I agree wholeheartedly that the low error rate of the Jewish scribes is very impressive, as you say. But to be accurate we must admit that it isn't perfect.
However, you say it is the "best error/time ratio of any document in existence" - I don't know about that. The Koran was copied for almost a thousand years before the printing press, and those transcriptions may be 100% exact, for all I know - we would need to ask a scholar in that field. (Yet even if it is, this may have something to do with the Jewish text changing punctuation during its history, which makes things hard - the Muslims have not had that difficulty.) There are also very old Hindu and Buddhist texts (much older than the Koran), but I know little about them. Likewise with Chinese ancient literature.
Perhaps someone here can enlighten us about these matters, it's a fascinating topic.
Compare the Dead Sea Scrolls to modern Torahs and you'll find it's letter-for-letter exact in the parts where they overlap.
Actually, that is false. A nice summary of this matter can be found here, pages 27-30 in the PDF version (note that the rest of the paper is interesting also).
Some examples: (1) there is anywhere between 1 character in 20 and 1 in 2000 difference between the dead sea scrolls fragments and the current text; (2) truly identical copies of the Torah are found only from the 16th century on, and those are not handwritten; (3) even today there are slightly-different versions of the Torah in use, e.g. the Yemenite version differs in 3 characters from the Koren (which is perhaps the 'standard').
So, by no means has the text been copied without error, at least not according to the people researching this topic.
Google news is unashamedly breaking copyright.. there's no argument there - the real question is why anyone would prosecute over something that's driving hits to their page and generating ad revenue?
For control. The sense of losing control scares them so much, they instinctively react. They figure that once they have everything in hand, they can then arrive at whatever arrangements they want (possibly similar ones to the status quo until a few days ago).
Apparently what is probably the premier desktop-oriented Linux distro doesn't think it's stable enough to include, but it's just as good - nay, better - than Aqua and Aero ?
Sounds like just another day in Linux-land to me:).
You are comparing them on one aspect, in which admittably the Linux offerings fall short - stability. But the people who say that Compiz/Beryl are better aren't talking about that, they are referring to other aspects - say, that they require less in the way of hardware (especially vs. Vista), or that they allow a lot more user customization.
I'd be interested to see if this might result in things like zfs being ported natively to the Linux kernel (rather than the current FUSE-based solution).
Well, based on TFA, that won't happen::
Using GPLv3 for Solaris likely would preclude Linux programmers from using Solaris software, and vice-versa. That would make it difficult for Solaris to benefit from hardware support built into Linux, or for Linux to benefit from performance tools built into Solaris.
And indeed, since Linux will stay GPL2, it can't utilize code from OpenSolaris, which will be (perhaps) GPL3. However, TFA is wrong about the other direction. Some 40% or so of Linux source code is licensed under "GPL2 or above", which means that OpenSolaris could use it, as I understand the issue.
I agree that the paper isn't revolutionary. In addition, it turns out that, after the 4-layer biologically-motivated system, they feed everything into a linear SVM (Support Vector Machine) or gentleBoost. For those that don't know, SVMs and Boosting are the 'hottest' topics in machine learning these days; they are considered the state of the art in that field. So basically what they are doing is providing some preprocessing before applying standard, known methods. (And if you say "but it's a linear SVM", well, it is linear because the training data is already separable.)
That said, they do present a simple and biologically-motivated preprocessing layer that appears to be useful, which reflects back on the brain. In summary, I would say that this paper helps more to understand brain functioning than to develop machines that can achieve human-like vision capabilities. So, very nice, but let's not over-hype it.
Please note that I said "If we don't get such a statement, that will mean something in itself." I did not say that they were guilty if they didn't make such a statement.
If they don't make such a statement, it would mean any of the following: (1) they are guilty, (2) they are innocent, but like to appear as if they are guilty (to scare people away from Linux), (3) they are innocent, but don't want to make any such general statements, as it may limit their future actions. Now, (1) is bad, (2) is also fairly bad IMO; (3) is debatable, I guess (perhaps they want to reserve the option to retaliate against patent threats against them).
But in any case, lack of a statement is informative, as I said. Personally I would like to hear such a statement from them, to clear things up.
From the information in the article, it appears that the economics of open source work much better than the economics of closed-source, proprietary software. The business model of OpenOffice.org is perfectly happy when local vendors sell their software at $2 per disk. The business model that Microsoft Office is based upon is violated when that happens.
Well said. But there is another perspective on this: the economics of FOSS and the real economics of Microsoft share some things in common. Both are based on the simple fact that software can be copied at virtually no cost; this lets FOSS exist (and explains why you don't see FOSS automobiles), and this allows Microsoft to (1) benefit from their product being pirated, as it creates market share that may pay later, (2) bundle whatever they want into the OS (it doesn't cost them any more), and kill competitors.
Piracy in the third world is really just part of MS's business plan. Bill Gates has even admitted as much (I can find the link if anyone wants it).
So I don't see how you could ever make an 'open source' DRM system, as long as there are a lot of talented programmers out there who would defang it as soon as they got their hands on the source code. So you'd either have to radically alter the definition of 'open source,' or you'd have to convince a lot of people that such a scheme was a Good Idea, to make it work.
You are 100% right, but you can look at the issue of DRM in another way.
If DRM is meant to prevent ALL users from using data in some way (copying it), then an OSS DRM would not achieve that goal, as you say. However, how about the goal of preventing casual users from using data in some way. Casual computer users are (making up a number here) 90% of computer users. So this would still have a use.
The remaining 'power-users' would have the source code, and would be able to modify it and circumvent the DRM, true. But the makers of the DRM might hope that the 'honor system' would stop most from doing so, and you know what - it might (if the entire system is more or less fair). Also, even if a few create a hacked DRM-less version that they give to casual users, the usual legal methods can be used against that. But even if they weren't used, most casual users would use iTunes from Apple and WMP from Microsoft, and not hacked versions of them, I am guessing.
So, OSS DRM would be a compromise between the current DRM and no DRM. I am personally in favor of no DRM, but I might be convinced to support some form of OSS DRM, in theory - as I said, if the entire system around it were very fair.
This is clearly a case of not just prosecutorial misconduct, but malicious prosecution of the worst kind.
It seems so, but TFA was lacking in details on one area (but perhaps it is my lack of knowledge) - weren't they first convicted by a jury of their peers? If so, then shouldn't we be wondering why they did so? Very odd.
It has already been established that DRM is bad. It doesn't work and it hurts everybody.
Also, Chillout isn't the only open-source DRM project, Sun has (had?) their DReaM initiative. But none of these attempts seem to be gaining any traction. The only widespread DRM scheme is Apple's, and Jobs himself says they would rather not be using it at all. The media companies should listen to him and finish the entire embarrassing affair.
Clearly MS hasn't learned its lesson yet. While I don't begrudge MS for trying to protect their copyrights against piracy, they do need to learn about subtlety.
MS have, in all likelihood, learned their lessons very well. What they have is actual global figures for their performance, and we Slashdot readers have only 3 anecdotes in this story. They can afford a policy that costs them a few users here but raises profits in many other places.
Yet, the Novell deal sounds new, doesn't it? Migration to a Linux desktop. Nice.
And the reason it won't get any funding to study whether or not it's a real cure for cancer is because there's no money in it! If it's a cheap solution and it magically cures cancer... where's the profit in that?
So people will continue to die from cancer, who could have been cured by this cheap drug, because it would offset the bottom line.
That would have been true, had our entire civilization been in the control of corporations. Luckily enough, this is not the case - this drug will, in all likelihood, get investigated, by academic researchers, using government funds and donations.
What motivation do they have? Well, if a virtual 'cure for cancer' is at hand, what researcher wouldn't want their name attached to it? What elderly rich person wouldn't feel good knowing they had contributed part of their fortune to that cause? What government official wouldn't want to support such a cause (if only to reduce healthcare costs)? And I am just speaking about non-altruistic motives here.
The for-profit mindset is not everything, even today. Both approaches have their place. Let's keep it that way.
DNA is far from perfect. Semen in a rape case, victims blood on murderers clothes, those are workable applications. But when you hoover a crimescene and test everything, suddenly people with even multiple degrees of seperation become suspects. You may not have commited the murder, but your eyelash was found on the victim. It fell onto that guys shoulder that you bumped into on the street. He's the murderer.
DNA evidence is like any tool: you can use it the right way, and you can use it the wrong way. Convincing someone based on an eyelash - and nothing more - is clearly misuse of DNA (by an incompetent DA and lazy jury). But finding an eyelash and using that as a lead, gathering further evidence that proves or disproves a suspect's innocence, is perfectly fine.
Yet, as mentioned elsewhere in this thread, Linus' statement that you quoted is of dubious legal standing. If a source file contains "GPL 2 or above", then which takes precedence - the license granted by the copyright owner of that file, or a claim made by Linus?
IANAL, I don't know. But it seems to me that the copyright owner decides.
Some of that may be for caches which would be released if an application requested more RAM. The OS might as well make use of it to reduce latency of other tasks whilst nothing else wants the RAM.
That is how Linux reports memory usage (in 'free', for example). But Windows has never done so, at least not Windows 95 til XP. Used RAM was RAM used by applications, not caching. However, perhaps Vista changes this, I don't know - never used it.
Well, I guess the lawyers will need to figure out the licensing issues, I certainly am not one. So I am not claiming to be sure of any of the legal issues; I just know what I read. Time will tell, I guess.
Changing the entire kernel to GPL3 would be very hard, as you say, while getting various parts of it would be much easier. Now, obviously you are right that 'bits and pieces' of the kernel are not functional by themselves. Yet OpenSolaris would want precisely just 'bits and pieces'; they already have a (apparently quite good) basic kernel. What they need are e.g. device drivers. Now, obviously you can't just paste Linux device driver code into OpenSolaris, but you might be able to (a) alter the Linux version much faster than writing it from scratch, or (b) write a 'compatibility' layer to translate some API calls, to automate or semi-automate the process. In either case, OpenSolaris gets a significant push in the device driver area, I would say. This by itself is a good reason for them to adopt the GPL3.
I may be wrong, but all I have read points in the other direction. Basically, various people contributed to the Linux kernel, and they wrote their own licensing texts. Some said "2 only", some "2", and some "2 or above". See Wikipedia for the following quote:
"[...]the terms of the GPL state that if no version is specified, then any version may be used, and Alan Cox pointed out that very few other Linux contributors [other than Linus] have specified a particular version of the GPL."
So OpenSolaris would be able to use most Linux code, it seems.
Well, from what I know, it would be really difficult to migrate the Linux Kernel from GPL2 to GPL3, even/if/ Linus liked to do so as each contribution is copyright of the contributor.
Difficult yes, impossible - no. And Linus may get additional motivation soon. If OpenSolaris goes GPL3, then it can use any "GPL2 or above" code from the Linux kernel (which I have heard is the majority). The Linux kernel, on the other hand, won't be able to use anything from OpenSolaris. This 'one-way-street' of code certainly isn't in Linux's interest.
well, I do think that most of the gnu programs (gcc, make, ncurses, wget), etc. will actually be GPLv3 when it came out.
Yes. Also GNOME (which is the default on SUSE these days), and Samba.
But lots of other stuff won't. So a sizable part of SUSE will be moving to GPL3 (barring any surprises), but not the overwhelming majority (e.g. the kernel).
The article is poorly written. For example, the GPL3 is referred to ("If the foundation decides to take action, the ban would apply to new versions of Linux covered under a licensing agreement due to take effect in March."), but the writer doesn't explain that 'Linux' won't be under this license - only parts of it.
However, the point is somewhat (perhaps) valid - if most of SUSE goes GPL3, and if the GPL3 is indeed in conflict with the Novell-Microsoft agreement, then there may be an issue (both qualifications seem likely, at present, but time will tell). The issue may be easily solvable, however, depending on the details of the Novell-Microsoft deal - which we do not know (Eben Moglen, however, supposedly does, or so we have been told).
Slashdot Mirror
I agree wholeheartedly that the low error rate of the Jewish scribes is very impressive, as you say. But to be accurate we must admit that it isn't perfect.
However, you say it is the "best error/time ratio of any document in existence" - I don't know about that. The Koran was copied for almost a thousand years before the printing press, and those transcriptions may be 100% exact, for all I know - we would need to ask a scholar in that field. (Yet even if it is, this may have something to do with the Jewish text changing punctuation during its history, which makes things hard - the Muslims have not had that difficulty.) There are also very old Hindu and Buddhist texts (much older than the Koran), but I know little about them. Likewise with Chinese ancient literature.
Perhaps someone here can enlighten us about these matters, it's a fascinating topic.
Some examples: (1) there is anywhere between 1 character in 20 and 1 in 2000 difference between the dead sea scrolls fragments and the current text; (2) truly identical copies of the Torah are found only from the 16th century on, and those are not handwritten; (3) even today there are slightly-different versions of the Torah in use, e.g. the Yemenite version differs in 3 characters from the Koren (which is perhaps the 'standard').
So, by no means has the text been copied without error, at least not according to the people researching this topic.
You are comparing them on one aspect, in which admittably the Linux offerings fall short - stability. But the people who say that Compiz/Beryl are better aren't talking about that, they are referring to other aspects - say, that they require less in the way of hardware (especially vs. Vista), or that they allow a lot more user customization.
So, you are all right.
I think it's a great idea, especially if the muted player gets a notification of the muting
Notification? How about if it plays his audio back to him, at normal volume times the number of people muting him? (a law of conservation of sorts)
Well, based on TFA, that won't happen:
I agree that the paper isn't revolutionary. In addition, it turns out that, after the 4-layer biologically-motivated system, they feed everything into a linear SVM (Support Vector Machine) or gentleBoost. For those that don't know, SVMs and Boosting are the 'hottest' topics in machine learning these days; they are considered the state of the art in that field. So basically what they are doing is providing some preprocessing before applying standard, known methods. (And if you say "but it's a linear SVM", well, it is linear because the training data is already separable.)
That said, they do present a simple and biologically-motivated preprocessing layer that appears to be useful, which reflects back on the brain. In summary, I would say that this paper helps more to understand brain functioning than to develop machines that can achieve human-like vision capabilities. So, very nice, but let's not over-hype it.
Please note that I said "If we don't get such a statement, that will mean something in itself." I did not say that they were guilty if they didn't make such a statement.
If they don't make such a statement, it would mean any of the following: (1) they are guilty, (2) they are innocent, but like to appear as if they are guilty (to scare people away from Linux), (3) they are innocent, but don't want to make any such general statements, as it may limit their future actions. Now, (1) is bad, (2) is also fairly bad IMO; (3) is debatable, I guess (perhaps they want to reserve the option to retaliate against patent threats against them).
But in any case, lack of a statement is informative, as I said. Personally I would like to hear such a statement from them, to clear things up.
Well, for starters, how about a statement from Microsoft, stating that such things do not occur?
If we don't get such a statement, that will mean something in itself.
Piracy in the third world is really just part of MS's business plan. Bill Gates has even admitted as much (I can find the link if anyone wants it).
You are 100% right, but you can look at the issue of DRM in another way.
If DRM is meant to prevent ALL users from using data in some way (copying it), then an OSS DRM would not achieve that goal, as you say. However, how about the goal of preventing casual users from using data in some way. Casual computer users are (making up a number here) 90% of computer users. So this would still have a use.
The remaining 'power-users' would have the source code, and would be able to modify it and circumvent the DRM, true. But the makers of the DRM might hope that the 'honor system' would stop most from doing so, and you know what - it might (if the entire system is more or less fair). Also, even if a few create a hacked DRM-less version that they give to casual users, the usual legal methods can be used against that. But even if they weren't used, most casual users would use iTunes from Apple and WMP from Microsoft, and not hacked versions of them, I am guessing.
So, OSS DRM would be a compromise between the current DRM and no DRM. I am personally in favor of no DRM, but I might be convinced to support some form of OSS DRM, in theory - as I said, if the entire system around it were very fair.
He's a black man running for president that actually has a chance.
If he gets elected, I hope he acknowledges his debt to David Palmer.
(Seriously, though, things like that may have an effect. Never underestimate the power of the media. In this case, at least, for good.)
This is clearly a case of not just prosecutorial misconduct, but malicious prosecution of the worst kind.
It seems so, but TFA was lacking in details on one area (but perhaps it is my lack of knowledge) - weren't they first convicted by a jury of their peers? If so, then shouldn't we be wondering why they did so? Very odd.
It has already been established that DRM is bad. It doesn't work and it hurts everybody.
Also, Chillout isn't the only open-source DRM project, Sun has (had?) their DReaM initiative. But none of these attempts seem to be gaining any traction. The only widespread DRM scheme is Apple's, and Jobs himself says they would rather not be using it at all. The media companies should listen to him and finish the entire embarrassing affair.
MS have, in all likelihood, learned their lessons very well. What they have is actual global figures for their performance, and we Slashdot readers have only 3 anecdotes in this story. They can afford a policy that costs them a few users here but raises profits in many other places.
Yet, the Novell deal sounds new, doesn't it? Migration to a Linux desktop. Nice.
That would have been true, had our entire civilization been in the control of corporations. Luckily enough, this is not the case - this drug will, in all likelihood, get investigated, by academic researchers, using government funds and donations.
What motivation do they have? Well, if a virtual 'cure for cancer' is at hand, what researcher wouldn't want their name attached to it? What elderly rich person wouldn't feel good knowing they had contributed part of their fortune to that cause? What government official wouldn't want to support such a cause (if only to reduce healthcare costs)? And I am just speaking about non-altruistic motives here.
The for-profit mindset is not everything, even today. Both approaches have their place. Let's keep it that way.
DNA evidence is like any tool: you can use it the right way, and you can use it the wrong way. Convincing someone based on an eyelash - and nothing more - is clearly misuse of DNA (by an incompetent DA and lazy jury). But finding an eyelash and using that as a lead, gathering further evidence that proves or disproves a suspect's innocence, is perfectly fine.
Yet, as mentioned elsewhere in this thread, Linus' statement that you quoted is of dubious legal standing. If a source file contains "GPL 2 or above", then which takes precedence - the license granted by the copyright owner of that file, or a claim made by Linus?
IANAL, I don't know. But it seems to me that the copyright owner decides.
Some of that may be for caches which would be released if an application requested more RAM. The OS might as well make use of it to reduce latency of other tasks whilst nothing else wants the RAM.
That is how Linux reports memory usage (in 'free', for example). But Windows has never done so, at least not Windows 95 til XP. Used RAM was RAM used by applications, not caching. However, perhaps Vista changes this, I don't know - never used it.
Well, I guess the lawyers will need to figure out the licensing issues, I certainly am not one. So I am not claiming to be sure of any of the legal issues; I just know what I read. Time will tell, I guess.
Changing the entire kernel to GPL3 would be very hard, as you say, while getting various parts of it would be much easier. Now, obviously you are right that 'bits and pieces' of the kernel are not functional by themselves. Yet OpenSolaris would want precisely just 'bits and pieces'; they already have a (apparently quite good) basic kernel. What they need are e.g. device drivers. Now, obviously you can't just paste Linux device driver code into OpenSolaris, but you might be able to (a) alter the Linux version much faster than writing it from scratch, or (b) write a 'compatibility' layer to translate some API calls, to automate or semi-automate the process. In either case, OpenSolaris gets a significant push in the device driver area, I would say. This by itself is a good reason for them to adopt the GPL3.
I may be wrong, but all I have read points in the other direction. Basically, various people contributed to the Linux kernel, and they wrote their own licensing texts. Some said "2 only", some "2", and some "2 or above". See Wikipedia for the following quote:
"[...]the terms of the GPL state that if no version is specified, then any version may be used, and Alan Cox pointed out that very few other Linux contributors [other than Linus] have specified a particular version of the GPL."
So OpenSolaris would be able to use most Linux code, it seems.
Well, from what I know, it would be really difficult to migrate the Linux Kernel from GPL2 to GPL3, even /if/ Linus liked to do so as each contribution is copyright of the contributor.
Difficult yes, impossible - no. And Linus may get additional motivation soon. If OpenSolaris goes GPL3, then it can use any "GPL2 or above" code from the Linux kernel (which I have heard is the majority). The Linux kernel, on the other hand, won't be able to use anything from OpenSolaris. This 'one-way-street' of code certainly isn't in Linux's interest.
well, I do think that most of the gnu programs (gcc, make, ncurses, wget), etc. will actually be GPLv3 when it came out.
Yes. Also GNOME (which is the default on SUSE these days), and Samba.
But lots of other stuff won't. So a sizable part of SUSE will be moving to GPL3 (barring any surprises), but not the overwhelming majority (e.g. the kernel).
The article is poorly written. For example, the GPL3 is referred to ("If the foundation decides to take action, the ban would apply to new versions of Linux covered under a licensing agreement due to take effect in March."), but the writer doesn't explain that 'Linux' won't be under this license - only parts of it.
However, the point is somewhat (perhaps) valid - if most of SUSE goes GPL3, and if the GPL3 is indeed in conflict with the Novell-Microsoft agreement, then there may be an issue (both qualifications seem likely, at present, but time will tell). The issue may be easily solvable, however, depending on the details of the Novell-Microsoft deal - which we do not know (Eben Moglen, however, supposedly does, or so we have been told).
"Deathly" is a perfectly good word. It means "like or resembling death."
Indeed. A common phrase, probably familiar to most bookreaders, is "deathly pale".