Yeah, it works so well that the glorious US army is attacking one weak country on the verge of self implosion after the other instead of going after someone who can defend himself, like, for example, uh, North Korea?
I suggest you read this Wired article to see how well it works. They are using Microsoft Chat on the battlefield, for crying out loud! Yes, the one with the comic characters, where the staff seargant looks like a big breasted bimbo. Nothing inspired a feeling of technical superiority like that, if you ask me.
First he takes over maintenance of BIND, then proceeds to blackmail his "customers" into paying him to be told about the copious amount of security problems, essentially capitalizing on the work of others.
Then he had BIND 9 written, which has nothing to do with BIND. We call it BIND then? Again, to capitalize on the work of others (not that the BIND name is something to carry proudly).
Now he patents what Linux could do at least since 2.0 (1996) with a patched squid 1 (been there, done that). Granted, his patent appears to be earlier than Akamai (1998) but Digital Island may have been early enough to invalidate the patent. They were bought by Cable and Wireless a while ago.
On the research level gcc is not as bleeding edge as other compilers. So if you run example code that shows the merits of a particular optimization, gcc may look not so good. But in practice, it's quite good.
My experiences with UltraSPARC are also a few years old, but gcc was faster and produced better code than Sun CC back then. You have to make sure to set -march=ultrasparc, of course. And I'm not sure about UltraSPARC but normally gcc -O4 does not do more than -O3, which basically is -O2 with function inlining. You can also get some boost with profile based optimization with gcc.
In summary, gcc produces very good code, but you might have to use some little known options for it. For example, gcc on Athlon XP and Pentium >= 3 may gain significant floating point performance with -mfpmath=sse,387 (I got >10% speed-up on lame, gcc's code was even faster than icc's with vectorizer). Another option worth knowing is -malign-double and the regparm attribute.
Another thing you have to keep in mind is: recent optimization advances normally are not big breakthroughs but small incremental advances. Many of them only help in a handful of special cases. gcc 3 has many more optimizations than gcc 2.95.3 and they were so proud of it that they said "much faster code on x86", and then there was whining and gnashing of teeth when most software was unaffected or even slower.
The only platform where I really would prefer the vendor cc is HP-UX on PA-RISC. The HP CC consistently produced 10-30% faster code than gcc (although that may have changed, I haven't used gcc > 2.7 on HP-UX).
It is quite difficult to produce better code than gcc, and my tests on powerpc (granted, those were a few years back using xlc on RS6000 with AIX 4) showed that xlc produced code of about the same quality -- sometimes worse, sometimes better.
The gcc "Haifa" scheduler was donated by IBM Haifa, by the way, so I think it's not surprising that gcc produces good code on powerpc.
On Intel it's quite the same, except that gcc does not vectorize code. From what I have seen, however, icc's vectorizer is not very useful either. I recently tested ogg-vorbis (which is a plain C floating point intensive benchmark) with icc 7 and gcc 3.3 and the gcc version was actually faster than the icc version (on my Athlon XP, target CPU pentium3) despite icc having vectorized several loops.
So all this "vendor-optimized C compiler" stuff is really besides the point. No C compiler will ever be able to match the quality of hand optimized assembler code, and the most important code (ffmpeg MPEG-2 decoder and MPEG-4 codec) has already been hand-optimized. You might be able to squeeze anoter 5 percent out of your code by using a vendor C compiler with insane optimizer settings, but what good is that if the end user is only going to use gcc anyway. I know I am, so I find the numbers for gcc actually more useful for comparison purposes than some vendor C compiler comparison.
Also, we don't want to encourage vendors to produce super vendor optimizing compilers, we want them to optimize gcc (so that everyone benefits, not just their users). So the more benchmarks are done using gcc, the better!
When I said "use it commercially" I meant "make money selling it under a commercial license". To make money with it, they need to modify it ("add value") and the GPL forces them to make their diffs available under GPL as well, which basically means someone will send the diffs to me and I may incorporate them in my version, making their added value available to everyone.
So the fact remains, if someone wants to make money selling my software (and I'm not talking about Red Hat or other distributors of my software here), he needs to talk to me.
Also, I wonder what you mean by "poor". I have a nice little family, can pay my bills, and get paid doing what I like to do -- what more could I possibly want from life? I don't have to be a millionaire. To me it's more about what remains when you die, and when I die, I will leave some offspring and some (fine?) free software behind. What (besides some mediocre slashdot trolls) are you planning to leave behind for future generations?
The point about GPL is that you can't get ripped off. If they rip you off, you can force them to release their derivative work also as GPL. If he chose the wrong license, he got what he deserved.
I put my embedded work under GPL and actually managed to get some funding. If it's GPL, people have to talk to you to use it commercially, you know? That's the beauty of GPL.
Anyway, I can't say I found LRP to be as great as this guy actually thinks it is. And this childish "look what you missed" bullshit is not going to get him anywhere either. The world is full of companies who are not making any money, Caldera and Lineo being two very good examples he cites himself. Don't expect them to pay you if they don't have to.
So far, almost every company that hired has tried to rip me off in the end. That's how it goes. So choose wisely, chose GPL.
BTW: A new init system? Got one of those as well... I even wrote my own libc. And you know what? People are helping with the projects, in fact, many people are helping me with the projects. Feel free to look at all the names in the dietlibc CHANGES file! I think it's how you treat people that makes them help you. If your code is readable and you treat people well, they will help. You won't get big front page articles on Wired, but you'll create a damn good project, people will know your name. And you will get invited, too! Meet me at Linuxtag 2003!;)
This is already law for newspapers, and why would internet sites be held to a lesser standard?
And what is the alternative? Facing countless lawsuits? I think it would be less easy to sue someone if he already had to publish your clarification.
And it doesn't say you would have to delete your original or that you can't make sure everyone understands you were forced by law to publish the "clarification" and you still stand by your original report.
Many GNU tools violate POSIX in some way, for example! To find examples, grep the GNU documentation for POSIXLY_CORRECT. One example is GNU du which will display the size in kilobytes instead of blocks (1 block = 512 bytes), which is a much more useful display.
Also, many open standards stink to the high heavens.
And how do you differentiate between willful and accidental violation of a standard? Yeah, I know, you don't care as long as Microsoft has to pay;-)
The MD5 sums will only match if the lines are identicaly.
Change 8 spaces to a tab? MD5 mismatch.
Change indentation? MD5 mismatch.
Add a comment? MD5 mismatch.
Rename a variable from i to counter? MD5 mismatch.
Rename a function from calculatechecksum to CalculateChecksum? MD5 mismatch.
I don't think we should be spending time thinking of ways to adhere to SCO's ridiculous claims. If they say the code is already published in the Linux kernel, how can any further damage happen if they publish it again? This is beyond ridiculous.
This is about denying Linux and BSD compatibility to their file system. NTFS is being reverse engineered as we speak, write access is being worked on, so obviously Microsoft needs something new to get rid of them pesky Linux people.
All the new and great features could be done with a (simple?) layer between NTFS and the application. There is no reason why Microsoft needs to invent a new file system here.
Every culture needs myths, fairy tails and mysteries. America needs it's own Loch Ness.
And it's good for the government, too! These kooks wasting their time on area 51 won't have any time left to poke around in current conspiracies. And the more really outlandish conspiracy theories are out there, the less likely the outlandish conspiracies that actually happen are, so nobody will believe them.
If we really had obtained alien technology, why would we still pollute our ecosystem to get from A to B? You'd think we wouldn't need to burn fossils any more for transportation.
I find it hilarious how every now and then some vaguely important sounding organization creeps out of the shadows, proclaims something completely unfounded regarding a topic that is undergoing heated discussion, yet nobody has ever heard of them before.
This happens all the time with those "think tanks", then all those bogus agencies selling certification in the security industry...
Is this a new art form or something? Hacking the stupid mainstream media who will just present it as fact if it comes from an important sounding institution, even if there is not much more about it than a letter box in Aruba?
Is that kind of slave labor contract even legal in the US? That would be quite shameful. How can they claim any rights to software he develops at home in his spare time?!
I did not say America is at fault for China's human rights behaviour, I said fear of US superiority is the reason for their space program. Or do you really want to argue here that wants to go to space out of pure villainy?;-)
Besides, US citizen should be really quiet about human rights after what they let their government do e.g. at Guantanamo Bay. Not to mention recent laws like PATRIOT.
Has has a perfectly fine other web page, and he certainly has enough money to buy a few hundred other domains and servers.
I don't get his complaining about freedom and oppression and stuff. Why can't he just publish his software on some other web page, maybe even under a pseudonym or something.
Slashdot Mirror
Yeah, it works so well that the glorious US army is attacking one weak country on the verge of self implosion after the other instead of going after someone who can defend himself, like, for example, uh, North Korea?
I suggest you read this Wired article to see how well it works. They are using Microsoft Chat on the battlefield, for crying out loud! Yes, the one with the comic characters, where the staff seargant looks like a big breasted bimbo. Nothing inspired a feeling of technical superiority like that, if you ask me.
At least the important systems run Linux.
First he takes over maintenance of BIND, then proceeds to blackmail his "customers" into paying him to be told about the copious amount of security problems, essentially capitalizing on the work of others.
Then he had BIND 9 written, which has nothing to do with BIND. We call it BIND then? Again, to capitalize on the work of others (not that the BIND name is something to carry proudly).
Now he patents what Linux could do at least since 2.0 (1996) with a patched squid 1 (been there, done that). Granted, his patent appears to be earlier than Akamai (1998) but Digital Island may have been early enough to invalidate the patent. They were bought by Cable and Wireless a while ago.
On the research level gcc is not as bleeding edge as other compilers. So if you run example code that shows the merits of a particular optimization, gcc may look not so good. But in practice, it's quite good.
My experiences with UltraSPARC are also a few years old, but gcc was faster and produced better code than Sun CC back then. You have to make sure to set -march=ultrasparc, of course. And I'm not sure about UltraSPARC but normally gcc -O4 does not do more than -O3, which basically is -O2 with function inlining. You can also get some boost with profile based optimization with gcc.
In summary, gcc produces very good code, but you might have to use some little known options for it. For example, gcc on Athlon XP and Pentium >= 3 may gain significant floating point performance with -mfpmath=sse,387 (I got >10% speed-up on lame, gcc's code was even faster than icc's with vectorizer). Another option worth knowing is -malign-double and the regparm attribute.
Another thing you have to keep in mind is: recent optimization advances normally are not big breakthroughs but small incremental advances. Many of them only help in a handful of special cases. gcc 3 has many more optimizations than gcc 2.95.3 and they were so proud of it that they said "much faster code on x86", and then there was whining and gnashing of teeth when most software was unaffected or even slower.
The only platform where I really would prefer the vendor cc is HP-UX on PA-RISC. The HP CC consistently produced 10-30% faster code than gcc (although that may have changed, I haven't used gcc > 2.7 on HP-UX).
It is quite difficult to produce better code than gcc, and my tests on powerpc (granted, those were a few years back using xlc on RS6000 with AIX 4) showed that xlc produced code of about the same quality -- sometimes worse, sometimes better.
The gcc "Haifa" scheduler was donated by IBM Haifa, by the way, so I think it's not surprising that gcc produces good code on powerpc.
On Intel it's quite the same, except that gcc does not vectorize code. From what I have seen, however, icc's vectorizer is not very useful either. I recently tested ogg-vorbis (which is a plain C floating point intensive benchmark) with icc 7 and gcc 3.3 and the gcc version was actually faster than the icc version (on my Athlon XP, target CPU pentium3) despite icc having vectorized several loops.
So all this "vendor-optimized C compiler" stuff is really besides the point. No C compiler will ever be able to match the quality of hand optimized assembler code, and the most important code (ffmpeg MPEG-2 decoder and MPEG-4 codec) has already been hand-optimized. You might be able to squeeze anoter 5 percent out of your code by using a vendor C compiler with insane optimizer settings, but what good is that if the end user is only going to use gcc anyway. I know I am, so I find the numbers for gcc actually more useful for comparison purposes than some vendor C compiler comparison.
Also, we don't want to encourage vendors to produce super vendor optimizing compilers, we want them to optimize gcc (so that everyone benefits, not just their users). So the more benchmarks are done using gcc, the better!
When I said "use it commercially" I meant "make money selling it under a commercial license". To make money with it, they need to modify it ("add value") and the GPL forces them to make their diffs available under GPL as well, which basically means someone will send the diffs to me and I may incorporate them in my version, making their added value available to everyone.
So the fact remains, if someone wants to make money selling my software (and I'm not talking about Red Hat or other distributors of my software here), he needs to talk to me.
Also, I wonder what you mean by "poor". I have a nice little family, can pay my bills, and get paid doing what I like to do -- what more could I possibly want from life? I don't have to be a millionaire. To me it's more about what remains when you die, and when I die, I will leave some offspring and some (fine?) free software behind. What (besides some mediocre slashdot trolls) are you planning to leave behind for future generations?
Ooooh, a bridge troll! ;)
No, I'm not giving you my money.
No wonder he didn't get a job.
;)
The point about GPL is that you can't get ripped off. If they rip you off, you can force them to release their derivative work also as GPL. If he chose the wrong license, he got what he deserved.
I put my embedded work under GPL and actually managed to get some funding. If it's GPL, people have to talk to you to use it commercially, you know? That's the beauty of GPL.
Anyway, I can't say I found LRP to be as great as this guy actually thinks it is. And this childish "look what you missed" bullshit is not going to get him anywhere either. The world is full of companies who are not making any money, Caldera and Lineo being two very good examples he cites himself. Don't expect them to pay you if they don't have to.
So far, almost every company that hired has tried to rip me off in the end. That's how it goes. So choose wisely, chose GPL.
BTW: A new init system? Got one of those as well... I even wrote my own libc. And you know what? People are helping with the projects, in fact, many people are helping me with the projects. Feel free to look at all the names in the dietlibc CHANGES file! I think it's how you treat people that makes them help you. If your code is readable and you treat people well, they will help. You won't get big front page articles on Wired, but you'll create a damn good project, people will know your name. And you will get invited, too! Meet me at Linuxtag 2003!
This is already law for newspapers, and why would internet sites be held to a lesser standard?
And what is the alternative? Facing countless lawsuits? I think it would be less easy to sue someone if he already had to publish your clarification.
And it doesn't say you would have to delete your original or that you can't make sure everyone understands you were forced by law to publish the "clarification" and you still stand by your original report.
Many GNU tools violate POSIX in some way, for example! To find examples, grep the GNU documentation for POSIXLY_CORRECT. One example is GNU du which will display the size in kilobytes instead of blocks (1 block = 512 bytes), which is a much more useful display.
;-)
Also, many open standards stink to the high heavens.
And how do you differentiate between willful and accidental violation of a standard? Yeah, I know, you don't care as long as Microsoft has to pay
The MD5 sums will only match if the lines are identicaly.
Change 8 spaces to a tab? MD5 mismatch.
Change indentation? MD5 mismatch.
Add a comment? MD5 mismatch.
Rename a variable from i to counter? MD5 mismatch.
Rename a function from calculatechecksum to CalculateChecksum? MD5 mismatch.
I don't think we should be spending time thinking of ways to adhere to SCO's ridiculous claims. If they say the code is already published in the Linux kernel, how can any further damage happen if they publish it again? This is beyond ridiculous.
Let them sue until they die of old age!
And this time I don't even care how much money is sunk into the greedy hands of their respective lawyers.
This is about denying Linux and BSD compatibility to their file system. NTFS is being reverse engineered as we speak, write access is being worked on, so obviously Microsoft needs something new to get rid of them pesky Linux people.
All the new and great features could be done with a (simple?) layer between NTFS and the application. There is no reason why Microsoft needs to invent a new file system here.
Do you actually believe Microsoft only spies on you if it's written in the EULA?
Does the EULA say that the Internet Explorer reports all web domains to the MSN search engine if it can't resolve them?
Oh, so you can turn it off alright. Does that change anything?
People don't trust Microsoft, and for good reasons.
And what is the U.N. supposed to do? Kick out SCO's ambassador?
When in doubt, ask NATO. They bomb first, ask later.
What do you mean, no harm to the hardware?
What mutilation to your hardware could possibly be worse than being used to run SCO?!
If the hardware is still alive, it certainly won't feel any more pain and deserves merciful, quick and painless death, at least a lengthy quarantine.
Every culture needs myths, fairy tails and mysteries. America needs it's own Loch Ness.
And it's good for the government, too! These kooks wasting their time on area 51 won't have any time left to poke around in current conspiracies. And the more really outlandish conspiracy theories are out there, the less likely the outlandish conspiracies that actually happen are, so nobody will believe them.
If we really had obtained alien technology, why would we still pollute our ecosystem to get from A to B? You'd think we wouldn't need to burn fossils any more for transportation.
Don't they normally embrace and extend?
No, I do not want to know what those pervs will extend to the kids!
Apple does not run on BSD, it runs on a microkernel like Nextstep.
oh, and Apple is most definitely not compliant to the Single Unix Specification as published by the open group. MacOS X does not even have poll(2)!
I find it hilarious how every now and then some vaguely important sounding organization creeps out of the shadows, proclaims something completely unfounded regarding a topic that is undergoing heated discussion, yet nobody has ever heard of them before.
This happens all the time with those "think tanks", then all those bogus agencies selling certification in the security industry...
Is this a new art form or something? Hacking the stupid mainstream media who will just present it as fact if it comes from an important sounding institution, even if there is not much more about it than a letter box in Aruba?
Sheesh.
Can't wait for the next season to air...
You are new here, right?
We always blame Microsoft here.
Have a nice day.
Is that kind of slave labor contract even legal in the US? That would be quite shameful. How can they claim any rights to software he develops at home in his spare time?!
I did not say America is at fault for China's human rights behaviour, I said fear of US superiority is the reason for their space program. Or do you really want to argue here that wants to go to space out of pure villainy? ;-)
Besides, US citizen should be really quiet about human rights after what they let their government do e.g. at Guantanamo Bay. Not to mention recent laws like PATRIOT.
Has has a perfectly fine other web page, and he certainly has enough money to buy a few hundred other domains and servers.
I don't get his complaining about freedom and oppression and stuff. Why can't he just publish his software on some other web page, maybe even under a pseudonym or something.