> SCO is going down, and they know it. This entire fiasco is a sad attempt to grasp some kind of life from a dying company and they have no problem taking the SCO name down with them.
Someone forgot to order the golden parachutes while the company was in the black, and now they're desperate to drum up enough money to buy a few on short notice.
> If it's already in the kernal, it's hardly secret. Where is it?
I thought they said it wasn't in the kernel. There were some hints that it had to do with the SysV init scripts, but of course Linux used those long before IBM started their Linux-for-enterprise push.
\methinks SCO is just making it up as they go...
BTW, does anyone have a summary of what enterprise-friendly code IBM has actually contributed? That should narrow the search for the Yeti down to a single continent at any rate.
> The longer they keep this info under their hats, the less I believe they have any leg to stand on.
That's also why I believe that this is an anti-Linux FUD campaign. If they were really concerned with IP then they have nothing to gain by keeping the code secret. If they announce it now it will get removed now (which is what they want, right?) but they'll still be entitled to any legal remedy they'd be entitled to without announcing it (assuming any at all). There's simply no IP-based reason not to announce it.
But as for FUD-based reasons, well, it's only FUDworthy so long as everything is up in the air and businesses thinking about making the switch have something to worry about. Point to the code and the argument switches to the facts of the claim, or the code gets ripped out, and the FUD-bubble bursts overnight.
The IP motivation says "announce it", and the FUD motivations says "mum's the word".
> I find it kind of ironic, then, that SCO would continue for two months to sell a product they claim to infringes on their IP rights, and profit (as much or as little as they have in that time) from someone else's work. Something's just wrong with that picture.
More ironic yet is the fact that they'd still be shipping it if they hadn't seen the GPL licensing argument on Slashdot!
We joke about Slashdot being a discussion site for The Register, but it looks like their press releases are a "discussion site" for what they read on Slashdot.
> Linguists have long considered the Washoe and Koko "signing" claims to be utter fraud
Got any examples? I know that the Chomskyites aren't too keen on it, but most of the linguists and other cognitive scientists that I know don't actually have much problem with it.
> the experimenters are almost criminally liberal in their interpretation of the signs.
It goes without saying that the proponents of the idea will give the animals too much credit - just as the opponents of the idea will give them too little. The truth surely lies somewhere in between, but even "somewhere in between" is astonishing, and would have been shocking to researchers a mere half a century ago.
What has been done with the signing or button-pushing apes and what has been seen by anthropologists observing them in the wild for the past ~40 years has forever changed the way we can think about these creatures - and about ourselves as well.
The point is that to say we are one-third daffodils because our DNA matches that of a daffodil 33% of the time, is not profound, it's ridiculous. There is hardly any biological comparison you can make which will find us to be one-third daffodil, except perhaps the DNA.
> I think thats an excellent point.
Not very insightful by my lights, because it completely ignores the important little stuff like DNA, cell structure, biochemistry, etc. We do have a lot in common with daffodils so long as you don't become to bemused by the big "size and shape" stuff.
He would do well to compare what we have in common with daffodils with what we have in common with a lump of copper ore; 33% may then look like an understatement of the similarity.
> I'm glad someone pointed this out. You may remember the study published last September from a Caltech researcher that concluded that the match between humans and chimps was LOWER than previously thought -- not higher. All depends on which genes you want to consider in the counting....and how you make the measurements. IIRC his method would count a displaced substring of the DNA as a whole lot of misses even though it might have resulted from a single mutation.
There are lots of ways of measuring this, and it really doesn't matter too much so long as you only compare numbers that were measured the same way. Using different methods of measurement don't make us any closer to or further from the chimps any more than switching between the metric and English systems changes the distance between two cities.
> In fact the point that the mosquitos mutated that rapidly is fairly bad news for evolutionists, as it begs the question of why there aren't transitional forms all over the planet right now.
Ah, but the planet is full of transitional forms.
> And if you say the platypus is transitional, from what to what???
From its ancestors to its descendents or a dead end - just like the rest of us.
Look at all the "aquatic mammals", ranging from housecats that get annoyed when they step in a puddle to whales that get annoyed because they have to come up for air now and then, and the whole range of acquatic adaptations in between.
The world is brim full of transitional forms, for those whose beliefs don't make them stumble through life with their eyes closed to it.
> > there isn't one point where you suddenly get a new species.
> Sure there is. One scientist says, "Hey, I think this constitutes a new species." A bunch of others say, "Yeah, I think you're right." Viola - new species.
> Well, until a multiverse theory has actualy observational data pointing to it, perhaps it should stay restrained to sci-fi and comic books.
> I'm not aware of any widely accepted theory that says we can make observations to prove or disprove any multiverse theory, so it hardly seems logical to classify them as scientific. SciAm should know better, or at least admit that the article is philosophical speculation, and not scientific.
It's all a part of the scientific method. Before you can test hypotheses you have to generate them, and that's exactly what's going on here. We know a lot about the universe, and our mathematical models for what we know have multiple interpretations. Rees and others are working out some of those interpretations as models for a multiverse, and it may be that some of those multiverse models will make testable predictions.
Think of this as the high-risk phase of the loop in the scientific method if you wish. When physicists spend gazillions of dollars on particle accelerators go get high enough energy for that coveted observation, there's no a priori guarantee that the experiment will actually turn up the expected result.
Same thing here: science of necessity relies on speculation. We just have to guide our speculation as best we can with the evidence at hand, and then see how things turn out when we get there.
> Except that counting the 'junk' DNA, you can easily find instances of people with greater genetic differences than chimps.
Possibly true, just as you can find similar examples when comparing the heights of human males and females.
All that means is that the extremes of two distributions overlap each other. What's important is the distributions themselves, not the extremes you can pull out of them. Even if what you say is true we would expect the variance of the distribution of genetic distances within humans or within chimps is less than the variance of human and chimp genetic distances when the two populations are pooled (just as we see with the variances of male and female heights vs. the variance of human height), and we would also expect the phenomenon to become more pronounced as we compared more and more distant species.
> The 'junk' DNA is really only useful for determining the breed/race of an animal. There are fewer differences in the junk DNA between two europeans than a european an an asian.
If the divergence has been fast enough to show up in that comparison then it should be even stronger between two different species.
Keep this in mind next time you hear someone say that the theory of evolution doesn't make any predictions...
> Except for species. We do have a hard definition of a species, and thats any group of life forms that can reproduce together (at least with sexually reproducing forms, not sure about asexually reproducing ones).
> It should be obvious to any cretin that there is a definite qualitative difference between human and chimp, indeed between human and all of (observable) nature. And that supposedly insignificant quality makes all the difference. The fact that we cannot (yet) measure its true magnitude in scientific terms does not make it any less ridiculously obvious. No human is just another monkey. Not even you.
So, is the chimpanzee intellect more similar to the human intellect or to the sea slug intellect? Or to the intellect of a rock, since you want to make us special with respect to "all of nature"?
Were merely an extremum along one dimension of measure that we're inordinately proud of. Chimps probably scoff because we don't have thumbs on our feet, and rocks because we're so fragile. Who's the fair arbitrater of excellence?
> (Question is, how could our definition of a genus be this open to debate?)
Because it's essentially arbitrary. We have a huge treee of life, and in order to get a handle on it we find it convenient to label some sub-trees as "species", "genuses", "families", etc. Unfortunately there isn't any firm rule for excactly where in the tree each division should be made. In fact the basic labelling dates back to a time when we were first putting the tree together and knew a lot less about it than we do now.
IMO it's very unlikely that anyone will move the "Homo" label higher in the tree to make it more inclusive as a result of this. DNA is a great way to do a sanity check on the structure of the tree, but it doesn't really have any bearing on how we label the nodes.
> Of course now with Hollywood and TV causing humans to devolve, the Chimps will have a chance to catch up.
They've got a long way to go... when a recent association broke up the human went on to be president of the USA and the chimp merely spent the rest of his life, I don't know, probably bonking all the female chimps in California or something similarly unrewarding.
> But I can tell you what's most special. Art. As soon as a chimp can draw a picture of their house, or of their parents or of anything else; then I'd think about moving them onto the same level with us. This is the clearest place where we see that humans differ from animals by type, not just degree. A human doesn't carve a sculpture well and a chimp does it badly. A human does it well and a chimp can't do it at all. This ability to imagine, mentally conceive and create is the genesis of all that makes humans unique.
> If you want to believe that you are almost a chimpanzee, that's fine, but I'm not believing. Note that the researchers ignored DNA that is not expressive. That may be a sensible idea, or it may be that the ignored DNA expresses itself in a way that has not been discovered.
Even when looking at the "junk" DNA that is believed to be non-coding, we are still found to be more similar to the chimps than to any other species - just not at the 99.4% similarity rate.
This is a natural fallout from the way evolution works: the non-coding DNA experiences less selective pressure and is thus likely to accumulate more random mutations over time than the highly selected stuff is. However, since humans and chimps started with the same pool of "junk" DNA and only diverged a few million years ago, fewer random differences have accumulated between us and them than between, say, us and chickens. (Arguably the "junk" DNA is a better way of calculating similarities and time since divergence, since any drift there is thought to be the result of accumulated random mutations rather than the result of any kind of environmental pressure, which obviously works differently in different environments.)
> If 99% of the important DNA is identical, then probability implies that 99% of the rest of the DNA is also identical.
No, the non-coding "junk" DNA will experience less selective pressure and thus drift more or less than random, so you would expect a higher difference rate there, more or less proportional to the time since divergence from the common ancestor.
The reason these researchers got the new 99.4% rather than the lower numbers previously obtained by other methods is because there rest of the DNA isn't 99.4% identical.
> We've only fully mapped the human genome so far. I bet if we fully mapped the chimp genome, we'd see many many more entries in the diff log than we thought.
...and many more similarities as well. Welcome to the concept of "percent".
However, the real reason for the bogosity of the claim is that clades aren't defined by thresholds in DNA differences. The tree of common descent is there, but it's somewhat arbitrary how far up from a leaf you go before you reach a node that you call "species", "genus", etc. They are merely labels of convenience, and if we suddenly do or don't find it convenient to put the chimps in the genus Homo it doesn't really tell us anything we didn't already know about the relationships in that branch of the tree. (I have a physical anthropology textbook published ten years ago that already mapped out this branch of the tree according to our current understanding of it, then already based on DNA comparisons as well.)
The real news from this is that by focusing on "functionally important" genes we now know that our "functionally important similarity" is 99.4% rather than the 97.whatever% that we previously got when looking at genes in general.
> The GPU allows you to do massively parallel computations, but penalizes you heavilly for things such as loops of variable length or reading memory back from the card outside of the once-per-cycle frame update, and the price of interrupting computation is prohibitive. Clearing the graphics pipeline can take a long, long time.
> Furthermore, while there have been a few papers published claiming the orders of magnitude increase in speed in these sorts of computations, none actually demonstrate this sort of speed-up. Everyone's speculating, but when it comes to it, results are lacking.
I looked in to using the GPU for vector * matrix multiplications over my Christmas vacation (yep, a Geek), and everywhere I turned I found people saying that whatever you gained in the number crunching you lost in the latency of sending your numbers to the GPU and reading them back when done. In the end I didn't even bother running an experiment on it.
But maybe conventional wisdom was wrong; elsewhere in the talkbacks I see links to a couple of.edu sites pushing this kind of thing, so I'm going to look at it some more.
Slashdot Mirror
You bled on my knife, you bastard! If you live I'm going to sue your ass off!!!
> SCO is going down, and they know it.
This entire fiasco is a sad attempt to grasp some
kind of life from a dying company and they
have no problem taking the SCO name down with them.
Someone forgot to order the golden parachutes while the company was in the black, and now they're desperate to drum up enough money to buy a few on short notice.
> If it's already in the kernal, it's hardly secret. Where is it?
I thought they said it wasn't in the kernel. There were some hints that it had to do with the SysV init scripts, but of course Linux used those long before IBM started their Linux-for-enterprise push.
\methinks SCO is just making it up as they go...
BTW, does anyone have a summary of what enterprise-friendly code IBM has actually contributed? That should narrow the search for the Yeti down to a single continent at any rate.
> The longer they keep this info under their hats, the less I believe they have any leg to stand on.
That's also why I believe that this is an anti-Linux FUD campaign. If they were really concerned with IP then they have nothing to gain by keeping the code secret. If they announce it now it will get removed now (which is what they want, right?) but they'll still be entitled to any legal remedy they'd be entitled to without announcing it (assuming any at all). There's simply no IP-based reason not to announce it.
But as for FUD-based reasons, well, it's only FUDworthy so long as everything is up in the air and businesses thinking about making the switch have something to worry about. Point to the code and the argument switches to the facts of the claim, or the code gets ripped out, and the FUD-bubble bursts overnight.
The IP motivation says "announce it", and the FUD motivations says "mum's the word".
No, this isn't about IP.
> I find it kind of ironic, then, that SCO would continue for two months to sell a product they claim to infringes on their IP rights, and profit (as much or as little as they have in that time) from someone else's work. Something's just wrong with that picture.
More ironic yet is the fact that they'd still be shipping it if they hadn't seen the GPL licensing argument on Slashdot!
We joke about Slashdot being a discussion site for The Register, but it looks like their press releases are a "discussion site" for what they read on Slashdot.
> The magazine has gone down hill in credibility in the last few years, and this article is the crowning achievement so far
Surely no worse than Carolyn Meinel's article on computer security back in 1998?
> Scientific American is the new OMNI..
FWIW, I still find several articles worth reading in each issue, plus lots of news material that I would have trouble picking up otherwise.
> Linguists have long considered the Washoe and Koko "signing" claims to be utter fraud
Got any examples? I know that the Chomskyites aren't too keen on it, but most of the linguists and other cognitive scientists that I know don't actually have much problem with it.
> the experimenters are almost criminally liberal in their interpretation of the signs.
It goes without saying that the proponents of the idea will give the animals too much credit - just as the opponents of the idea will give them too little. The truth surely lies somewhere in between, but even "somewhere in between" is astonishing, and would have been shocking to researchers a mere half a century ago.
What has been done with the signing or button-pushing apes and what has been seen by anthropologists observing them in the wild for the past ~40 years has forever changed the way we can think about these creatures - and about ourselves as well.
> I think thats an excellent point.
Not very insightful by my lights, because it completely ignores the important little stuff like DNA, cell structure, biochemistry, etc. We do have a lot in common with daffodils so long as you don't become to bemused by the big "size and shape" stuff.
He would do well to compare what we have in common with daffodils with what we have in common with a lump of copper ore; 33% may then look like an understatement of the similarity.
> I'm glad someone pointed this out. You may remember the study published last September from a Caltech researcher that concluded that the match between humans and chimps was LOWER than previously thought -- not higher. All depends on which genes you want to consider in the counting.
There are lots of ways of measuring this, and it really doesn't matter too much so long as you only compare numbers that were measured the same way. Using different methods of measurement don't make us any closer to or further from the chimps any more than switching between the metric and English systems changes the distance between two cities.
> What about lemurs? they are closer than 'rabbits'
As would be expected, since we and the lemurs are both primates.
> In fact the point that the mosquitos mutated that rapidly is fairly bad news for evolutionists, as it begs the question of why there aren't transitional forms all over the planet right now.
Ah, but the planet is full of transitional forms.
> And if you say the platypus is transitional, from what to what???
From its ancestors to its descendents or a dead end - just like the rest of us.
Look at all the "aquatic mammals", ranging from housecats that get annoyed when they step in a puddle to whales that get annoyed because they have to come up for air now and then, and the whole range of acquatic adaptations in between.
The world is brim full of transitional forms, for those whose beliefs don't make them stumble through life with their eyes closed to it.
> > there isn't one point where you suddenly get a new species.
> Sure there is. One scientist says, "Hey, I think this constitutes a new species." A bunch of others say, "Yeah, I think you're right." Viola - new species.
Ah, so that's how God did it!
> I was absolutely astounded when FireFly got cancelled and the absolute mind-raping garbage that is John Doe stayed on the air.
But at least it gives you some insight into why the RIAA is publishing what it does these days.
For some, Sturgeon's Law isn't an observation, it's a marketing plan. Crap draws bigger audiences than stuff that challenges people.
> Well, until a multiverse theory has actualy observational data pointing to it, perhaps it should stay restrained to sci-fi and comic books.
> I'm not aware of any widely accepted theory that says we can make observations to prove or disprove any multiverse theory, so it hardly seems logical to classify them as scientific. SciAm should know better, or at least admit that the article is philosophical speculation, and not scientific.
It's all a part of the scientific method. Before you can test hypotheses you have to generate them, and that's exactly what's going on here. We know a lot about the universe, and our mathematical models for what we know have multiple interpretations. Rees and others are working out some of those interpretations as models for a multiverse, and it may be that some of those multiverse models will make testable predictions.
Think of this as the high-risk phase of the loop in the scientific method if you wish. When physicists spend gazillions of dollars on particle accelerators go get high enough energy for that coveted observation, there's no a priori guarantee that the experiment will actually turn up the expected result.
Same thing here: science of necessity relies on speculation. We just have to guide our speculation as best we can with the evidence at hand, and then see how things turn out when we get there.
> Except that counting the 'junk' DNA, you can easily find instances of people with greater genetic differences than chimps.
Possibly true, just as you can find similar examples when comparing the heights of human males and females.
All that means is that the extremes of two distributions overlap each other. What's important is the distributions themselves, not the extremes you can pull out of them. Even if what you say is true we would expect the variance of the distribution of genetic distances within humans or within chimps is less than the variance of human and chimp genetic distances when the two populations are pooled (just as we see with the variances of male and female heights vs. the variance of human height), and we would also expect the phenomenon to become more pronounced as we compared more and more distant species.
> The 'junk' DNA is really only useful for determining the breed/race of an animal. There are fewer differences in the junk DNA between two europeans than a european an an asian.
If the divergence has been fast enough to show up in that comparison then it should be even stronger between two different species.
Keep this in mind next time you hear someone say that the theory of evolution doesn't make any predictions...
> Except for species. We do have a hard definition of a species, and thats any group of life forms that can reproduce together (at least with sexually reproducing forms, not sure about asexually reproducing ones).
Actually, even that is somewhat problematic in lots of cases.
> It should be obvious to any cretin that there is a definite qualitative difference between human and chimp, indeed between human and all of (observable) nature. And that supposedly insignificant quality makes all the difference. The fact that we cannot (yet) measure its true magnitude in scientific terms does not make it any less ridiculously obvious. No human is just another monkey. Not even you.
So, is the chimpanzee intellect more similar to the human intellect or to the sea slug intellect? Or to the intellect of a rock, since you want to make us special with respect to "all of nature"?
Were merely an extremum along one dimension of measure that we're inordinately proud of. Chimps probably scoff because we don't have thumbs on our feet, and rocks because we're so fragile. Who's the fair arbitrater of excellence?
> (Question is, how could our definition of a genus be this open to debate?)
Because it's essentially arbitrary. We have a huge treee of life, and in order to get a handle on it we find it convenient to label some sub-trees as "species", "genuses", "families", etc. Unfortunately there isn't any firm rule for excactly where in the tree each division should be made. In fact the basic labelling dates back to a time when we were first putting the tree together and knew a lot less about it than we do now.
IMO it's very unlikely that anyone will move the "Homo" label higher in the tree to make it more inclusive as a result of this. DNA is a great way to do a sanity check on the structure of the tree, but it doesn't really have any bearing on how we label the nodes.
> Of course now with Hollywood and TV causing humans to devolve, the Chimps will have a chance to catch up.
They've got a long way to go... when a recent association broke up the human went on to be president of the USA and the chimp merely spent the rest of his life, I don't know, probably bonking all the female chimps in California or something similarly unrewarding.
> Bonobos are even more evolved than Chimps because they settle things by having sex rather than by fighting.
Or have the best of both worlds with penis fencing.
If you go for that kind of thing...
> But I can tell you what's most special. Art. As soon as a chimp can draw a picture of their house, or of their parents or of anything else; then I'd think about moving them onto the same level with us. This is the clearest place where we see that humans differ from animals by type, not just degree. A human doesn't carve a sculpture well and a chimp does it badly. A human does it well and a chimp can't do it at all. This ability to imagine, mentally conceive and create is the genesis of all that makes humans unique.
> Art is the signature of man.
Check out Koko's paintings. I'm especially fond of the bird.
But Koko's a gorilla rather than a mere chimp...
> If you want to believe that you are almost a chimpanzee, that's fine, but I'm not believing. Note that the researchers ignored DNA that is not expressive. That may be a sensible idea, or it may be that the ignored DNA expresses itself in a way that has not been discovered.
Even when looking at the "junk" DNA that is believed to be non-coding, we are still found to be more similar to the chimps than to any other species - just not at the 99.4% similarity rate.
This is a natural fallout from the way evolution works: the non-coding DNA experiences less selective pressure and is thus likely to accumulate more random mutations over time than the highly selected stuff is. However, since humans and chimps started with the same pool of "junk" DNA and only diverged a few million years ago, fewer random differences have accumulated between us and them than between, say, us and chickens. (Arguably the "junk" DNA is a better way of calculating similarities and time since divergence, since any drift there is thought to be the result of accumulated random mutations rather than the result of any kind of environmental pressure, which obviously works differently in different environments.)
> If 99% of the important DNA is identical, then probability implies that 99% of the rest of the DNA is also identical.
No, the non-coding "junk" DNA will experience less selective pressure and thus drift more or less than random, so you would expect a higher difference rate there, more or less proportional to the time since divergence from the common ancestor.
The reason these researchers got the new 99.4% rather than the lower numbers previously obtained by other methods is because there rest of the DNA isn't 99.4% identical.
> We've only fully mapped the human genome so far. I bet if we fully mapped the chimp genome, we'd see many many more entries in the diff log than we thought.
However, the real reason for the bogosity of the claim is that clades aren't defined by thresholds in DNA differences. The tree of common descent is there, but it's somewhat arbitrary how far up from a leaf you go before you reach a node that you call "species", "genus", etc. They are merely labels of convenience, and if we suddenly do or don't find it convenient to put the chimps in the genus Homo it doesn't really tell us anything we didn't already know about the relationships in that branch of the tree. (I have a physical anthropology textbook published ten years ago that already mapped out this branch of the tree according to our current understanding of it, then already based on DNA comparisons as well.)
The real news from this is that by focusing on "functionally important" genes we now know that our "functionally important similarity" is 99.4% rather than the 97.whatever% that we previously got when looking at genes in general.
> The GPU allows you to do massively parallel computations, but penalizes you heavilly for things such as loops of variable length or reading memory back from the card outside of the once-per-cycle frame update, and the price of interrupting computation is prohibitive. Clearing the graphics pipeline can take a long, long time.
> Furthermore, while there have been a few papers published claiming the orders of magnitude increase in speed in these sorts of computations, none actually demonstrate this sort of speed-up. Everyone's speculating, but when it comes to it, results are lacking.
I looked in to using the GPU for vector * matrix multiplications over my Christmas vacation (yep, a Geek), and everywhere I turned I found people saying that whatever you gained in the number crunching you lost in the latency of sending your numbers to the GPU and reading them back when done. In the end I didn't even bother running an experiment on it.
But maybe conventional wisdom was wrong; elsewhere in the talkbacks I see links to a couple of
> The Fizzer-uninstaller posted there creates the file '%WinDir%\uninstall.pky', which then causes Fizzer to remove all of its registry keys.
Why didn't they provide a UNIX version, too?