Three words, my ignorant-of-SE-Michigan-demographics friend:
Dearborn Arab-Americans.
(I.E. there are a *lot* of people in Michigan who have relatives in the Middle East. A *lot*. And most of their relatives tend to be the reasonably well-off types, or they'd still be back in the Middle East, unable to afford to come overseas.)
AD likes to charge *bank* for their parts, though. I love a lot of their chips, they make some great stuff (the SSM series has some fun toys, especially), but they do charge for them.
They don't need millimeter accuracy, for one thing, so what the hell are you talking about?
Part 2: every once in a while, a cop goes and checks that the device and the criminal are in the same location. If they aren't, the criminal gets his parole revoked. The payoff for the (technically difficult) hack is nowhere near the downside, that of being thrown back in jail and probably having some time tacked on for violating parole/removing a tracking device/whatever they want to do you for.
Seriously: do you have any idea how hard it would be to pull off that hack? GPS simulators go for tens of thousands of dollars; you're not going to pull it off with a PC and some solder.
These people are on parole. An occasional random check (which is made pretty easy, since you know exactly where the GPS device is) to make sure the GPS device and the parolee are in close proximity, and if you find out they aren't, they've violated parole and get to go back to jail and serve the rest of their sentence in the ever so entertaining confines of the penal system.
Re:So what's the big deal for the rest of us?
on
SHA-1 Broken
·
· Score: 3, Informative
Here's the worry.
Let's say someone trusts a digital signature, signed with SHA-1, to the point of allowing money to be predicated on the validity of this signature. If the message is signed and valid, the payer pays the payee $X dollars, where X is some very large amount.
Message #1 is generated and sent. It validates.
The money is paid. At which point the payee produces a second message which hashes the same as the first but claims to be turning down the deal, or modifying the terms of the deal s.t. they don't have to do anything to earn that money, and they claim that's what was actually sent.
This is a problem, since the break apparently allows the construction of two (relatively) arbitrary message sequences that hash to the same value, which is an easier and much different problem from constructing an arbitrary message that hashes the same as a pre-existing message.
Information theory criticisms are common amongst creationists; thus, I assumed you were working from one. I have no clue what you are trying to refer to with "meaningful sidebands", but the principle works just the same for however many bits of information you want to extend it to; its just math, it doesn't care how many bits you want to use. I wouldn't even have brought it in, if it weren't so typical that it would be brought in by the creationist side of this argument. It isn't really the right framework to address the problem, but it does hold gross truth to apply to the problem at hand. If you want to drop the information theoretic aspect of it, I'd be happy to.
Your analogy is wrong. Why? It's very simple.
99.9% of the time, replacing that heat tile with something random would kill the shuttle. 0.1% of the time (probably less, but the exact probability is relatively unimportant) we get something that works - is useful, to use your terminology. 0.1% of those work *better* than the original. Are we agreed that these are all possible? Because that is exactly why randomness (increased entropy) is essential.
The second stage is natural selection; in this stage, we look at the designs and reproduce the most successful ones. Thus, over time our shuttle would become a veritable brick spacehouse, nearly impervious to heat. Lots of trials? Lots of time? Absolutely. But the mechanism works.
You're struggling to explain it not because its unclear, but because you are wrong.
I've played Netrek on and off for years (mostly off, for the past few). It's fun. Xpilot was too. I've been video gaming since 1986 or so. Computer gaming since I first got a computer, an Apple II. Don't talk history to me without knowing something yourself.
Gee, isn't Nethack itself a copy of earlier commercial games, like (e.g.) Rogue? Isn't that why the entire class of games is called roguelikes? And was Rogue commercial? (Answer: yes. It was sold as part of BSD UNIX, distributed freely with the system but binary only, then commercialized for profit by the authors, and is not considered open source.) And didn't Rogue have randomized maps? Yes. It did. You have no sense of history, and are completely full of shit.
Claiming that Diablo was just "Nethack with 3D graphics" ignores all the work of the artists, musicians... and the game designers who put quite a bit of time into making the game enjoyable. Not just good looking, but enjoyable.
What you're seeing with proprietary games today are cookie cutters with improved graphics.
Yes, because Katamari Damacy was a cookie-cutter. So was DDR (which I personally hate, but lots of people seem to like). And open-source definitely originated the RTS game. Civilization was totally an open-source effort. MULE, too. The RPG as we know it owes a lot more to Final Fantasy, Ultima, and Wizardry than it does to anything open source has put up. Let's not forget Infocom, for that matter. FPSs were started by open source.... oh, wait, no. Name to me one significant genre of game that open source developed. Go ahead. Occasional innovations that then spread to other games in the same genre (ala the addition of build queues to RTSs, textbinds to FPS comms systems, observer ports for net games) does not mean open source gaming did more than commercial gaming; the fact remains, ask most any gamer - even the ones who *do* like open source - what games most advanced the state of the art, and they'll talk about commercial games. They'll talk about the genre-definers, games like Rogue, Civilization, DOOM, Dune2, Warcraft, Ultima, Wizardry. Open source games, alas, are really just reimplementations of these.
Proprietary != one size fits all. Many proprietary games hit a niche and stick there, without reaching the mainstream at all.
Fun without tinkering != one size fits all. It might be fun for only a certain subset of the population (think Bridge Construction Set - is it a good game? Hell yes! But someone who doesn't think math/physics is neat, who isn't interested in static and dynamic structural stability, might not think so.)
Adding randomness/entropy is not adding information. As you said, selection selects from existing information, it doesn't add anything. You still lack a mechanism for adding information. Selection does indeed operate on randomness, where it is made accessible by expression. Selection removes the entropy again. Selection is, in a way, a limiting entropy filter.
Adding randomness *is*, in this case, adding information. Information can be defined as the change in entropy as you transition from a priori knowledge to a posteriori knowledge. If you, a priori, know exactly what your input and output are going to be, your information content is zero. However, this also implies that your input must be uniform (have very low entropy). By increasing the randomness (disorder, entropy) of the input, and hence lowering your a priori knowledge, mutation *does* add information. Let's take the extreme example of a 100% incident homozygote AA. The entropy of this is 0 (log2(1) is 0.) Let's assume a mutation occurs, and that we now have a 50/50 distribution of A and B (i.e. AA, AB, BA, and BB are all equipresent genotypes), just to make the math a little quicker. The entropy has decreased, and the information passed in the message (Hbefore - Hafter) is 1 bit. Thus, the mutation *has* added information. Then natural selection provides a "filter", as you put it, eliminating unsuccessful information.
You're using Dembski's arguments, but he fails elementary information theory miserably.
Yeah, I must be an awful person because I do like FPSs. And it's impossible for a good RPG to be good because it has a story (aka production values).
Actually, the games that have seen the most playtime from me recently are Metroid Prime 2, Katamari Damacy, NCAA 2004, Bridge Construction Set, and EV Nova. There isn't a FPS in the bunch (no, Metroid is *not* an FPS, despite having a first person vantage.) I'd be playing more RPGs if I: had more time, had seen anything worth buying last time I was at the store 2 or 3 months ago.
Tecmo is a bunch of assholes, but most open source games just aren't that good.
What about those of us who just want to play a fun game?
Sorry, but most open source games are just not very good. The ones that are fun, are almost without exception the ones that are just ripoff versions of commercial software.
Have fun with your open source games; I like to play games with production value, which (unfortunately) limits me to commercial software. There are small commercial houses that produce cool stuff (Introversion, ChronicLogic), but even they are closed-source and commercial.
Enjoy FrozenBubble while I go play Metroid Prime. Enjoy TuxRacer while I get down to Galactic Civilizations II. And we won't even start with MMOs.
Actually, this is a real comment I wrote in some DSP code.// I have no idea what this line of code does.// But if I get rid of it, the DSP won't boot. code code code// So leave it the fuck alone.
You realize that *real* companies push their patches out to their machines automatically... after they've tested those patches?
We're not talking about Windows Update. We're talking about "OK, the patch is good... force all 25,000 workstations to upgrade now." That's an auto-update.
Support for 192/24 audio, if it doesn't already have it. HDMI/HDCP switching and upconversion from RCA/SV/Component. Possibly DVI-HDCP conversion, for switching in a computer source. 7.1 surround, or 9.1, or whatever they decide to do next. The next Dolby Labs crap.
You can argue you may not need these, and you don't necessarily need them, but they're in the same class as signal upconversion, component I/O, auto sound adjustment, etc.
(I'm in the same boat - I have a roughly 8 year old Sony receiver that, sooner or later, I am going to want to replace in order to get component switching, upconversion, etc... in the meantime, I make do with an Audio Authority component switching box.)
Your title is my whole point. Natural selection, yes, evolution yes, Evolution... maybe. No strong evidence for Evolution yet. Of course, it is a natural extension to evolution.
As to the destructive change argument... you're wrong. I could easily argue that the change away from hem-S to hem-A is a destructive one; it results in the loss of disease immunity and the loss of genetic information. Natural selection is predicated on information *loss* - less valuable information is lost, replaced by more valuable information, so that argument is just an invalid attack on selection. Your argument is almost intelligent design - you seem to be claiming that there's something more than blind mathematics and chance guiding this.
Further, you have your argument wrong. We want added informational entropy - we want more randomness in order for natural selection and Evolution to work. Without some measure of randomness being added, natural selection quickly stops. This is where information is added; mutation adds, while selection loses, information. The net result is an increase, because the survival gain (breeding) for successful information is large enough that all suitable organisms will eventually obtain that added information.
I've seen Sarfati's information theory critiques of evolution, and he ignores the huge survival gain for a favorable mutation; in effect, you only need one good mutation to wipe out a hundred million bad ones.
Look at it this way. Take a CD, with, say, Britney Spears on it. Burn a couple trillion copies. Scratch each one with a nail, slightly differently. Listen to each one. If we assume that certain types of music are objectively "better" than Britney, and we can write a goal function that can tell whether what we're listening to is "better" (this is where survival of the fittest comes in), then we can evaluate each of those CDs. The one that is most fit is going to spread its genotype, at which point we select it, burn a trillion copies, and repeat the process. Eventually we wind up with Mozart. The trick is that we add information randomly, and remove it selectively. Because of the selectivity, we can assure that changes which add bad information are ignored, while changes that add good information are propagated. Think of it like the Linux kernel. Patches you write are rejected as being unsuitable. Patches Linus writes are accepted as being good. Even if there are 10 million of you to 1 Linus, the kernel improves.
I am arguing for natural selection, and little-E evolution (intraspecies adaptation to environment). I am, as I stated in the beginning, unconvinced regarding any particular argument for the mechanism for speciation and (by extension) Evolution, although it seems reasonable to me that these evolutionary changes would, over time in completely seperated populations, lead to speciation. There is no prevarication in this argument.
In non-malarial areas, and areas where malaria treatment is easy and effective, hem-S has no benefit, and hem-A should be ~100%. And, as you'll note, it is. Northern Europeans don't suffer sickle cell issues. When people from malaria-susceptible areas move to non-malarial areas, or when malaria treatment is improved, the incidence of hem-S begins to drop towards zero (tempered only by the availability of medical treatment for sickle cell disease, which helps explain why the trait hasn't yet disappeared entirely in non-malarial areas like the US, as well as the constant influx of people from malaria-susceptible regions who might be carriers).
Unless you can prevent an alternate theory which better explains why this mapping of hem-S incidence to malaria mortality rate is so good (and it is), I think I'll stick with the best available explanation, which would be natural selection. Any explanation that assumes an unproven creator is flawed from the beginning.
Sorry, I mistyped. Hemoglobin-C is believed to be replacing hemoglobin-S, due to the significantly less severe symptoms for CC individuals as compared to SS ones, while it still provides malaria resistance. (Modiano 2001).
Is the mutation really nett-beneficial? When the mosquitos go away again, the population will be left with a 25% mortality rate from this "benefit" and two thirds of the survivors weakened. It may have reduced the death-rate temporarily (or not, we may have simply seen more Africans born instead), but in the long term, the toll (in deaths and less-productive members) from it is going to be humungous.
You do know that there's no penalty for having a single hemoglobin-S gene, right? So none of the survivors will be weakened. The only exception is that people with two variant hemoglobin genes will often suffer sickle cell, or similar. However, the variant sickle cell genes
Further, I make no assumption that its designed to get rid of mosquitos - its designed to *protect* humans. In fact, you could argue that it would be mutually beneficial to the mosquitos, as it is likely to keep more humans alive, giving the mosquitos increased feeding sources. So everything you say about "When the mosquitos go away" is meaningless - if the mosquitos went away, the trait would disappear fairly quickly. The argument on natural selection as evidenced by the hem-S gene has nothing to do with mosquitos, except insofar as they provide the external selective pressure. The presence or absence of hem-S has no real effect on the mosquitos.
Also, 25% of your population is not expressive, and therefore vulnerable to malaria anyway. From this we learn that the mutation is not necessary to the survival of humans in that area.
Not necessary. Just useful. Natural selection doesn't require that something be required for it to be selected; it just requires that it be better than the alternatives.
Absolutely true, but it has not improved the species, and neither of the two population groups look like dying out or speciating.
I would argue that yielding a higher local survival rate *is* improving the species.
As to your die-out claim, there's another malaria-preventing mutation, hemoglobin-C. Hemoglobin-C's distribution in populations at risk of malaria is slowly dwindling, as it provides less protection from malaria than hemoglobin-S. So, in fact, there is a population group dying out. Also, the gene incidence for hemoglobin-S in Africa can be as high as 46% (areas in Uganda), while their descendents in the US have significantly lower incidences, less than 10%. That shows me two population groups dying out due to their genotype being less suitable than an available alternative.
How do you explain the gene incidence distribution for variant hemoglobins following the distribution of malaria, if not by natural selection? Wherever malaria is a large problem, you find a variant hemoglobin - Latin America, Africa, SE Asia. Where malaria isn't a problem (and not because of treatment, but because of climate), you don't find these variants - Northern Europe, Russia, Canada. Explain that without natural selection.
You could only imagine it being an improvement on a planet totally swamped in mossies, with nowhere to run - and even then it's still not really an improvement, only a destructive second-best coping mechanism which is revealed as a massive burden again if the mossies are ever removed from the equation.
It is an improvement wherever malaria is a large problem. I wouldn't argue it is an improvement to the species as a whole, but it is an improvement to the species in a given environment. Which is all natural selection is about.
Basically, you seem to have a strange idea about what natural selection implies. The variant hemoglobin/malaria issue provides strong evidence for the correctness of the theory of natural selection as a mechanism for intraspecies change.
Well, hell, it is. I mean, can you tell me DJ Assault's "Ass 'n' Titties" is not the finest song ever written? Can you argue that Deeon's "Let Me Bang" puts "Let It Be" to shame?
Can not the whole world agree that Paris the Black-Fu is a god amongst men?
(F'real. Ghetto-tech is the best. OK. Not really. But you gotta love the booty.)
Shit, anyone who listens to "Night Drive" should be able to hear the italo influence. And anyone who listens to "Night Drive" and tells me cheap gear isn't usable for good music is going to get a kicking.
(I don't really think only Detroit puts out good music; I just say shit like that to piss off the under-educated. That said, I think that on balance, a lot/most of the good electronic stuff I wind up hearing is American in origin. Then again, I could give a fuck about trance and its many little subgenres, which wipes out a lot of the Euro scene right there.)
Aw, fuck. Somebody had to call me on KDJ, didn't they.
Well, look. We kind of look at Moodymann as our own homegrown form of trance - it sucks now, we know it sucks now, but we have to tolerate it for its prior contributions. And even so, I'll take Moodymann over most trance producers.
Slashdot Mirror
Three words, my ignorant-of-SE-Michigan-demographics friend:
Dearborn Arab-Americans.
(I.E. there are a *lot* of people in Michigan who have relatives in the Middle East. A *lot*. And most of their relatives tend to be the reasonably well-off types, or they'd still be back in the Middle East, unable to afford to come overseas.)
AD likes to charge *bank* for their parts, though. I love a lot of their chips, they make some great stuff (the SSM series has some fun toys, especially), but they do charge for them.
They don't need millimeter accuracy, for one thing, so what the hell are you talking about?
Part 2: every once in a while, a cop goes and checks that the device and the criminal are in the same location. If they aren't, the criminal gets his parole revoked. The payoff for the (technically difficult) hack is nowhere near the downside, that of being thrown back in jail and probably having some time tacked on for violating parole/removing a tracking device/whatever they want to do you for.
Seriously: do you have any idea how hard it would be to pull off that hack? GPS simulators go for tens of thousands of dollars; you're not going to pull it off with a PC and some solder.
These people are on parole. An occasional random check (which is made pretty easy, since you know exactly where the GPS device is) to make sure the GPS device and the parolee are in close proximity, and if you find out they aren't, they've violated parole and get to go back to jail and serve the rest of their sentence in the ever so entertaining confines of the penal system.
Here's the worry.
Let's say someone trusts a digital signature, signed with SHA-1, to the point of allowing money to be predicated on the validity of this signature. If the message is signed and valid, the payer pays the payee $X dollars, where X is some very large amount.
Message #1 is generated and sent. It validates.
The money is paid. At which point the payee produces a second message which hashes the same as the first but claims to be turning down the deal, or modifying the terms of the deal s.t. they don't have to do anything to earn that money, and they claim that's what was actually sent.
This is a problem, since the break apparently allows the construction of two (relatively) arbitrary message sequences that hash to the same value, which is an easier and much different problem from constructing an arbitrary message that hashes the same as a pre-existing message.
Information theory criticisms are common amongst creationists; thus, I assumed you were working from one. I have no clue what you are trying to refer to with "meaningful sidebands", but the principle works just the same for however many bits of information you want to extend it to; its just math, it doesn't care how many bits you want to use. I wouldn't even have brought it in, if it weren't so typical that it would be brought in by the creationist side of this argument. It isn't really the right framework to address the problem, but it does hold gross truth to apply to the problem at hand. If you want to drop the information theoretic aspect of it, I'd be happy to.
Your analogy is wrong. Why? It's very simple.
99.9% of the time, replacing that heat tile with something random would kill the shuttle. 0.1% of the time (probably less, but the exact probability is relatively unimportant) we get something that works - is useful, to use your terminology. 0.1% of those work *better* than the original. Are we agreed that these are all possible? Because that is exactly why randomness (increased entropy) is essential.
The second stage is natural selection; in this stage, we look at the designs and reproduce the most successful ones. Thus, over time our shuttle would become a veritable brick spacehouse, nearly impervious to heat. Lots of trials? Lots of time? Absolutely. But the mechanism works.
You're struggling to explain it not because its unclear, but because you are wrong.
I've played Netrek on and off for years (mostly off, for the past few). It's fun. Xpilot was too. I've been video gaming since 1986 or so. Computer gaming since I first got a computer, an Apple II. Don't talk history to me without knowing something yourself.
Gee, isn't Nethack itself a copy of earlier commercial games, like (e.g.) Rogue? Isn't that why the entire class of games is called roguelikes? And was Rogue commercial? (Answer: yes. It was sold as part of BSD UNIX, distributed freely with the system but binary only, then commercialized for profit by the authors, and is not considered open source.) And didn't Rogue have randomized maps? Yes. It did. You have no sense of history, and are completely full of shit.
Claiming that Diablo was just "Nethack with 3D graphics" ignores all the work of the artists, musicians... and the game designers who put quite a bit of time into making the game enjoyable. Not just good looking, but enjoyable.
What you're seeing with proprietary games today are cookie cutters with improved graphics.
Yes, because Katamari Damacy was a cookie-cutter. So was DDR (which I personally hate, but lots of people seem to like). And open-source definitely originated the RTS game. Civilization was totally an open-source effort. MULE, too. The RPG as we know it owes a lot more to Final Fantasy, Ultima, and Wizardry than it does to anything open source has put up. Let's not forget Infocom, for that matter. FPSs were started by open source.... oh, wait, no. Name to me one significant genre of game that open source developed. Go ahead. Occasional innovations that then spread to other games in the same genre (ala the addition of build queues to RTSs, textbinds to FPS comms systems, observer ports for net games) does not mean open source gaming did more than commercial gaming; the fact remains, ask most any gamer - even the ones who *do* like open source - what games most advanced the state of the art, and they'll talk about commercial games. They'll talk about the genre-definers, games like Rogue, Civilization, DOOM, Dune2, Warcraft, Ultima, Wizardry. Open source games, alas, are really just reimplementations of these.
Proprietary != one size fits all. Many proprietary games hit a niche and stick there, without reaching the mainstream at all.
Fun without tinkering != one size fits all. It might be fun for only a certain subset of the population (think Bridge Construction Set - is it a good game? Hell yes! But someone who doesn't think math/physics is neat, who isn't interested in static and dynamic structural stability, might not think so.)
Adding randomness/entropy is not adding information. As you said, selection selects from existing information, it doesn't add anything. You still lack a mechanism for adding information. Selection does indeed operate on randomness, where it is made accessible by expression. Selection removes the entropy again. Selection is, in a way, a limiting entropy filter.
Adding randomness *is*, in this case, adding information. Information can be defined as the change in entropy as you transition from a priori knowledge to a posteriori knowledge. If you, a priori, know exactly what your input and output are going to be, your information content is zero. However, this also implies that your input must be uniform (have very low entropy). By increasing the randomness (disorder, entropy) of the input, and hence lowering your a priori knowledge, mutation *does* add information. Let's take the extreme example of a 100% incident homozygote AA. The entropy of this is 0 (log2(1) is 0.) Let's assume a mutation occurs, and that we now have a 50/50 distribution of A and B (i.e. AA, AB, BA, and BB are all equipresent genotypes), just to make the math a little quicker. The entropy has decreased, and the information passed in the message (Hbefore - Hafter) is 1 bit. Thus, the mutation *has* added information. Then natural selection provides a "filter", as you put it, eliminating unsuccessful information.
You're using Dembski's arguments, but he fails elementary information theory miserably.
The problem is, I actually wrote this code.
I didn't copy it from anyone, I wrote it in the first place. Still had no idea what it did, though.
Yeah, I must be an awful person because I do like FPSs. And it's impossible for a good RPG to be good because it has a story (aka production values).
Actually, the games that have seen the most playtime from me recently are Metroid Prime 2, Katamari Damacy, NCAA 2004, Bridge Construction Set, and EV Nova. There isn't a FPS in the bunch (no, Metroid is *not* an FPS, despite having a first person vantage.) I'd be playing more RPGs if I: had more time, had seen anything worth buying last time I was at the store 2 or 3 months ago.
Tecmo is a bunch of assholes, but most open source games just aren't that good.
What about those of us who just want to play a fun game?
Sorry, but most open source games are just not very good. The ones that are fun, are almost without exception the ones that are just ripoff versions of commercial software.
Have fun with your open source games; I like to play games with production value, which (unfortunately) limits me to commercial software. There are small commercial houses that produce cool stuff (Introversion, ChronicLogic), but even they are closed-source and commercial.
Enjoy FrozenBubble while I go play Metroid Prime. Enjoy TuxRacer while I get down to Galactic Civilizations II. And we won't even start with MMOs.
Fucking Slashdot.
// I have no idea what this line of code does.
// But if I get rid of it, the DSP won't boot.
// So leave it the fuck alone.
I, of course, meant to write:
Actually, this is a real comment I wrote in some DSP code.
long meaningless bit of code that still isn't as ugly as Perl;
Actually, this is a real comment I wrote in some DSP code. // I have no idea what this line of code does. // But if I get rid of it, the DSP won't boot. // So leave it the fuck alone.
code
code
code
USC is not attached to the University of California system; it is a private school.
Look, I like early KDJ as much as the next Detroit scenester.
But the stuff he's put out lately is... ech. Just ech. Much like pretty much every trance track I've ever heard, just ech.
You realize that *real* companies push their patches out to their machines automatically... after they've tested those patches?
We're not talking about Windows Update. We're talking about "OK, the patch is good... force all 25,000 workstations to upgrade now." That's an auto-update.
Receiver upgrades:
Support for 192/24 audio, if it doesn't already have it.
HDMI/HDCP switching and upconversion from RCA/SV/Component. Possibly DVI-HDCP conversion, for switching in a computer source.
7.1 surround, or 9.1, or whatever they decide to do next. The next Dolby Labs crap.
You can argue you may not need these, and you don't necessarily need them, but they're in the same class as signal upconversion, component I/O, auto sound adjustment, etc.
(I'm in the same boat - I have a roughly 8 year old Sony receiver that, sooner or later, I am going to want to replace in order to get component switching, upconversion, etc... in the meantime, I make do with an Audio Authority component switching box.)
It's a repeat. Spectrum ran an article on it as well, and it's actually been pretty well covered, just not in the past... mmm... month or so.
Your title is my whole point. Natural selection, yes, evolution yes, Evolution... maybe. No strong evidence for Evolution yet. Of course, it is a natural extension to evolution.
As to the destructive change argument... you're wrong. I could easily argue that the change away from hem-S to hem-A is a destructive one; it results in the loss of disease immunity and the loss of genetic information. Natural selection is predicated on information *loss* - less valuable information is lost, replaced by more valuable information, so that argument is just an invalid attack on selection. Your argument is almost intelligent design - you seem to be claiming that there's something more than blind mathematics and chance guiding this.
Further, you have your argument wrong. We want added informational entropy - we want more randomness in order for natural selection and Evolution to work. Without some measure of randomness being added, natural selection quickly stops. This is where information is added; mutation adds, while selection loses, information. The net result is an increase, because the survival gain (breeding) for successful information is large enough that all suitable organisms will eventually obtain that added information.
I've seen Sarfati's information theory critiques of evolution, and he ignores the huge survival gain for a favorable mutation; in effect, you only need one good mutation to wipe out a hundred million bad ones.
Look at it this way. Take a CD, with, say, Britney Spears on it. Burn a couple trillion copies. Scratch each one with a nail, slightly differently. Listen to each one. If we assume that certain types of music are objectively "better" than Britney, and we can write a goal function that can tell whether what we're listening to is "better" (this is where survival of the fittest comes in), then we can evaluate each of those CDs. The one that is most fit is going to spread its genotype, at which point we select it, burn a trillion copies, and repeat the process. Eventually we wind up with Mozart. The trick is that we add information randomly, and remove it selectively. Because of the selectivity, we can assure that changes which add bad information are ignored, while changes that add good information are propagated. Think of it like the Linux kernel. Patches you write are rejected as being unsuitable. Patches Linus writes are accepted as being good. Even if there are 10 million of you to 1 Linus, the kernel improves.
I am arguing for natural selection, and little-E evolution (intraspecies adaptation to environment). I am, as I stated in the beginning, unconvinced regarding any particular argument for the mechanism for speciation and (by extension) Evolution, although it seems reasonable to me that these evolutionary changes would, over time in completely seperated populations, lead to speciation. There is no prevarication in this argument.
In non-malarial areas, and areas where malaria treatment is easy and effective, hem-S has no benefit, and hem-A should be ~100%. And, as you'll note, it is. Northern Europeans don't suffer sickle cell issues. When people from malaria-susceptible areas move to non-malarial areas, or when malaria treatment is improved, the incidence of hem-S begins to drop towards zero (tempered only by the availability of medical treatment for sickle cell disease, which helps explain why the trait hasn't yet disappeared entirely in non-malarial areas like the US, as well as the constant influx of people from malaria-susceptible regions who might be carriers).
Unless you can prevent an alternate theory which better explains why this mapping of hem-S incidence to malaria mortality rate is so good (and it is), I think I'll stick with the best available explanation, which would be natural selection. Any explanation that assumes an unproven creator is flawed from the beginning.
Sorry, I mistyped. Hemoglobin-C is believed to be replacing hemoglobin-S, due to the significantly less severe symptoms for CC individuals as compared to SS ones, while it still provides malaria resistance. (Modiano 2001).
Is the mutation really nett-beneficial? When the mosquitos go away again, the population will be left with a 25% mortality rate from this "benefit" and two thirds of the survivors weakened. It may have reduced the death-rate temporarily (or not, we may have simply seen more Africans born instead), but in the long term, the toll (in deaths and less-productive members) from it is going to be humungous.
You do know that there's no penalty for having a single hemoglobin-S gene, right? So none of the survivors will be weakened. The only exception is that people with two variant hemoglobin genes will often suffer sickle cell, or similar. However, the variant sickle cell genes
Further, I make no assumption that its designed to get rid of mosquitos - its designed to *protect* humans. In fact, you could argue that it would be mutually beneficial to the mosquitos, as it is likely to keep more humans alive, giving the mosquitos increased feeding sources. So everything you say about "When the mosquitos go away" is meaningless - if the mosquitos went away, the trait would disappear fairly quickly. The argument on natural selection as evidenced by the hem-S gene has nothing to do with mosquitos, except insofar as they provide the external selective pressure. The presence or absence of hem-S has no real effect on the mosquitos.
Also, 25% of your population is not expressive, and therefore vulnerable to malaria anyway. From this we learn that the mutation is not necessary to the survival of humans in that area.
Not necessary. Just useful. Natural selection doesn't require that something be required for it to be selected; it just requires that it be better than the alternatives.
Absolutely true, but it has not improved the species, and neither of the two population groups look like dying out or speciating.
I would argue that yielding a higher local survival rate *is* improving the species.
As to your die-out claim, there's another malaria-preventing mutation, hemoglobin-C. Hemoglobin-C's distribution in populations at risk of malaria is slowly dwindling, as it provides less protection from malaria than hemoglobin-S. So, in fact, there is a population group dying out. Also, the gene incidence for hemoglobin-S in Africa can be as high as 46% (areas in Uganda), while their descendents in the US have significantly lower incidences, less than 10%. That shows me two population groups dying out due to their genotype being less suitable than an available alternative.
How do you explain the gene incidence distribution for variant hemoglobins following the distribution of malaria, if not by natural selection? Wherever malaria is a large problem, you find a variant hemoglobin - Latin America, Africa, SE Asia. Where malaria isn't a problem (and not because of treatment, but because of climate), you don't find these variants - Northern Europe, Russia, Canada. Explain that without natural selection.
You could only imagine it being an improvement on a planet totally swamped in mossies, with nowhere to run - and even then it's still not really an improvement, only a destructive second-best coping mechanism which is revealed as a massive burden again if the mossies are ever removed from the equation.
It is an improvement wherever malaria is a large problem. I wouldn't argue it is an improvement to the species as a whole, but it is an improvement to the species in a given environment. Which is all natural selection is about.
Basically, you seem to have a strange idea about what natural selection implies. The variant hemoglobin/malaria issue provides strong evidence for the correctness of the theory of natural selection as a mechanism for intraspecies change.
Well, hell, it is. I mean, can you tell me DJ Assault's "Ass 'n' Titties" is not the finest song ever written? Can you argue that Deeon's "Let Me Bang" puts "Let It Be" to shame?
Can not the whole world agree that Paris the Black-Fu is a god amongst men?
(F'real. Ghetto-tech is the best. OK. Not really. But you gotta love the booty.)
Shit, anyone who listens to "Night Drive" should be able to hear the italo influence. And anyone who listens to "Night Drive" and tells me cheap gear isn't usable for good music is going to get a kicking.
(I don't really think only Detroit puts out good music; I just say shit like that to piss off the under-educated. That said, I think that on balance, a lot/most of the good electronic stuff I wind up hearing is American in origin. Then again, I could give a fuck about trance and its many little subgenres, which wipes out a lot of the Euro scene right there.)
Aw, fuck. Somebody had to call me on KDJ, didn't they.
Well, look. We kind of look at Moodymann as our own homegrown form of trance - it sucks now, we know it sucks now, but we have to tolerate it for its prior contributions. And even so, I'll take Moodymann over most trance producers.