbitrary.

We think there's some absolute human, that somehow adheres to a Platonic ideal of humanness. But, really that's not how things work. I mean, would one consider Neandertals a different species than human beings? They broke off the main human lines some 600,000 years ago. But, it turns out that modern humans interbred with them, and anybody with European ancestry is like 3% Neandertal. So are they really human? Were they always human? The Denisova hominims broke off of the Neandertal stock, and share a common ancestry with homo sapiens some 600,000 years ago too. But, they too interbred with humans in modern Melanesians (people in Oceania, like Vanatu where you get black-skinned kids with blond hair likely due to such influence). And apparently Denisova also from time to time (if a newly found toe bone is to be believed) bred with their closer cousins the Neandertals. So were any of these really different species? Under different definitions there are different answers, because the term species is absolute and evolution is gradual and progressive.

Species in some sense are a sort of hind sight thing. It is certain that at one time all of the rodents and all of the lagomorphs (rabbits, hares, pikas) had a last common ancestor (which DNA analysis suggests lived ~90.1 million years ago). This last common ancestor had at least two offspring. One of which gave rise to all the mice, rats, capybara, beavers, squirrels, porcupines, etc, and the other which gave rise to every rabbit, hare, and pika. Somewhere along the line, you can draw a line, but it would an arbitrary line. And is somewhat irrelevant to the actual history that actually happened with regard to their populations. And more and more this is what we're finding. Taxonomy is largely what we tend to call things, we give populations names. But they aren't that exact thing, there is no exact thing. They gradually change all the time.

This is largely why more and more we're trying to move away from such things. Replacing the typical system of taxonomy with a nested hierarchy so we can say that this last common ancestor of rodents and lagomorphs was a glires and so are all of its offspring. But, this ancestor doesn't necessarily fit into any of the subgroups of it's progeny.

So if for example in the distant future, many animals have died off, and bats have diversified to take over their now open niches. Let's say there are bats that swarm during the day, bats that swim under the ocean, bats the size of mosquitoes that suck blood from bats that that graze in the fields. All of these bats would be different species, but they would never stop being bats. There would be more groups inside the old groups, but the old groups, which would be the species we currently know, wouldn't stop being real. Perhaps the bats under the ocean and the bats that graze in the fields, and the ones that burrow into trees to hunt for insects, would all be fruit bats. So fruit bats wouldn't be a species but rather a group of distinct species (to some extent they are as there's many species of fruit bats), just as the glires are a group that gave rise to every rabbit and rodent on the planet.

Which is a long way around to the question you initially asked. Largely this creationist is confused as to what evolution says and what problems are really problems for the theory. The problem here is with our old ideas of taxonomy and what species are. This isn't a problem for evolution. Reality is what reality is. Our taxonomy needs to fit with reality rather than our general beliefs about names and putting things in boxes (Platonic ideals). It's all gradual, and there are no absolute lines, and we know this because we know evolution. It isn't that there's flaws in evolution, it's that there are flaws in the way we think about biology and evolution lays these errors out for us.

How in layman's terms did the first humans evolve from apes and then proceed to separate themselves from the apes?
It doesn't work like that. We didn't separate ourselves from the apes.  At some point some of our ancestors took different paths than their siblings. One brother went one way, and the other brother went the other way, and they never got back together. And these ancestors of ours were apes, so we are apes. We don't ever stop being apes. Populations split and divide. The group that gave rise to the orangutans went one way and the group that gave rise to the gorillas/humans/chimps/bonobos went another and didn't get back together. Then the group that gave rise to the mountain and lowland gorillas went one way and the group that gave rise to the humans/chimps/bonobos went the other, and didn't get back together. And the group that gave rise to humans went one way and chimps/bonobos went the other. And then the chimps went one way and the bonobos went the other. There isn't some set in stone principle that says this is X and it perfectly exemplifies Xness. There are populations and they divide. Sometimes they get back together and sometimes they don't. When things started, they were no different than one brother is from another brother, in fact, necessarily you could find a set of siblings that gave rise to every divided group. You could find a pair of siblings where one gave rise to all the reptiles, dinosaurs, and birds, and the other to all the mammals.

At what point did breeding with apes stop by humans?
 There's no absolute lines there. Humans are apes. And humans breed with humans. So in a very real sense humans still breed with apes (namely humans). But, looking back to our last common ancestor with chimpanzees (our closest relatives), and it may well be that humans might still be able to breed with them! It's not impossible for lions and tigers to breed or horses and donkeys. They produce offspring, and it might well be possible to have human-chimp hybrids. We just don't try it. So at one point did the ancestors of all humans and the ancestors of all chimpanzees stop breeding? -- They stopped breeding when they stopped. Seemingly about 7.8 million years ago or so.

There are some humans in some distant parts of the rain forest who do not know civilization. The large interbreeding population of humans haven't bred with them since we stopped breeding with them. If for some reason we never started back up, eventually we couldn't and we may well end up with two species of humans, such as the Morlocks and the Eloy (from H.G. Well's the Time Machine). It all depends on when the populations stop and happen to stay stopped.

So how did we separate from the other groups of apes. We went one way, and the other group went another. At what point did we stop breeding with them? We didn't breed with them anymore, when we didn't breed with them anymore.

The actual answers seem sort of anticlimactic, and they are. But, mostly because the problems are with our understanding and not evolutionary theory. We tend to see species as these distinct things which fit into these specific boxes. But, that's not what they are. They are siblings and a few million years of isolation. If you want to draw a line, at some point and say your 30,000th cousin is no longer the same species, you can, but realize that it's an arbitrary line, and our 30,000ths cousins the Neandertals bred with us just fine.
You could go further back, but it's still all just where you draw the line and that has nothing to do with evolution, that has to do with you and your line.

" The other day I saw a post from a creationist stating that the Theory of Evolution has to have some fault in it because there is no interbreeding of species."
No. There's a problem there but it's our idea of species that has the fault, not the theory of evolution.

Tuesday, September 20, 2011

The invisible hand of the market and God.

Apparently every third-order emergent property is God at work. There's apparently a belief gaining some steam that the invisible hand of the market is God at work.We see this same thing in various other places too, in Evolution by natural selection we see that it's God at work, and within the human brain we see that that too is mysteriously God.
a third-order emergent structure is a consequence of shape, time, and heritable instructions. For example, an organism's genetic code sets boundary conditions on the interaction of biological systems in space and time.
Well, sometimes things like markets, intelligence, science, and evolution which all rely on evolutionary algorithms are easy to confuse. Early on we develop a theory of minds, and are often keen to assign minds where they don't belong and end up painting with a broad brush to catch all human minds and any other third-order emergent structures, so we see the products of nature as the products of minds. And the invisible hand of the market as God's hand. And the amazingness of our own minds. We see minds in every third order emergent structure, because realizing something is/has too much design to be the result of first or second order emergent structures typically in our lives means it's the result of a mind. Though, it could similarly be the result of anything implementing the evolutionary algorithm, which basically is just the non-random selection of randomly occurring differences.

Friday, September 16, 2011

I don't know which is sadder.

http://thisiscolossal.com/2011/07/world-record-connect-the-dots-mona-lisa-in-6239-dots/

I'm not sure which is sadder, the fact that in instantly recognized that collection of dots as the Mona Lisa or that I instantly knew what it was and that I could do significantly better.

Wednesday, September 14, 2011

The case against Corbett was finally thrown out.

Apparently as a teacher, you can say creationism is superstitious nonsense, not because it's absolutely true but because teachers have qualified immunity and should get a bit of leeway rather than the hard boot.

http://www.csmonitor.com/USA/Justice/2011/0819/US-judges-rule-for-teacher-who-called-creationism-superstitious-nonsense


Also the court finally reversed their decision on some math teacher who was hanging up religious posters. When the initially decision came down I thought it was amazingly stupid. It's not about the teacher's freedom of speech but rather about the teacher in his role as an agent of the government. He doesn't get first amendment protected because the first amendment is specific to restrict what he can do.

http://www.csmonitor.com/USA/Justice/2011/0913/God-Bless-America-Not-in-high-school-math-class-US-judges-rule

How to MacGyver a knife sharpener.



I tried it. It actually works.

Monday, September 12, 2011

The typical Weighted Euclidean color distance values are wrong.

There's a rather common set of weights for Euclidean Color Distance, to better approximate the human eye.

(30*(r1-r2))**2 + (59*(g1-g2))**2 + (11*(b1-b2))**2

30, 59, and 11 for RGB weights.

Well, they seemed rather pulled out of one's ass. And I'm not the first to notice. But, I did bother to code up a routine to compare CIELab to color distance weights and find out, by brute forcing each and every color distance to find out where the weights there should be.

22.216091748149788, 42.88860259783791, 34.895305654012304

Or 22, 43, 35 since we're very likely going with integer math. This is just brute force average best approximation of weighted Euclidean to CIELab.

From the cited Color metric article:

  • Several individuals suggested a weighted Euclidean distance in R'G'B', according to the formula:
    ΔC=3×ΔR2+4×ΔG2+2×ΔB2
    This function has practically the same result as YUV. Its simplicity and speed of calculation make it a better choice than YUV.
  • As explained in the section "gamma correction" below, the perception of brightness by the human eye is non-linear. From the experiments it appears that the curve for this non-linearity is not the same for each colour. The weighted Euclidean distance presented works quite well for the subset of colours where the "red" signal is 128 or more (on a scale of 0-255). For the other half of the full R'G'B' cube, this different weighting produced better results:
    ΔC=2×ΔR2+4×ΔG2+3×ΔB2

 2, 4, 3, are pretty close to my values (they are weights and thus need to be proportional to each other not similar in absolute magnitude). 2/9, 4/9, 3/9 or 22.22, 44.44, 33.33 when you bring the weights up to how many parts I have. So 22.22 vs 22.21...  44.4 vs 42.8 ... 33.3 vs 34.9.

Update:
I also ran the colors for CIELuv.

R G B weights:
0.2753667361197665, 0.3846815980739518, 0.3399516658062818

28, 38, 34

Saturday, September 10, 2011

Evangelist says things in response to atheist.

So part of Google+'s features is something called sparks. It's actually pretty cool for finding relevant bits of fun things. Like this interesting article which vaguely consists of an Evangelist saying things in response to an unlinked Atheist in some state somewhere.

There is a lot of really pathethic bits, once again highlighting the difference between answering or countering something and simply responding.

“The greatest leap of faith is assuming that order evolves out of complete chaos,” -- This statement particularly stuck in my maw. That's not a leap of faith, that's actually demonstratively true. From Benard Cells, to hexagon's on Saturn, snowflakes, round planets, orderly solar systems. The fact is you can mix a crap ton of whatever in a blender, and leave it for a few days on a table and it will get very orderly. It turns out that chaos is actually rather unstable and things tend to get orderly. While it is true that the second law of thermodynamics insures that higher energy and greater entropic points will necessarily equalize out, that's actually making things more uniform over time. There was a lot of chaos in the early solar system with rocks and debris flying every which direction, but over time, order evolved and the Earth is the only big rock in this very close to perfect circle we orbit in. This is especially true for higher degree emergent phenomenon like evolution, which as a natural consequence of heredity and natural selection creates organisms which bias towards the preservation of better genes and better adapted bodies coded by those genes.

And as for the balanced ecosystems and perfect design of nature, he's just wrong. Everything in nature is poised on a knife's edge, being eaten by parasites and killed by predators, and often starving to death. Most animals suffer through most of their lives with the smallest amount of bad luck separating one from suffering through another week or dying what must be a very painful death. If a loving God designed such a system, that God's only love is for suffering and pain, because nature is red in tooth and claw. And the life of animals is solitary, poor, nasty, brutish, and short. The only thing such a system manages is to make the smallest genetic advantage thrive within that population's gene pool and thereby selectively bias the future generations towards having rather than lacking that advantage.
And then there's his other really terrible bit of claiming:


“Nothing in life is free. The more valuable the object, the higher the price. Nothing is more valuable to God than the human race, which is why such a high price had to be paid, and in this case with a divine human life.” --- Paid to whom?

Really, to whom?

While it's certainly the case that propitiation by blood is a common theme throughout religion and religious views, it's not something that actually would be fundamentally important in our 13.75 billion year old universe. The idea that that the oxygen carry molecules of some species, on this tiny little planet, around this tiny star, in this large galaxy, which is simply one of billions, is obviously absurd to anybody who really questions the core bit of Christian dogma, that Blood Sacrifice of God to God makes any real sense.

And the thing is, it does make sense. This is exactly what we should expect to find if a culture obsessed with primitive blood magic were to make a religion. Because if we give blood for blood for criminal offenses, it's rather quick to give blood for sin in religious ones. And then the better the blood, the more it will atone for, so how great and powerful must be the blood of God? This argument is found wholecloth here by Jason Frenn as well as in Hebrews 9. Christianity is exactly what we should expect to find coming out of a culture rife in belief in blood magic, but not what we should expect to find being true. And lo and behold, we find Christianity first coming out of exactly a place and region and people obsessed with exactly that. So it really does make sense, as long as Christianity is false.

Monday, September 5, 2011

The evolution of lines. That's clever.




I'm a huge fan of evolutionary algorithms and some general crowd sourcing stuff.

There's stuff we haven't seen! Save the world.




So if there's a bunch of critters that we know everything about is it okay to kill them all?