Ditto to what Trepidity and Seumas said. But I have questions. The article has a quote: "If you’re not paying for it, then you’re the product being sold." So: I just installed OpenBSD on some hardware I had lying around. You see where this is going.
I didn't buy the CD set (yet). It's about an 80% chance that I will now, and 99% if I turn the box into a packet filter as planned.
Questions: (1) In what way am I the product being sold (by OpenBSD)? (2) If I'm not "the product", then how can I tell? I'm dissatisfied with any answer that says it's obvious on its face: "I just know in my gut that Google has the capacity for evil, yet OpenBSD is good." (3) If I am "the product", then by what mechanism am I suddenly relieved of that status when I buy the CD sets?
I'm not trying to argue that OpenBSD is good therefore Google is good, nor that Google is bad therefore OpenBSD is bad. I genuinely can't reconcile my experience with the quote, so I think the quote is an overgeneralization. Both Google and OpenBSD may or may not be selling me as a product, the fact that they're free tells me very little, and I have to research them myself.
You're right that we would have just one species of bacteria. We ought to convince ourselves that we do have more than one species, though, and the authors of the article are hard to convince. They use the idea that the incredible diversity and rapid evolution that you mention aren't enough. This indeed relates to the term "evolutionary species."
The authors define "evolutionary species" as a population that evolves independently from others, meaning that genotypes can't cross the line between two evolutionary species. In the term "evolutionary species," I'd put the emphasis on the word species. Yes, all species are evolutionary, but not all evolutionary entities are species.
Why would anyone doubt that asexual organisms can split into evolutionary species? The article says that sexually reproducing organisms can diversify but also experience cohesion within groups, and "asexuals might not diversify into distinct species, because there is no interbreeding to maintain cohesive units above the level of the individual." Distinct is the key word in that sentence. Now, if asexuals have no source of cohesion (including homologous recombination), then they don't have a good way to form truly independent subpopulations, and they can diversify to the extreme, but yes, a group of (true) asexuals would remain always just one big evolutionary species. These authors and others point out that bacteria do form distinct populations (across which genotypes presumably can't cross), but those results are less convincing because the bacteria in question can all undergo recombination, if not sex proper, and that could induce cohesion. The problem of "one big species" still exists for totally asexual organisms like the rotifers in this study.
In the end, this study shows that the characteristics of some truly asexual critters fit the model of cohesive, independent clusters better than not, and they also fit a model of selection based on more than just geographic isolation, for example. How do they cohere? Probably by ending up in niches, the authors say. It has always made sense that true asexuals could do this, but it sounds like we had little evidence about whether they did in fact, before this paper.
Turning to the Times and Slashdot articles, they are both right to point out that it's interesting that organisms can replicate themselves asexually for so long and still compete with sexual reproducers, but they don't emphasize why that's interesting. The comments are right to point out that both sexual and asexual reproduction can generate diversity, so that's not why asexuals are interesting. They're interesting because they lack the usual source of cohesion and independence. In short, the question was "can organisms diversify and cohere into real species or species-like units without any combining of their genes?" The answer is "it looks like it." As an aside, I have to apply a lot of effort to feel good when all tertiary sources seem to miss subtle points like this all the time.
First, because I haven't seen anyone point it out yet, the actual journal article is Neurology 2006;67:1208-1214. Go to http://dx.doi.org/10.1212/01.wnl.0000238082.13860. 50 and the server will redirect you appropriately. The journal's Web site should let you read the abstract for free. To read the whole article, you have to pay, or find a suitable institution with online or print access to this journal.
Now, some comments. The idea that correlation doesn't imply causation is correct, but this paper used a multivariate analysis to attempt to control for several possible confounding factors. I count twelve that the authors thought about and included in some of their models: age, sex, educational level, diabetes, systolic blood pressure, daily alcohol intake, physical activity, perceived health score, perceived stress score, energy, social isolation, and region of residence. It looks like the paper acknowledges more confounders than anyone's mentioned here on Slashdot so far. Ultimately, though, this paper is a cohort study, so you can still argue that they missed a confounding factor. If you can think of a legitimate one, you stand a good chance getting it published in the journal Neurology.
Next, naming intelligent friends with high BMIs or famous thin people with questionable smarts does not change what this paper says, of course. Let's even pretend to add those people to the data. Now we have 2243 subjects instead of 2223. I doubt that changes the results much, but I admit I can't prove that. Counterexamples do tell us something very important, though. If high BMI really causes worse word-list learning, it is still one of a staggering number of other effects on this measure, and it by no means excludes anyone from higher intelligence.
Lastly, people are right to wonder what cognitive tests like word-list learning really measure. This paper didn't use IQ directly, but the point still stands. The authors know this and address it, too. "The functional significance of cognitive changes in our sample is difficult to assess.... We did not collect any direct index of work performance." In fact, they don't know whether differences in these psychometric test scores apply to "this healthy working population." BMI, too, may represent a surrogate marker. The association in this paper still stands, although I don't see anything about whether active weight loss attempts change cognitive point measures or decline. Yes, there are other markers of cardiovascular risk, and these include waist circumference (Am J Cardiol 2006;98:1053-1056), which someone could study in the same way that the Neurology paper studies BMI.
So what's the point? The point is, the differences in these cognitive tests concern some people. The results suggest that some real effect on cognition exists, and the authors mention a few reasonable mechanisms for the effect. If you agree that a normal BMI leads successively to less diabetes, less coronary artery disease, and less chest pain when you walk around, then it makes sense to try for a normal BMI if it's even possible that it will save blood vessels in your brain, or your brain cells directly, or whatever mechanism you believe. It wouldn't surprise me, though, if weight loss merely slows cognitive decline or lessens the risk, rather than positively improving intelligence or some similar claim. The other point is that newspapers check sources and strive to do it very well, but they rarely offer substantial analysis of original research. They will quote authorities regarding the research but leave item-by-item discussion to commentary articles in specialty journals. Even my couple hundred words here only begin to address the reasonable analysis of this or any scholarly article.
From experience, I can tell you that anything less than a teravoxel is utter crap. If I need to sculpt an enlargement of my hand measuring seven feet tall, I don't want to see pixellation close up. Of course, if you just want to make desktop-size HoloSnaps to send to Grandma, disregard everything I've said. P.S. Does anyone have a fast connection I can use for the seven-foot holovideo of my hand? And a really good way to do z-buffering?
Even if judges and/or law clerks don't follow the "collected resources" link he left on that page, how much do you want to bet that a few of them spot the title of the previous node and click that just out of curiosity?
We expect to release version 1.0 of GNU/PNG at about the same time as the GNU/HURD is complete.
Surely you'll give this project a better name: something worthy of the GNU project. I think you should choose GNP. It starts with 'GN', which everyone likes, and it plays on the original abbreviation in the tradition of GPG instead of PGP. And GNP must stand for GNP's not PNG, of course.
Basically, both players could move at the same time.
Slashdot has probably covered this already, but a site called Kung Fu Chess hosts a version of chess "where you never wait your turn", along with a few other games based on the no-waiting concept. I'm sure it does brain damage to your regular chess game if you play it too much, but I find it quite enjoyable.
Slashdot Mirror
Ditto to what Trepidity and Seumas said. But I have questions. The article has a quote: "If you’re not paying for it, then you’re the product being sold." So: I just installed OpenBSD on some hardware I had lying around. You see where this is going.
I didn't buy the CD set (yet). It's about an 80% chance that I will now, and 99% if I turn the box into a packet filter as planned.
Questions:
(1) In what way am I the product being sold (by OpenBSD)?
(2) If I'm not "the product", then how can I tell? I'm dissatisfied with any answer that says it's obvious on its face: "I just know in my gut that Google has the capacity for evil, yet OpenBSD is good."
(3) If I am "the product", then by what mechanism am I suddenly relieved of that status when I buy the CD sets?
I'm not trying to argue that OpenBSD is good therefore Google is good, nor that Google is bad therefore OpenBSD is bad. I genuinely can't reconcile my experience with the quote, so I think the quote is an overgeneralization. Both Google and OpenBSD may or may not be selling me as a product, the fact that they're free tells me very little, and I have to research them myself.
You're right that we would have just one species of bacteria. We ought to convince ourselves that we do have more than one species, though, and the authors of the article are hard to convince. They use the idea that the incredible diversity and rapid evolution that you mention aren't enough. This indeed relates to the term "evolutionary species."
The authors define "evolutionary species" as a population that evolves independently from others, meaning that genotypes can't cross the line between two evolutionary species. In the term "evolutionary species," I'd put the emphasis on the word species. Yes, all species are evolutionary, but not all evolutionary entities are species.
Why would anyone doubt that asexual organisms can split into evolutionary species? The article says that sexually reproducing organisms can diversify but also experience cohesion within groups, and "asexuals might not diversify into distinct species, because there is no interbreeding to maintain cohesive units above the level of the individual." Distinct is the key word in that sentence. Now, if asexuals have no source of cohesion (including homologous recombination), then they don't have a good way to form truly independent subpopulations, and they can diversify to the extreme, but yes, a group of (true) asexuals would remain always just one big evolutionary species. These authors and others point out that bacteria do form distinct populations (across which genotypes presumably can't cross), but those results are less convincing because the bacteria in question can all undergo recombination, if not sex proper, and that could induce cohesion. The problem of "one big species" still exists for totally asexual organisms like the rotifers in this study.
Figure 2 in the article http://dx.doi.org/10.1371/journal.pbio.0050087 shows all of these ideas about cohesion, species, populations, and so forth relatively elegantly.
In the end, this study shows that the characteristics of some truly asexual critters fit the model of cohesive, independent clusters better than not, and they also fit a model of selection based on more than just geographic isolation, for example. How do they cohere? Probably by ending up in niches, the authors say. It has always made sense that true asexuals could do this, but it sounds like we had little evidence about whether they did in fact, before this paper.
Turning to the Times and Slashdot articles, they are both right to point out that it's interesting that organisms can replicate themselves asexually for so long and still compete with sexual reproducers, but they don't emphasize why that's interesting. The comments are right to point out that both sexual and asexual reproduction can generate diversity, so that's not why asexuals are interesting. They're interesting because they lack the usual source of cohesion and independence. In short, the question was "can organisms diversify and cohere into real species or species-like units without any combining of their genes?" The answer is "it looks like it." As an aside, I have to apply a lot of effort to feel good when all tertiary sources seem to miss subtle points like this all the time.
First, because I haven't seen anyone point it out yet, the actual journal article is Neurology 2006;67:1208-1214. Go to http://dx.doi.org/10.1212/01.wnl.0000238082.13860. 50 and the server will redirect you appropriately. The journal's Web site should let you read the abstract for free. To read the whole article, you have to pay, or find a suitable institution with online or print access to this journal.
Now, some comments. The idea that correlation doesn't imply causation is correct, but this paper used a multivariate analysis to attempt to control for several possible confounding factors. I count twelve that the authors thought about and included in some of their models: age, sex, educational level, diabetes, systolic blood pressure, daily alcohol intake, physical activity, perceived health score, perceived stress score, energy, social isolation, and region of residence. It looks like the paper acknowledges more confounders than anyone's mentioned here on Slashdot so far. Ultimately, though, this paper is a cohort study, so you can still argue that they missed a confounding factor. If you can think of a legitimate one, you stand a good chance getting it published in the journal Neurology.
Next, naming intelligent friends with high BMIs or famous thin people with questionable smarts does not change what this paper says, of course. Let's even pretend to add those people to the data. Now we have 2243 subjects instead of 2223. I doubt that changes the results much, but I admit I can't prove that. Counterexamples do tell us something very important, though. If high BMI really causes worse word-list learning, it is still one of a staggering number of other effects on this measure, and it by no means excludes anyone from higher intelligence.
Lastly, people are right to wonder what cognitive tests like word-list learning really measure. This paper didn't use IQ directly, but the point still stands. The authors know this and address it, too. "The functional significance of cognitive changes in our sample is difficult to assess.... We did not collect any direct index of work performance." In fact, they don't know whether differences in these psychometric test scores apply to "this healthy working population." BMI, too, may represent a surrogate marker. The association in this paper still stands, although I don't see anything about whether active weight loss attempts change cognitive point measures or decline. Yes, there are other markers of cardiovascular risk, and these include waist circumference (Am J Cardiol 2006;98:1053-1056), which someone could study in the same way that the Neurology paper studies BMI.
So what's the point? The point is, the differences in these cognitive tests concern some people. The results suggest that some real effect on cognition exists, and the authors mention a few reasonable mechanisms for the effect. If you agree that a normal BMI leads successively to less diabetes, less coronary artery disease, and less chest pain when you walk around, then it makes sense to try for a normal BMI if it's even possible that it will save blood vessels in your brain, or your brain cells directly, or whatever mechanism you believe. It wouldn't surprise me, though, if weight loss merely slows cognitive decline or lessens the risk, rather than positively improving intelligence or some similar claim. The other point is that newspapers check sources and strive to do it very well, but they rarely offer substantial analysis of original research. They will quote authorities regarding the research but leave item-by-item discussion to commentary articles in specialty journals. Even my couple hundred words here only begin to address the reasonable analysis of this or any scholarly article.
From experience, I can tell you that anything less than a teravoxel is utter crap. If I need to sculpt an enlargement of my hand measuring seven feet tall, I don't want to see pixellation close up. Of course, if you just want to make desktop-size HoloSnaps to send to Grandma, disregard everything I've said. P.S. Does anyone have a fast connection I can use for the seven-foot holovideo of my hand? And a really good way to do z-buffering?
Even if judges and/or law clerks don't follow the "collected resources" link he left on that page, how much do you want to bet that a few of them spot the title of the previous node and click that just out of curiosity?
--
AJZ
We expect to release version 1.0 of GNU/PNG at about the same time as the GNU/HURD is complete.
Surely you'll give this project a better name: something worthy of the GNU project. I think you should choose GNP. It starts with 'GN', which everyone likes, and it plays on the original abbreviation in the tradition of GPG instead of PGP. And GNP must stand for GNP's not PNG, of course.
--
Andy Z.
Basically, both players could move at the same time.
Slashdot has probably covered this already, but a site called Kung Fu Chess hosts a version of chess "where you never wait your turn", along with a few other games based on the no-waiting concept. I'm sure it does brain damage to your regular chess game if you play it too much, but I find it quite enjoyable.