Right, so based on that chart, one might well find code from 4.3BSD in both Linux and SCO, which was in the SCO codebase first but is nonetheless ok (because Linux got it from BSD).
So my question is, can we do a diff for matches between 4.3BSD and Linux 2.4, or do people know offhand what might still be there?
Is the SHA1 actually coded in python, or in C libraries used by the python interpreter? I wouldn't be surprised if it's the latter, in which case recoding the rest isn't much help.
Oh, please. The network and algorythms will be the limit on this kind of application; python will be fine if you're not trying to run it on a 286 with a T3.
Seriously, pull up "top" or something and tell me if bittorrent actually uses nontrivial CPU. I could be wrong, but I'd be very surprised.
The British bombed German dams during World War II...
Could a conventional explosive (truck bomb? boat bomb?) damage this dam, or would it take nonconventional weapons, in which case we're probably all screwed anyway? (Actually, I can't think offhand of a sccenario where somebody is bombing central China where things haven't gone to heck anyway, but...)
Right. All the radiation will be absorbed by your stomach/intestinal lining, which is essentially disposable anyway, and is replaced every few weeks, so cancer there isn't really an issue unless you get such a huge dose of radiation that it eats away your stomach lining.
Lungs are quite different; get measureable plutonium dust in there and lung cancer is pretty much a sure thing.
Caffene, by contrast, is quite simple; too much at once and you get a heart attack.
Well, that Linux hippie chick is the lost bastard child of UNIX, recently dumped SCO, who is the stepson of Novell, and IBM is currently trying to get Linux to go steady, but tolerates her sleeping around...
Yes, wrong. Veritas started as a vendor of fault-tolerant UNIX boxen, and eventually converted to a software-only strategy when it became clear that building hardware was not a value-add.
No DRM, off-site backup of my data, and worked fine with Galeon on Debian Linux. If you aren't jumpy about your tax info being shipped off to intuit.com, I can definately recommend it.
I was puzzled by this at first, but it's actually not surprising. Once you get into a court of law, any damn thing can happen; some fool judge might even give SCO a chance. Right here and now, though, SCO is hemmoraging money by the day, so IBM will sit, and grin, and stall, and stall, for the year or so it takes SCO to crash and burn.
Heh. Did somebody at Caldera^H^HSCO finally notice that they were violating the GPL by shipping Linux while claiming property rights against it?
The funny thing is, they've therefore stolen all the non-infringing code in the kernel, as it's from other people and they can only redistribute it by releasing their own.
(Assuming, of course, that there is any actual infringemnet, which seems unlikely.)
Expect increasingly shrill announcements as IBM blackens the sky with lawyers and SCO tries to give Linux a black eye to force IBM to buy them out before the case is thrown out of court.
The STMP protocol should be extended; the receiver can require the sender to factor a large prime number before the message will be accepted. A few seconds CPU time per legitimate message is no biggie, but...
Even if the test is fair, who the heck cares? I mean, have you looked at the test results?
A RHAS 2.1 box with 4 CPUs and 4 (!) gigabit ethernet cards can drive the cards to >50% utilization with SAMBA. NT2K3 can get close to 100%. But who the $#%#$ has or needs SMB servers with multiple gigabit cards?
And, of course, this is a beta NT version running filesharing in kernel versus a 2.4 Linux kernel running filesharing in userspace.
Slashdot Mirror
Birthday paradox or no, 128 bits is a darn big number. You're basically looking at
Product (2^128-x)/(2^128) for X over [1..N].
I don't trust my intuition on how big that is as N increases; what are the actual numbers?
Right, so based on that chart, one might well find code from 4.3BSD in both Linux and SCO, which was in the SCO codebase first but is nonetheless ok (because Linux got it from BSD).
So my question is, can we do a diff for matches between 4.3BSD and Linux 2.4, or do people know offhand what might still be there?
Hrm... even if she's right and it's not some strange conincidence, is there old BSD code in Linux? That should be checkable.
That should be free and clear copyright-wise, but System 5 could well have the same BSD code (quite possibly orignally stolen from BSD).
Because even SCO realizes it would all be rewritten in about 1 week if it were revealed and even marginally questionable?
Is the SHA1 actually coded in python, or in C libraries used by the python interpreter? I wouldn't be surprised if it's the latter, in which case recoding the rest isn't much help.
Oh, please. The network and algorythms will be the limit on this kind of application; python will be fine if you're not trying to run it on a 286 with a T3.
Seriously, pull up "top" or something and tell me if bittorrent actually uses nontrivial CPU. I could be wrong, but I'd be very surprised.
The British bombed German dams during World War II...
Could a conventional explosive (truck bomb? boat bomb?) damage this dam, or would it take nonconventional weapons, in which case we're probably all screwed anyway? (Actually, I can't think offhand of a sccenario where somebody is bombing central China where things haven't gone to heck anyway, but...)
Right. All the radiation will be absorbed by your stomach/intestinal lining, which is essentially disposable anyway, and is replaced every few weeks, so cancer there isn't really an issue unless you get such a huge dose of radiation that it eats away your stomach lining.
Lungs are quite different; get measureable plutonium dust in there and lung cancer is pretty much a sure thing.
Caffene, by contrast, is quite simple; too much at once and you get a heart attack.
The community needs to come up with a way to respond to this incident, and to other things like it.
Well, there was the build-your-own-cruise-missle story on slashdot last week...
Well, that Linux hippie chick is the lost bastard child of UNIX, recently dumped SCO, who is the stepson of Novell, and IBM is currently trying to get Linux to go steady, but tolerates her sleeping around...
Have the executives been dumping while this plays out during the recent spike? Could be fun for the SEC.
Of course, they're probably just stupid rather than devious, but you never know...
Well, they did buy a pig in a poke. The reason you shouldn't is that the pig may be - as it was here - a cat.
Which they would have know if Novell had Let the cat out of the bag...
Linus, of course, is cleverly hedging his bets here. He knows there's no chance that a Slashdot editor will catch a duplicate!
Executives pay selves huge severance packages as company goes under.
Isn't it a felony to lie in congressional testimony?
Yes, wrong. Veritas started as a vendor of fault-tolerant UNIX boxen, and eventually converted to a software-only strategy when it became clear that building hardware was not a value-add.
You want to run a bunch of AIs and a simulation of the world, so you run it on a Beowulf Cluster of human brains.
This also partially explains Neo's powers; because part of the simulation is running in unused areas of his brain, he is able to alter it.
I understand an early draft of the script actually included some of this, but it was 'simplified' away.
Worked fine for me.
No DRM, off-site backup of my data, and worked fine with Galeon on Debian Linux. If you aren't jumpy about your tax info being shipped off to intuit.com, I can definately recommend it.
The ticker is SCOX.
Last quarter they lost $725k on sales of $13.5M. They have about $5M cash, 340 employees, and a total stock value of about $40M.
Their revenues dropped 25% last quarter; if that continues they have only a year or two to live.
I was puzzled by this at first, but it's actually not surprising. Once you get into a court of law, any damn thing can happen; some fool judge might even give SCO a chance. Right here and now, though, SCO is hemmoraging money by the day, so IBM will sit, and grin, and stall, and stall, for the year or so it takes SCO to crash and burn.
Heh. Did somebody at Caldera^H^HSCO finally notice that they were violating the GPL by shipping Linux while claiming property rights against it?
The funny thing is, they've therefore stolen all the non-infringing code in the kernel, as it's from other people and they can only redistribute it by releasing their own.
(Assuming, of course, that there is any actual infringemnet, which seems unlikely.)
Expect increasingly shrill announcements as IBM blackens the sky with lawyers and SCO tries to give Linux a black eye to force IBM to buy them out before the case is thrown out of court.
Waiting doesn't help much; the spammer can just multiplex with multiple connections (maybe you can counter this, but it gets to be a real PITA).
Conversion doesn't have to be all-or-nothing; mail for which compute tax wasn't paid can just be flagged with an X-Might-Be-Spam header.
The STMP protocol should be extended; the receiver can require the sender to factor a large prime number before the message will be accepted. A few seconds CPU time per legitimate message is no biggie, but...
Even if the test is fair, who the heck cares? I mean, have you looked at the test results?
A RHAS 2.1 box with 4 CPUs and 4 (!) gigabit ethernet cards can drive the cards to >50% utilization with SAMBA. NT2K3 can get close to 100%. But who the $#%#$ has or needs SMB servers with multiple gigabit cards?
And, of course, this is a beta NT version running filesharing in kernel versus a 2.4 Linux kernel running filesharing in userspace.
And then he could demand a ransom of... one million dollars.