Hi John. I think we've both responded to the AC above but it's ended up looking like we're talking to each other or some such thing. I can't quite decipher the thread of this conversation. However, if the question you're referring to is the one about i686 binaries, this page has some interesting studies. He's mostly concerned with tuning Athlon XP binaries via --fxxx type flags, but begins with tests of march=foo. That parameter doesn't pay off until i686 and gets another nice bump at march=athlon-xp. The skinny from his benchmark in that case yields a 9% improvement over plain ol' i386.
He got some pretty significant gains by tweaking the more detailed flags beyond march, but the code in question is floating point computational stuff. That's good to know since it means tweaking a renderer or whatnot will be worth the effort, but it says nothing about yer meat and potatoes binaries.
Incidentally, an anonymous poster pointed out that Debian has an i686 libc in unstable now. I'll install it the next time I do an upgrade, but I ain't gonna bother recompiling 850 packages.
The reason debian wont be optimised is because the gatekeepers on debian-devel argue that optimisation dont make programs run faster (seriously).
That's patently false. They just don't think the speed (around 9%) is worth the effort.
I think the real reason is pride, they are afraid of lossing face and admitting they were wrong.
LOL! You've got to be kidding. This distribution is regularly the butt of/. jokes that run along the lines of, "Hey, so I hear Debian is about drop a.out and maybe even make the jump to libc6." If they wanted to invest some ego in a public face there are other things that would play second fiddle to technical matters before they got around to "looking cool" in front of the "133t".
Admitting they were wrong would make it harder to start arguments in the future.
Uhh... Huh?
They argue that debian packages are optimised, the kernel for example has multiple packages each optimised for a different cpu.
Well, that is the most important optimization to make.
The minimum that needs to be done is to modify policy to require packages that can be optimised to have support for end users compiling optimised for themself.
Why bother when you can just install a Debian package that adds that functionality?
You know if you want you could just install "pentium-builder". Barring that it would take the addition of one line of shell code to the rules file to add any CFLAGS you like.
I'm not a guru, but I am pretty sure that i586 optimized code runs slower (or even poorly) on anything that isn't a 586. The whole foo-i586.rpm thing was a kind of marketing gimmick of Mandrake's back when they were packaging pgc and the K6-3 was considered a fast processor. The notion of "optimizing" builds like that has been brought up on the Debian mailing lists before but dismissed as being not worth the effort. After all, Debian has a huge amount of software to keep in the repository. If one starts subdividing x86 where does one stop? Should one offer libpng3 compiled for i386, i486, i586, i686, K7, and P4? Pragmatically, it isn't worth it. They might not even package 64-bit binaries for quite some time.
The biggest optimizations to be had are already packaged in the kernels to begin with, though there's probably something to be said for tuning up glibc too (IIRC Redhat does this).
I put it to use recently as well and also give it props. It's still text based but more streamlined. If you just want to get to a boot off the disk and a login prompt you don't have to waste a lot of time. IIRC, the design allows for graphical front ends too for those who prefer them, but why miss out on that ncurses TAB TAB ENTER speed and rhythmn? It's just like playing Defender...
Well, I hope that smiley means you took my post as humorously as I meant it. The stuff I highlighted has a pretty different tone from the usual, "I know blah, blah, blah..." we like to throw about around here.
I don't know what I mean. I'm confused too... I honestly don't know what I'm talking about... Like I said, I honestly don't know what I'm talking about... I haven't the foggiest idea why it never occurred to me.
I don't think it would kill you to note that some people here -- in a thread that was supposed to be about Gentoo Linux -- have offered unsolicited assistance with your camera. You've received one negative response from an anonymous smart ass who never got modded out of oblivion. Like others, I immediately wanted to pitch in with pointers as I just learned how to link my new camera with my computer but saw that gphoto2 has already been mentioned.
Incidentally, there is no fragmentation in this case. libgphoto2 is the defacto resource for connecting *nix to PTP cameras.
That looks really interesting. I take it the point is to let dd run until "somefile" takes up all remaining space on the partition and dd exits with a write error?
tar doesn't concern itself with compression. It transforms many files to one file and back again. Since the original example using bz2 was dealing with a stream read right off of hda adding tar to the mix contributes nothing. That stream is just one big file to anything on the receiving end of the pipe.
That policy isn't about "playing stupid politics". It has to do with restricting the interaction between closed binaries and kernel innards. If Microsoft released a closed binary kernel module for some piece of hardware they sell would you want it to have unrestricted access to kernel innards? If it was unreliable would you want the kernel maintainers to waste their time trying to make it work?
Stop waiting for the installer and just get it right now. The business card cd is only 30 megs and IIRC you can skip testing and go straight to unstable if you like.
Every time the government grants a copy right or patent to someone, it denies everyone else their right to do the same thing.
Well we're into a broader subject now. This article was about a crackdown on software pirates. Then there was that half rambling, repetitive, "14 point" rant that confounded software piracy, mp3 file swapping, and a percieved attitude amongst fans of CopyLeft software that is probably exhibited by a small minority at best.
It was the harping against p2p (ahem) liberation of mp3 files that I noted the most in his comment. That is the subject I presumed you were addressing.
I don't think we can honestly lump copyright and patent issues together as being two heads of the same evil hydra. "Blowin' in the Wind" isn't patented. That's the province of some groovy and cheap way to make ammonia. Either way I don't see how you could extrapolate an actual infringement of rights. Voting, ownership, running for office, being able to publish your opinions, avoid self-incrimination, etc... those are the kinds of things that are rights. The effects of patents are a strange beast, but copyrights seem pretty straightforward. I really don't think anyone has been prevented from writting "Visions of Johanna" just because Dylan got a copyright on it. Nor do I see how the GPL could work without copyright.
...Oligopoly is more like it.
I have no truck whatsoever with that characterization.
Q:If they do let up on the downward pressure they are apparently exerting on your life will you respond in kind?
A:My attitude would be much different.
I had the whole p2p/music thing in mind here. What I couldn't get around when formulating a response the first time was the way I don't think anything could be different in principal than the way it was before Napster and p2p applications threw the recording industry for a loop. Acknowledging the rights of the copyright holder(s) can't mean anything less than keeping a ripped track off of a p2p network unless they say they don't mind. If we can't repect that then we have no right to expect reciprocal treatment of anything GPL'd.
The big record companies and our own legislature have pirated our rights to free speech.
Huh? In what way? What are you talking about?
I'll concede that it's worth compromising free speech...
Personally, I'm not sure anything is worth compromising free speech. I can understand the laws in Germany that prohibit Nazi paraphenelia or propaganda, but even in that case my mind kinda hits a wall. The cultural need for such regulation seems apparent enough, but it can't be done without cramping something vital. I think freedom of speech as a civic principal has a special relation to the other tenents of democracy. It is as Samuel Johnson (IIRC) said of courage: it's the one virtue that facilitates all the others.
...by granting exclusive copy rights to writers/performers so that there will be an incentive for people to create.
I'm not grasping an unabiguous logic here. In what way does noting that Bob Dylan wrote "Just Like a Woman" infringe upon the free speech of everyone who isn't Bob Dylan? What does any of that have to do with putting mp3s on a p2p network? Surely you aren't claiming that you're politically oppressed by David Geffen and the like are you?
But those rights should only be short term. The founding fathers stated something like 14 years with a one-time 14 year extension.
That sounds alright. But why is this a topic of discussion? Are the mp3s on Gnutella a product of civil disobediance?
Things happen *much* faster now, so those terms should be shorter, not longer.
How do we measure our current "speed" relative to that of the 18th Century so as to come up with a theoretically proper expiration date on copyrights?
When they stop infringing my rights, I'll start caring about theirs.
What rights of yours are "they" infringing? If you sincerely believe yourself to be engaged in a struggle for "rights" then how will "not caring" about their rights ever advance your cause? If they do let up on the downward pressure they are apparently exerting on your life will you respond in kind? If so, what rights of theirs will you begin to care about?
This debate has been hashed and rehashed on/. for years now without any forward movement towards a fresh, positive resolution. That may be because there isn't one to be found. With that in mind I would like to suggest that asking the author of a work for permission to trade something on a p2p network seems like a fair guiding principal. It reflects (or at least harmonizes with) the dictates of respect that most of us express and rely upon in our daily lives.
That may be what we know from the articles, but the fact that this spat between partners in crime has become public at all says something that makes any injunction to get-with-the-suing-mother-fuckers a bit beside the point.
Baystar doesn't have their money yet, and if they came out and said "SCO's case has no merit" the stock might tank to the point where they couldn't get it.
With this in mind I have to wonder if Baystar's disposition towards SCO is even more negative than the public story indicates. This is still a serious vote of no confidence that damages Baystar's investment and, of course, they know it. So how badly do they really want to bail out in light of their willingness to let this degree of friction get public?
Slashdot Mirror
He got some pretty significant gains by tweaking the more detailed flags beyond march, but the code in question is floating point computational stuff. That's good to know since it means tweaking a renderer or whatnot will be worth the effort, but it says nothing about yer meat and potatoes binaries.
Incidentally, an anonymous poster pointed out that Debian has an i686 libc in unstable now. I'll install it the next time I do an upgrade, but I ain't gonna bother recompiling 850 packages.
I feel certain that I've heard this injunction before, but now that I'm actually thinking about it I've got to wonder. Why not?
That's patently false. They just don't think the speed (around 9%) is worth the effort.
I think the real reason is pride, they are afraid of lossing face and admitting they were wrong.
LOL! You've got to be kidding. This distribution is regularly the butt of /. jokes that run along the lines of, "Hey, so I hear Debian is about drop a.out and maybe even make the jump to libc6." If they wanted to invest some ego in a public face there are other things that would play second fiddle to technical matters before they got around to "looking cool" in front of the "133t".
Admitting they were wrong would make it harder to start arguments in the future.
Uhh... Huh?
They argue that debian packages are optimised, the kernel for example has multiple packages each optimised for a different cpu.
Well, that is the most important optimization to make.
The minimum that needs to be done is to modify policy to require packages that can be optimised to have support for end users compiling optimised for themself.
Why bother when you can just install a Debian package that adds that functionality?
You know if you want you could just install "pentium-builder". Barring that it would take the addition of one line of shell code to the rules file to add any CFLAGS you like.
The biggest optimizations to be had are already packaged in the kernels to begin with, though there's probably something to be said for tuning up glibc too (IIRC Redhat does this).
I put it to use recently as well and also give it props. It's still text based but more streamlined. If you just want to get to a boot off the disk and a login prompt you don't have to waste a lot of time. IIRC, the design allows for graphical front ends too for those who prefer them, but why miss out on that ncurses TAB TAB ENTER speed and rhythmn? It's just like playing Defender...
This has been on the table for a while. Mozilla has "reasonable" (as they describe it at mozilla.org) support for it.
Well, I hope that smiley means you took my post as humorously as I meant it. The stuff I highlighted has a pretty different tone from the usual, "I know blah, blah, blah..." we like to throw about around here.
'cause he screwed up?
For the record I don't think there's any ground left to cover.
WOW!
I can't believe I just read that on Slashdot!
It's kind of a classic that keeps coming up on gentoo stories. The poster isn't trying to pass it off as his own.
Incidentally, there is no fragmentation in this case. libgphoto2 is the defacto resource for connecting *nix to PTP cameras.
That looks really interesting. I take it the point is to let dd run until "somefile" takes up all remaining space on the partition and dd exits with a write error?
tar doesn't concern itself with compression. It transforms many files to one file and back again. Since the original example using bz2 was dealing with a stream read right off of hda adding tar to the mix contributes nothing. That stream is just one big file to anything on the receiving end of the pipe.
Is that really the case? I thought their response was that they didn't want multiple warnings issued when one should suffice.
That policy isn't about "playing stupid politics". It has to do with restricting the interaction between closed binaries and kernel innards. If Microsoft released a closed binary kernel module for some piece of hardware they sell would you want it to have unrestricted access to kernel innards? If it was unreliable would you want the kernel maintainers to waste their time trying to make it work?
If you only want to give it to "real life friends" then you really shouldn't be concerned about giving them your email address as well.
Stop waiting for the installer and just get it right now. The business card cd is only 30 megs and IIRC you can skip testing and go straight to unstable if you like.
Exactly. Building Debian in adherance to "constitutional" principals ensures it's availability as a reference platform.
Touche.
Well we're into a broader subject now. This article was about a crackdown on software pirates. Then there was that half rambling, repetitive, "14 point" rant that confounded software piracy, mp3 file swapping, and a percieved attitude amongst fans of CopyLeft software that is probably exhibited by a small minority at best.
It was the harping against p2p (ahem) liberation of mp3 files that I noted the most in his comment. That is the subject I presumed you were addressing.
I don't think we can honestly lump copyright and patent issues together as being two heads of the same evil hydra. "Blowin' in the Wind" isn't patented. That's the province of some groovy and cheap way to make ammonia. Either way I don't see how you could extrapolate an actual infringement of rights. Voting, ownership, running for office, being able to publish your opinions, avoid self-incrimination, etc... those are the kinds of things that are rights. The effects of patents are a strange beast, but copyrights seem pretty straightforward. I really don't think anyone has been prevented from writting "Visions of Johanna" just because Dylan got a copyright on it. Nor do I see how the GPL could work without copyright.
I have no truck whatsoever with that characterization.
Q:If they do let up on the downward pressure they are apparently exerting on your life will you respond in kind?
A:My attitude would be much different.
I had the whole p2p/music thing in mind here. What I couldn't get around when formulating a response the first time was the way I don't think anything could be different in principal than the way it was before Napster and p2p applications threw the recording industry for a loop. Acknowledging the rights of the copyright holder(s) can't mean anything less than keeping a ripped track off of a p2p network unless they say they don't mind. If we can't repect that then we have no right to expect reciprocal treatment of anything GPL'd.
Huh? In what way? What are you talking about?
I'll concede that it's worth compromising free speech...
Personally, I'm not sure anything is worth compromising free speech. I can understand the laws in Germany that prohibit Nazi paraphenelia or propaganda, but even in that case my mind kinda hits a wall. The cultural need for such regulation seems apparent enough, but it can't be done without cramping something vital. I think freedom of speech as a civic principal has a special relation to the other tenents of democracy. It is as Samuel Johnson (IIRC) said of courage: it's the one virtue that facilitates all the others.
I'm not grasping an unabiguous logic here. In what way does noting that Bob Dylan wrote "Just Like a Woman" infringe upon the free speech of everyone who isn't Bob Dylan? What does any of that have to do with putting mp3s on a p2p network? Surely you aren't claiming that you're politically oppressed by David Geffen and the like are you?
But those rights should only be short term. The founding fathers stated something like 14 years with a one-time 14 year extension.
That sounds alright. But why is this a topic of discussion? Are the mp3s on Gnutella a product of civil disobediance?
Things happen *much* faster now, so those terms should be shorter, not longer.
How do we measure our current "speed" relative to that of the 18th Century so as to come up with a theoretically proper expiration date on copyrights?
When they stop infringing my rights, I'll start caring about theirs.
What rights of yours are "they" infringing? If you sincerely believe yourself to be engaged in a struggle for "rights" then how will "not caring" about their rights ever advance your cause? If they do let up on the downward pressure they are apparently exerting on your life will you respond in kind? If so, what rights of theirs will you begin to care about?
This debate has been hashed and rehashed on /. for years now without any forward movement towards a fresh, positive resolution. That may be because there isn't one to be found. With that in mind I would like to suggest that asking the author of a work for permission to trade something on a p2p network seems like a fair guiding principal. It reflects (or at least harmonizes with) the dictates of respect that most of us express and rely upon in our daily lives.
That may be what we know from the articles, but the fact that this spat between partners in crime has become public at all says something that makes any injunction to get-with-the-suing-mother-fuckers a bit beside the point.
With this in mind I have to wonder if Baystar's disposition towards SCO is even more negative than the public story indicates. This is still a serious vote of no confidence that damages Baystar's investment and, of course, they know it. So how badly do they really want to bail out in light of their willingness to let this degree of friction get public?