He starts by comparing the SSDF computer ratings with the FIDE Elo rating, and claims that they would lead to the conclusion that computers will surpass humans in a couple of months. This is non-sense. Nobody has ever claimed that. The SSDF ratings always come with the disclaimer that they are only based on computer vs computer games, so that the ratings cannot be compared to human ratings.
Then he refutes this claim that noone has ever made by citing the results of human-vs-computer matches in the last 15 years. We see draws in the last 5 years against opponents that had gradually increasing ratings, who probably were taking the matches increasingly more seriously. And that altogether had 49 games. Sorry, I can hardly believe that Sonas is a statistician if he is deriving his claim from such a small sample.
All we can say is that these matches in the last 5 years do not show that computers will be clearly better than humans in a few years. But they do nothing to destroy the very strong evidence to the contrary, either. Evidence gathered in 30 years of computer chess development, e.g. showing that a doubling in computing power leads to about 60 points ELO rating increase.
Hmm, this triggeredmore discussion that I had hoped for, thanx all for your answers. Although I think I had answered that already in the original post, let me first get this straight:
Why do you ask this? Does it matter?
Is there something wrong with being Mormon? Does it have anything to do with them being a good/bad person?
Of course there is nothing wrong with being a Mormon. It does not make a person good or bad.
But religion and atheism both shape your view towards the world, and I would usually expect it being mentioned in a portrait of a person which camp he belongs to, just as I would expect to read from which background he is from, where he is living, whether he is married/has children etc. Of course, mentioning McBride's beliefs in an article dissection his IP claims would be non-sense and would weaken your points. But there were articles going through his entire resume to better understand his actions. I would have expected it to get mentioned there.
I was a little suprised about the many "It does not matter" posts. Surely I would have expected them, but not in that majority. Surely, such is not the typical American opinion when it comes to politicians, as we all know. (I don't think you will be surprised about the story of a respectable woman who told me how she round-about denied her atheism when she was running for an office in a conservative neighbourhood.)
Finally, although this is really dangerous territory, let me try to answer this question:
How exactly would his being mormon change your reaction to some of his statements? And your understanding of him as a person?
I am really not sure. And note that I know little about Mormons in particular. So let me start by answering in very general terms. In my perception, people with strong religious beliefs are often more firm holding up their convictions even when there are standing against a big majority. Pick your favourite example; let me just mention the most positive that comes to my mind, which is that several groups of strong religious believers (catholic, evangelists, "Zeugen Jehovah) were among the last to give up resistance against Nazis in Germany in the 30s (and hence were among those who suffered most badly).
Back to McBride: when he talked about all IP in the open source community, that it would be based on a habit of copyright infringement and a generally lax attitude towards IP, I took that as pure FUD. Now that I know he is a strong religious believer, I have second thoughts whether he might actually be convinced that he is right.
Again, this adds nothing to judging his actions, just to perceiving his personality.
I have a question to the few American slashdot readers, if there are any:
In a German article on SCO/Cannopy, I recently learned that Darl McBride is a "devout mormon". Now, I have never seen this mentioned anywhere in the many articles on SCO/McBride, even when they were specifically talking about McBride's personality.
Of course, his beliefs do in no way at all affect my factual views on his mission against Linux/GPL. But nevertheless, knowing he is a believing Mormon changes my understanding of him as a person, and possibly of his motivations. Maybe it should not, but it does change my reaction to many of his strongly worded statements.
Now I wonder why I never saw this mentioned in American articles. Is this because Americans don't think it matters? Is this because, while they know it might matter, they want to keep that out of the discussion? Is it an unwritten rule that religion is kept out of controversal discussions? Even when analyzing personalities? Or is it just not widely known that he is Mormon? Or is it, to the contrary, just implicitly assumed by US readers that a person such as McBride is probably strongly religious?
I am posting this despite knowing that it might be misunderstood as being negative on religious people, or religion in general. It is not meant that way. It is just that knowing whether someone is deeply religious or not is an important aspect in understanding his motivations.
As a point in favour of your reasoning: When there was the big debate in Germany about Linux use in the German parliament, there was also the question about Windows source code being made available to the German government.
But the source code would never have been allowed to go to the BSI (Federal agency of IT security), which would be the only department of the government with
the resources
the competence
for just a partial audit of the sources. So I agree all this shared-source is just a PR stunt.
Just to add to this: At least two friends of mine that have dual-boot machines say they usually use Windows over Linux because they don't want to wait 2 minutes just to check their e-mail.
(These 2 minutes include X and KDE startup time, as it should.)
Much of this time seems to be spent waiting for disk seeks. so I have always mow much one could have optimize harddisk for boot time.
...but this partnership between DHS and CERT makes me feel uneasy. Would you, if you had discovered a security hole in, say, a widely used FS/OSS application, still notify CERT about it? If you have to assume that the first organization they will share the information with is the DHS?
Yes I know, the DHS hasn't done anything so far to earn this level of distrust from me, in fact they seem to be trying to build up trust, but...I'd definitely feel better just contacting security@mylinuxdistribution.com.
One thing you didn't mention is that GC is deemed to have pretty high processor cache-miss costs. The obvious part is that the GC run itself is basically pointer chasing, i.e. pretty much the worst thing you can do cache-wise. And after the GC run, the cache is clobbered with stuff useless for continuing the work.
There is another indirect cost pointed out by Linus Torvalds in a lengthy post to the gcc mailining list.
The executive summary is that (he thinks that) memory that is not to be used anymore should be freed immediately. Otherwise, the data in there will keep lying around in the data cache.
Also, he claims that explicit ref-counting gives you advantages for optimization: Assume you have to
make some modifications to a data structure, but you don't want other parts of the program to see the modifications. Without ref-counting, you have to copy all the data structure before modifying it. With ref-couting, you can omit the copying if you are the only one with access to the data structure.
And finally, he thinks that GC makes it too easy to write pointer-chasing-heavy code---as that kind of code is bad for cache behaviour all the time.
It is an ongoing discussion whether GC really has that bad effects on performance of GCC. But Linus Torvalds seems to have very good points. (And some of them certainly cannot be taken into account in a "GC cost is less than hand-written memory management"-paper.)
...how to write good user interfaces. With coders like you we will never achieve complete world domination. The correct program is, of course, s.th. like this:
int main() {
int i;
printf("Comparing source trees...\n");
sleep(2);
printf("Check started.\n");
for (i = 1000; i--;) {
printf(".");
sleep(1);
if (i % 100 == 0)
printf("\n%d0 percent remaining\n", i / 100);
}
printf("\n\nThese source trees appear to be entirely different!\n");
return 0; }
Open source laws have often been criticized because they might favour one solution over another for ideological reasons, ignoring the techincal ones.
This should be an obvious case where even the general public might be possible to convince that all the software in such a system must be open source. There is no excuse for not doing so.
Of course, this is not yet the complete solution, but without it I cannot think of one.
The company has shown a recent preference for more moderate courses of action, such as sending invoices to Linux users rather than taking them to court.
Wow. How bad must you behave until sending out invoices to end users, without backing up your claims by any substantial public explanations, is considered a "moderate course of action"???
I am reluctant to give the answer, as every timechess programs come up on slashdot, someone has to make a post referring to Go:
Computers are still very weak at this Asian board game. And despite many people trying to make substantial progress with that. The best open-source one, GNU Go, is btw not very far away from the best commercial ones.
The German "Innenministerium" directly funded the development of gnupg. However, after seeing the small success of the project (in terms of adoption rate), and maybe other reasons, the ministery finally decided to stop direct funding
of OSS development, and instead relying on other means to support OSS.
The other high-profile project funded by the German government is Kollaborate. This was done by the "BSI" (Federal Agency for IT security), which is known to be very Linux-friendly (and equally MS-unfriendly).
Well okay, depends on what you mean by "original". Yeah, at the very beginning, it may have been planned for late 90s. It was pushed backed several times as you say. Then, after 9-11, Congress set a deadline for January 30, 2003. At that time the plans were definitely assuming a much later date for going into production.
...as the political decisions on this project misrespected one of most fundamental and easiest to understand rules of software project management: Never change the release date to be earlier than you originally intended.
After September 11, it was decided that the system had to be used starting early 2003. This was years earlier than in the original plans.
...to see what a biased analysis this is. "Linux unaffected" with 100% certainty? Come on, we all have seen things go horribly wrong in courts. And have seen impact of pure FUD with no basis whatsoever.
Yeah, I know I will get modded down for just suggesting that SCO's action might have some effect, but well...
The problem starts much earlier than with a law suit shutting down a project. Just the impression that a project might get into troubles with IP issues can severely hinder its acceptance in the business world.
The Linux kernel mailing list had long flamewars over patents for realtime Linux due to this. And, IIRC, in one of the MS memos posted by ESR they stated that possible trouble with patents was the most successful message to deter managers from steering towards FS/OSS.
I wanted to moderate on this topic, but sorry, there were hardly any posts worth moderating...
Why can't we once in a while have an interesting non-tech article here without getting hundreds of comments that do nothing but expressing their boredom?
If you think the story is crap, you are free to move on. But this being a discussion forum, and "Isn't it ironic..." being on of its favourite phrases, why shouldn't some of us be interested in reflecting the original (yeah, avoiding "correct" here...) usage of this term, and how it is most commonly used instead these days. After all, with some sensitivity for language subtilities you can be much wittier, impress girls, most important get more slashdot karma... (If you don't believe me, try making jokes in any than your first language -- I had to learn this the hard way when I first came to an English speaking country.)
...by continually continueing to spread any news having anything to do with this case? After all, the strategy of SCO seems to create as many news as possible to transmit their FUD.
Once it stopped appearing daily on all the news sites around, everything would calm down a bit, and the case would generally be seen as a private law suit about trade secrets between two companies having a long history of cross-licensing agreements, i.e. nothing that extraordinary.
But I see that at least some news sites seem to have started understanding this already:
Those lobbying in favour of software patents wanted to have the final vote next week. But it has now been
decided (sorry, link in German) that it will be held in September, as planned originally.
There is nothing special about those architectures with respect to floating point.
(Talking of x86 and x86-64 architectures.)
Well there is one thing that may be a nasty surprise: The fact that x86 processors use registers with 80 bit precisions can mean that two absolutely identical looking computations, when compiled (e.g.) with an optimizing C compiler, can lead to non-identical results. That's because just storing a result from register to memory changes the result (truncating it).
(If using GCC, you can use -ffloat-store to avoid this problem.)
Why would somebody using an open source code be called a 'looter'?
Did you read the article? Someone announcing to sue everyone using Linux while continueing to sell it to its customers can certainly be called a looter, no?
Slashdot Mirror
He starts by comparing the SSDF computer ratings with the FIDE Elo rating, and claims that they would lead to the conclusion that computers will surpass humans in a couple of months. This is non-sense. Nobody has ever claimed that. The SSDF ratings always come with the disclaimer that they are only based on computer vs computer games, so that the ratings cannot be compared to human ratings.
Then he refutes this claim that noone has ever made by citing the results of human-vs-computer matches in the last 15 years. We see draws in the last 5 years against opponents that had gradually increasing ratings, who probably were taking the matches increasingly more seriously. And that altogether had 49 games. Sorry, I can hardly believe that Sonas is a statistician if he is deriving his claim from such a small sample.
All we can say is that these matches in the last 5 years do not show that computers will be clearly better than humans in a few years. But they do nothing to destroy the very strong evidence to the contrary, either. Evidence gathered in 30 years of computer chess development, e.g. showing that a doubling in computing power leads to about 60 points ELO rating increase.
Why do you ask this? Does it matter? Is there something wrong with being Mormon? Does it have anything to do with them being a good/bad person?
Of course there is nothing wrong with being a Mormon. It does not make a person good or bad.
But religion and atheism both shape your view towards the world, and I would usually expect it being mentioned in a portrait of a person which camp he belongs to, just as I would expect to read from which background he is from, where he is living, whether he is married/has children etc. Of course, mentioning McBride's beliefs in an article dissection his IP claims would be non-sense and would weaken your points. But there were articles going through his entire resume to better understand his actions. I would have expected it to get mentioned there.
I was a little suprised about the many "It does not matter" posts. Surely I would have expected them, but not in that majority. Surely, such is not the typical American opinion when it comes to politicians, as we all know. (I don't think you will be surprised about the story of a respectable woman who told me how she round-about denied her atheism when she was running for an office in a conservative neighbourhood.)
Finally, although this is really dangerous territory, let me try to answer this question: How exactly would his being mormon change your reaction to some of his statements? And your understanding of him as a person?
I am really not sure. And note that I know little about Mormons in particular. So let me start by answering in very general terms. In my perception, people with strong religious beliefs are often more firm holding up their convictions even when there are standing against a big majority. Pick your favourite example; let me just mention the most positive that comes to my mind, which is that several groups of strong religious believers (catholic, evangelists, "Zeugen Jehovah) were among the last to give up resistance against Nazis in Germany in the 30s (and hence were among those who suffered most badly).
Back to McBride: when he talked about all IP in the open source community, that it would be based on a habit of copyright infringement and a generally lax attitude towards IP, I took that as pure FUD. Now that I know he is a strong religious believer, I have second thoughts whether he might actually be convinced that he is right.
Again, this adds nothing to judging his actions, just to perceiving his personality.
In a German article on SCO/Cannopy, I recently learned that Darl McBride is a "devout mormon". Now, I have never seen this mentioned anywhere in the many articles on SCO/McBride, even when they were specifically talking about McBride's personality.
Of course, his beliefs do in no way at all affect my factual views on his mission against Linux/GPL. But nevertheless, knowing he is a believing Mormon changes my understanding of him as a person, and possibly of his motivations. Maybe it should not, but it does change my reaction to many of his strongly worded statements.
Now I wonder why I never saw this mentioned in American articles. Is this because Americans don't think it matters? Is this because, while they know it might matter, they want to keep that out of the discussion? Is it an unwritten rule that religion is kept out of controversal discussions? Even when analyzing personalities? Or is it just not widely known that he is Mormon? Or is it, to the contrary, just implicitly assumed by US readers that a person such as McBride is probably strongly religious?
I am posting this despite knowing that it might be misunderstood as being negative on religious people, or religion in general. It is not meant that way. It is just that knowing whether someone is deeply religious or not is an important aspect in understanding his motivations.
Try googling for "Darl McBride" --- and google will answer "Did you mean Dark McBride?"
But the source code would never have been allowed to go to the BSI (Federal agency of IT security), which would be the only department of the government with
- the resources
- the competence
for just a partial audit of the sources. So I agree all this shared-source is just a PR stunt.A big thank you!
(These 2 minutes include X and KDE startup time, as it should.)
Much of this time seems to be spent waiting for disk seeks. so I have always mow much one could have optimize harddisk for boot time.
Yes I know, the DHS hasn't done anything so far to earn this level of distrust from me, in fact they seem to be trying to build up trust, but...I'd definitely feel better just contacting security@mylinuxdistribution.com.
There is another indirect cost pointed out by Linus Torvalds in a lengthy post to the gcc mailining list. The executive summary is that (he thinks that) memory that is not to be used anymore should be freed immediately. Otherwise, the data in there will keep lying around in the data cache. Also, he claims that explicit ref-counting gives you advantages for optimization: Assume you have to make some modifications to a data structure, but you don't want other parts of the program to see the modifications. Without ref-counting, you have to copy all the data structure before modifying it. With ref-couting, you can omit the copying if you are the only one with access to the data structure.
And finally, he thinks that GC makes it too easy to write pointer-chasing-heavy code---as that kind of code is bad for cache behaviour all the time.
It is an ongoing discussion whether GC really has that bad effects on performance of GCC. But Linus Torvalds seems to have very good points. (And some of them certainly cannot be taken into account in a "GC cost is less than hand-written memory management"-paper.)
...how to write good user interfaces. With coders like you we will never achieve complete world domination. The correct program is, of course, s.th. like this:
int main()
{
int i;
printf("Comparing source trees...\n");
sleep(2);
printf("Check started.\n");
for (i = 1000; i--;) {
printf(".");
sleep(1);
if (i % 100 == 0)
printf("\n%d0 percent remaining\n", i / 100);
}
printf("\n\nThese source trees appear to be entirely different!\n");
return 0;
}
This should be an obvious case where even the general public might be possible to convince that all the software in such a system must be open source. There is no excuse for not doing so.
Of course, this is not yet the complete solution, but without it I cannot think of one.
The company has shown a recent preference for more moderate courses of action, such as sending invoices to Linux users rather than taking them to court.
Wow. How bad must you behave until sending out invoices to end users, without backing up your claims by any substantial public explanations, is considered a "moderate course of action"???
Computers are still very weak at this Asian board game. And despite many people trying to make substantial progress with that. The best open-source one, GNU Go, is btw not very far away from the best commercial ones.
There is a Go Wiki with, among other things, a short introdcution.
WOW!
(Late post as the authors has been busy downloading, configuring and compiling 2.6.0-test3 once he saw the OP.)
The other high-profile project funded by the German government is Kollaborate. This was done by the "BSI" (Federal Agency for IT security), which is known to be very Linux-friendly (and equally MS-unfriendly).
Well okay, depends on what you mean by "original". Yeah, at the very beginning, it may have been planned for late 90s. It was pushed backed several times as you say. Then, after 9-11, Congress set a deadline for January 30, 2003. At that time the plans were definitely assuming a much later date for going into production.
After September 11, it was decided that the system had to be used starting early 2003. This was years earlier than in the original plans.
Yeah, I know I will get modded down for just suggesting that SCO's action might have some effect, but well...
The Linux kernel mailing list had long flamewars over patents for realtime Linux due to this. And, IIRC, in one of the MS memos posted by ESR they stated that possible trouble with patents was the most successful message to deter managers from steering towards FS/OSS.
Look how seriously the guy on the right side is watching a fish being drawn...
If you think the story is crap, you are free to move on. But this being a discussion forum, and "Isn't it ironic..." being on of its favourite phrases, why shouldn't some of us be interested in reflecting the original (yeah, avoiding "correct" here...) usage of this term, and how it is most commonly used instead these days. After all, with some sensitivity for language subtilities you can be much wittier, impress girls, most important get more slashdot karma... (If you don't believe me, try making jokes in any than your first language -- I had to learn this the hard way when I first came to an English speaking country.)
Once it stopped appearing daily on all the news sites around, everything would calm down a bit, and the case would generally be seen as a private law suit about trade secrets between two companies having a long history of cross-licensing agreements, i.e. nothing that extraordinary.
But I see that at least some news sites seem to have started understanding this already:
We unsuccessfully tried to ignore the SCO v. IBM fracas, mostly because ...
and
We know just how you feel (lwn.net commenting on the above quote from Linux journal...)
Those lobbying in favour of software patents wanted to have the final vote next week. But it has now been decided (sorry, link in German) that it will be held in September, as planned originally.
Well there is one thing that may be a nasty surprise: The fact that x86 processors use registers with 80 bit precisions can mean that two absolutely identical looking computations, when compiled (e.g.) with an optimizing C compiler, can lead to non-identical results. That's because just storing a result from register to memory changes the result (truncating it).
(If using GCC, you can use -ffloat-store to avoid this problem.)
Did you read the article? Someone announcing to sue everyone using Linux while continueing to sell it to its customers can certainly be called a looter, no?