The DOM is well documented at W3C. At least if all you need is a reference. The whole point of Mozilla is that you should be able to use open standards to write your web pages, not vendor specific standards.
What, exactly, did they steal? A disk containing the source code? Copies of relevant patents? Keep in mind that copyright and patent violations are not theft, at least not in any legal sense, no matter how often corporate spokesmen call it theft.
In what way would that prevent it from being preemptive? Pre-emptive multitasking is when the scheduler can interrupt (and suspend) a running task to run another task. The Amiga had this. How the scheduler decides which task to run next is beside the point.
If the RIAA moves slowly this will only encourage people to copy more music illegally, since 99%+ of the songs on Napster were originally ripped from legitimately purchased CDs. Napster should spur them on to find ways to make it possible for people to easily get music over the net. Since searching for a completely secure method of distributing music is going to be futile, anyway, they should concentrate on getting money from people by giving them what they want.
Of course they may well be too stupid/naive to understand this, but then they should be replaced with people who aren't.
AMD has provided libraries containing 3DNow code for various applications, and there are quite a number of people who write 3DNow! assembly for at least games and video card drivers. This probably gives you a lot better performance than a compiler that automatically generates SIMD code from C/C++, assuming you're clever enough, though it incurs a significant extra expense.
Wow, I was wrong, the Intel C++ compiler does do automatic vectorisation of code, which supposedly gives significant speedups for the P4.
I doubt any compiler supports 3DNow, though, since AMD doesn't do in-house compilers. Though I guess they may have cooperated with, say, Metrowerks, to get 3DNow optimisations in their compiler.
I don't think any C/C++ compilers support SIMD (whether 3DNow, SSE or something else). Possibly on Cray vector computers, but I don't think even they do, at least not without special extensions. Fortran compilers, OTOH, have had support for that kind of stuff on vector computers for a long time, and that technology has probably been adapted for use with at least SSE, since Intel has a pretty good compiler team. Compaq's Visual Fortran compiler may support it, but their web site doesn't actually say it does.
If you want to use 3DNow or SSE in assembly, however, that is supported with recent binutils.
What instructions does the Athlon have that "do more" than instructions in the P4? 3DNow (even the extended version in the Athlon) should be roughly comparable to SSE, and since the P4 has SSE2, it should have approximately the same capabilities.
As for memory technologies, the P4 was designed for Rambus memory, and that's what they're using in these benchmarks.
There is also (AFAIK) no compiler that has real support for optimising code for the Athlon, at least not comparable to Intel's compilers, which do support the P4 (although there probably aren't very many apps compiled for the P4 available yet:). The P4 will probably benefit more from compiler support, though.
The only thing you should have to architect in an unportable way is the guts of the 3D engine. The rest of the game probably won't have to be tailored for the Emotion Engine (including the higher level parts of the 3D engine), if you do your job well. Of course the guts of the 3D engine can be a pretty big part of a modern game...
IANAPS2D (I Am Not A PS2 Developer) so I may be completely wrong about this.
Re:Well, I've been using the nightlies for ���
on
Send Some Mo' Zilla
·
· Score: 2
Mozilla ignores GTK+ theme engines completely, at least for me. It does appear to use the colors of the theme, but buttons, scroll bars, menus, etc. are completely custom, defined by the Mozilla theme you use.
The GPL on the Linux kernel covers only the kernel, not application software, which is what all, or at least most, of the.NET implementation would be. The GPL does not protect things like file formats in any way whatsoever, so if MS wants to replace all config files in a hypothetical Linux.NET with XML files, they can do that, as long as they rewrite the tools or release source for tools licensed under GPL/similar licenses. The GPL only applies to derivative works, not new works.
As for whether people would tolerate it, that's another matter, one which I can't answer.
No SMP chipset for the Athlon has been released yet. The AMD 760MP chipset, which supports SMP and DDR memory, should be released in late 2000, according to AMD. Motherboards will probably follow soon after.
I'm simply saying that the credit for being the first to bring a good graphical user interface to a Unix belongs with Next, not Apple.
Apple, OTOH, may well be the company that finally brings Unix to the masses. They deserve credit for trying to do that. They do not deserve credit for finally bringing a good GUI to Unix, because that was done in 1989, by Next, which was not Apple, even if Jobs and probably some other people had been at Apple before Next.
NextStep has provided a good graphical user interface on top of a Unix variant for 11 years now. Apple bought Next, freshened it up, and added Mac compatibilty and eye candy. I fail to see how that gives them all the credit for it.
Especially as they have broken some of the nice things in Next, like having both buttons on a scroll bar in the same place. Ok, so that's the only thing I know is missing so far, but I haven't even seen Mac OS X in action yet, only a few screenshots.:-)
You can't blame Red Hat for not giving back to the community, because they do. They pay Alan Cox, Stephen C. Tweedie and Ingo Molnar to hack the kernel (probably more people too), they funded a lot of early work on Gnome, they bought Cygnus and freed Source Navigator, not to mention that they release all the software they as part of their distribution as free software. These are just a few of the things Red Hat has done for the community.
Claiming they don't give free stuff back to the community is ridiculous. Though personally I prefer Debian to the Red Hat distribution, I am grateful for all the things that Red Hat has done for all of us.
Gecko has been pretty damn good for a very long time now, it's the rest of Mozilla that's lagging. Konqueror isn't anywhere near as ambitious as Mozilla, even if it may be as standards-compliant once both it and Mozilla are done. Mozilla runs on Windows, Mac and various Unices, Konqueror only (AFAIK) runs on KDE, which only runs on Unices. Mozilla is also supposed to support mail and news, neither of which Konqueror does. I also think Konqueror reused an existing Javascript implementation, whereas the Mozilla one is written from scratch for Mozilla.
This is not to say Konqueror isn't impressive; it is, even though I personally like Mozilla better. And all the stuff in the previous paragraph is AFAIK, I could well be wrong on some of the points, because I haven't done much research.
Since Corel is porting Mozilla to KDE/QT for their own purposes, not as part of the main Mozilla development effort, I don't see how this will delay Mozilla by more than the time it would take the Mozilla crew to say "No, we don't need a QT port, but thanks for asking".
Of course, Corel could have had people help on the main Mozilla effort instead, which could have a positive effect, but let's not forget about the mythical man-month.
The fact that RSA released the RSA algorithm into the public domain two weeks before it would have become public domain anyway says very little about the security of RSA. In fact, RSA keys of 4096 bits are still very hard to crack, AFAIK.
And there are other encryption algorithms in use in open source software already, like Diffie-Hellman, another public key algorithm which is supported by NSS 3.1.
The cheapest possible G4 costs $1549 (from Apples online store). That gives me a 400MHz G4, 64 megs of memory, 20 gig hard disk, DVD-ROM drive, 16 meg ATI Rage 128 Pro graphics.
The closest Dell Dimension 4100 I can build costs $1179, with an 800MHz PIII, 20 gig hard disk, 64 megs of memory, a DVD-ROM drive, a 16 meg Rage 128 Pro, an SB64 PCI (I'm not sure how that compares to a G4, I don't know what their sound hardware does), a 3C905C NIC, Altec Lansing speakers with subwoofer.
So I save $380 for something that is probably faster than a G4 in real use (unless I run Photoshop filters all day). And Dell aren't exactly the cheapest PC maker, and they don't sell Athlons. So it seems to me Macs are still expensive compared to PCs.
I do prefer Apple's design, though. Macs are much sexier than any PC I have seen. Though older SGI boxen are even sexier.
I don't agree with this. Games are for playing, not winning. Contests are for winning. They are not one and the same.
Some people play games to win. Some people play games to experience the game, much like you read a book or watch a movie, except it's interactive. Some people do both. For the second type of gaming, the AI isn't as important - you're not trying to beat the computer, you just want to see what happens when you do something.
This is not to say that AI is only important in contest-like games. Good AI in adventure games could make them far better (good enough, and they could actually be real role-playing games). But Baldur's Gate is fun without good AI. Planescape: Torment is among the best games I've ever played, and it has nearly as poor AI as Baldur's Gate. Whereas Counterstrike against current AIs would most likely be fairly boring.
What law is it that gives you an obligation to enforce a license using the legal system? And there were legitimate breaches of the GPL (the KDE article admits a few instances of such). However, AFAIK no FSF software was involved, so RMS and the FSF couldn't do very much about it, except complain.
In fact, in all the cases so far where someone has used GPL code in violation of the license have been solved outside the court system. Examples are the Next Objective C compiler (based on gcc) and ncftp (used readline), both of which were released under the GPL (though NcFTP seems to be Artistic License now, and no longer uses readline).
Re:License wars are a waste of energy
on
KDE Strikes Back
·
· Score: 2
What people are bitching about is the fact that KDE is not legally redistributable in binary form. KDE is licensed under the GNU GPL, which prohibits distribution unless all of the source can be distributed under the terms of the GPL. KDE uses QT, but QT can't be distributed under the terms of the GPL. Therefore it is not legal to distribute KDE, at least not when linked to QT. Quite simple, really.
This has little to do with ideology. KDE could fix their problems by removing all outside GPL code and changing their license to one that would permit distribution. Troll Tech could fix the problem by making QT available under a GPL-compatible license. Why they don't do this I don't know, but I doubt there is any ideology involved.
MySQL didn't take part in the TPC-C part of the test, because MySQL can't handle TPC-C. It doesn't implement enough of SQL2 for that.
In my professional opionion, MySQL needs to drop the My part and adhere better to the SQL part. At least that would make my life easier.:)
The performance increase over PIII is considerable for the SPECviewperf benchmarks. On most of the benchmarks the Athlon is 10-20% faster than a PIII with RDRAM, and as much as 25% faster if the PIII has 133MHz SDRAM (the same as the Athlon). A 20% performance advantage is not negligible, and the fact that it's also cheaper doesn't exactly hurt.
Slashdot Mirror
The DOM is well documented at W3C. At least if all you need is a reference. The whole point of Mozilla is that you should be able to use open standards to write your web pages, not vendor specific standards.
What, exactly, did they steal? A disk containing the source code? Copies of relevant patents? Keep in mind that copyright and patent violations are not theft, at least not in any legal sense, no matter how often corporate spokesmen call it theft.
In what way would that prevent it from being preemptive? Pre-emptive multitasking is when the scheduler can interrupt (and suspend) a running task to run another task. The Amiga had this. How the scheduler decides which task to run next is beside the point.
If the RIAA moves slowly this will only encourage people to copy more music illegally, since 99%+ of the songs on Napster were originally ripped from legitimately purchased CDs. Napster should spur them on to find ways to make it possible for people to easily get music over the net. Since searching for a completely secure method of distributing music is going to be futile, anyway, they should concentrate on getting money from people by giving them what they want.
Of course they may well be too stupid/naive to understand this, but then they should be replaced with people who aren't.
AMD has provided libraries containing 3DNow code for various applications, and there are quite a number of people who write 3DNow! assembly for at least games and video card drivers. This probably gives you a lot better performance than a compiler that automatically generates SIMD code from C/C++, assuming you're clever enough, though it incurs a significant extra expense.
Wow, I was wrong, the Intel C++ compiler does do automatic vectorisation of code, which supposedly gives significant speedups for the P4.
I doubt any compiler supports 3DNow, though, since AMD doesn't do in-house compilers. Though I guess they may have cooperated with, say, Metrowerks, to get 3DNow optimisations in their compiler.
I don't think any C/C++ compilers support SIMD (whether 3DNow, SSE or something else). Possibly on Cray vector computers, but I don't think even they do, at least not without special extensions. Fortran compilers, OTOH, have had support for that kind of stuff on vector computers for a long time, and that technology has probably been adapted for use with at least SSE, since Intel has a pretty good compiler team. Compaq's Visual Fortran compiler may support it, but their web site doesn't actually say it does.
If you want to use 3DNow or SSE in assembly, however, that is supported with recent binutils.
What instructions does the Athlon have that "do more" than instructions in the P4? 3DNow (even the extended version in the Athlon) should be roughly comparable to SSE, and since the P4 has SSE2, it should have approximately the same capabilities. :). The P4 will probably benefit more from compiler support, though.
As for memory technologies, the P4 was designed for Rambus memory, and that's what they're using in these benchmarks.
There is also (AFAIK) no compiler that has real support for optimising code for the Athlon, at least not comparable to Intel's compilers, which do support the P4 (although there probably aren't very many apps compiled for the P4 available yet
The only thing you should have to architect in an unportable way is the guts of the 3D engine. The rest of the game probably won't have to be tailored for the Emotion Engine (including the higher level parts of the 3D engine), if you do your job well. Of course the guts of the 3D engine can be a pretty big part of a modern game...
IANAPS2D (I Am Not A PS2 Developer) so I may be completely wrong about this.
Mozilla ignores GTK+ theme engines completely, at least for me. It does appear to use the colors of the theme, but buttons, scroll bars, menus, etc. are completely custom, defined by the Mozilla theme you use.
The GPL on the Linux kernel covers only the kernel, not application software, which is what all, or at least most, of the .NET implementation would be. The GPL does not protect things like file formats in any way whatsoever, so if MS wants to replace all config files in a hypothetical Linux.NET with XML files, they can do that, as long as they rewrite the tools or release source for tools licensed under GPL/similar licenses. The GPL only applies to derivative works, not new works.
As for whether people would tolerate it, that's another matter, one which I can't answer.
No SMP chipset for the Athlon has been released yet. The AMD 760MP chipset, which supports SMP and DDR memory, should be released in late 2000, according to AMD. Motherboards will probably follow soon after.
I'm simply saying that the credit for being the first to bring a good graphical user interface to a Unix belongs with Next, not Apple.
Apple, OTOH, may well be the company that finally brings Unix to the masses. They deserve credit for trying to do that. They do not deserve credit for finally bringing a good GUI to Unix, because that was done in 1989, by Next, which was not Apple, even if Jobs and probably some other people had been at Apple before Next.
NextStep has provided a good graphical user interface on top of a Unix variant for 11 years now. Apple bought Next, freshened it up, and added Mac compatibilty and eye candy. I fail to see how that gives them all the credit for it. :-)
Especially as they have broken some of the nice things in Next, like having both buttons on a scroll bar in the same place. Ok, so that's the only thing I know is missing so far, but I haven't even seen Mac OS X in action yet, only a few screenshots.
You can't blame Red Hat for not giving back to the community, because they do. They pay Alan Cox, Stephen C. Tweedie and Ingo Molnar to hack the kernel (probably more people too), they funded a lot of early work on Gnome, they bought Cygnus and freed Source Navigator, not to mention that they release all the software they as part of their distribution as free software. These are just a few of the things Red Hat has done for the community.
Claiming they don't give free stuff back to the community is ridiculous. Though personally I prefer Debian to the Red Hat distribution, I am grateful for all the things that Red Hat has done for all of us.
Gecko has been pretty damn good for a very long time now, it's the rest of Mozilla that's lagging. Konqueror isn't anywhere near as ambitious as Mozilla, even if it may be as standards-compliant once both it and Mozilla are done. Mozilla runs on Windows, Mac and various Unices, Konqueror only (AFAIK) runs on KDE, which only runs on Unices. Mozilla is also supposed to support mail and news, neither of which Konqueror does. I also think Konqueror reused an existing Javascript implementation, whereas the Mozilla one is written from scratch for Mozilla.
This is not to say Konqueror isn't impressive; it is, even though I personally like Mozilla better. And all the stuff in the previous paragraph is AFAIK, I could well be wrong on some of the points, because I haven't done much research.
That's because Galeon is spelled Galeon.
Since Corel is porting Mozilla to KDE/QT for their own purposes, not as part of the main Mozilla development effort, I don't see how this will delay Mozilla by more than the time it would take the Mozilla crew to say "No, we don't need a QT port, but thanks for asking".
Of course, Corel could have had people help on the main Mozilla effort instead, which could have a positive effect, but let's not forget about the mythical man-month.
The fact that RSA released the RSA algorithm into the public domain two weeks before it would have become public domain anyway says very little about the security of RSA. In fact, RSA keys of 4096 bits are still very hard to crack, AFAIK.
And there are other encryption algorithms in use in open source software already, like Diffie-Hellman, another public key algorithm which is supported by NSS 3.1.
The cheapest possible G4 costs $1549 (from Apples online store). That gives me a 400MHz G4, 64 megs of memory, 20 gig hard disk, DVD-ROM drive, 16 meg ATI Rage 128 Pro graphics.
The closest Dell Dimension 4100 I can build costs $1179, with an 800MHz PIII, 20 gig hard disk, 64 megs of memory, a DVD-ROM drive, a 16 meg Rage 128 Pro, an SB64 PCI (I'm not sure how that compares to a G4, I don't know what their sound hardware does), a 3C905C NIC, Altec Lansing speakers with subwoofer.
So I save $380 for something that is probably faster than a G4 in real use (unless I run Photoshop filters all day). And Dell aren't exactly the cheapest PC maker, and they don't sell Athlons. So it seems to me Macs are still expensive compared to PCs.
I do prefer Apple's design, though. Macs are much sexier than any PC I have seen. Though older SGI boxen are even sexier.
I don't agree with this. Games are for playing, not winning. Contests are for winning. They are not one and the same.
Some people play games to win. Some people play games to experience the game, much like you read a book or watch a movie, except it's interactive. Some people do both. For the second type of gaming, the AI isn't as important - you're not trying to beat the computer, you just want to see what happens when you do something.
This is not to say that AI is only important in contest-like games. Good AI in adventure games could make them far better (good enough, and they could actually be real role-playing games). But Baldur's Gate is fun without good AI. Planescape: Torment is among the best games I've ever played, and it has nearly as poor AI as Baldur's Gate. Whereas Counterstrike against current AIs would most likely be fairly boring.
What law is it that gives you an obligation to enforce a license using the legal system? And there were legitimate breaches of the GPL (the KDE article admits a few instances of such). However, AFAIK no FSF software was involved, so RMS and the FSF couldn't do very much about it, except complain.
In fact, in all the cases so far where someone has used GPL code in violation of the license have been solved outside the court system. Examples are the Next Objective C compiler (based on gcc) and ncftp (used readline), both of which were released under the GPL (though NcFTP seems to be Artistic License now, and no longer uses readline).
What people are bitching about is the fact that KDE is not legally redistributable in binary form. KDE is licensed under the GNU GPL, which prohibits distribution unless all of the source can be distributed under the terms of the GPL. KDE uses QT, but QT can't be distributed under the terms of the GPL. Therefore it is not legal to distribute KDE, at least not when linked to QT. Quite simple, really.
This has little to do with ideology. KDE could fix their problems by removing all outside GPL code and changing their license to one that would permit distribution. Troll Tech could fix the problem by making QT available under a GPL-compatible license. Why they don't do this I don't know, but I doubt there is any ideology involved.
MySQL didn't take part in the TPC-C part of the test, because MySQL can't handle TPC-C. It doesn't implement enough of SQL2 for that. :)
In my professional opionion, MySQL needs to drop the My part and adhere better to the SQL part. At least that would make my life easier.
The performance increase over PIII is considerable for the SPECviewperf benchmarks. On most of the benchmarks the Athlon is 10-20% faster than a PIII with RDRAM, and as much as 25% faster if the PIII has 133MHz SDRAM (the same as the Athlon). A 20% performance advantage is not negligible, and the fact that it's also cheaper doesn't exactly hurt.