Sorry for shouting but there are so many messages here chastizing FSF for not backing up. They were backing up but they were compromised in March '03 and they can't trust backups that were made since the intrusion.
The FSF site was cracked back in March so it is likely that a lot of the FSF software in the Gentoo system is also compromised. You'd have much better luck with Debian because its stable tree is much less up-to-date then Gentoo's.
IMO this move by SCO was to intentially anger the Linux community. They might get a few measly dollars from a few fools but for the most part they are trying to get more publicity and it seems to be working.
It is really a wonderful meller-drama. In this world filled with ambiguity, here we have a story where good and evil are well defined and unambiguous. SCO is playing up their role to the hilt twirling their mustacios. We are all falling for it but in return we get to be on a great emotional rollercoaster ride.
When SCO lose to IBM they (SCO) will have no money and no assets left. Anyone who pays money to SCO (to buy a license or to buy stock) will never get any of it back after the suit.
The theory is that SCO can't complain about somebody distributing these hashes because you can't get the source code from it.
I agree with you that the md5 scheme is clever but the problem is that you can get the SCO source code from it since the lines in the Linux files (which everyone has access to) are identical to the lines in the SCO files.
To take an extreme case, what if I did an md5 sum, not on every block of 5 lines but instead on every character? If I compared the SCO source with a file that contains all the ascii characters then the hash results would disclose the entire SCO source.
Don't get me wrong. SCO totally sucks for doing what they are doing. I believe that McBride has already made reservations for himself in one of the hottest corners of hell. I also think that the md5 scheme is a brilliant technique for finding overlapping code. I just disagree with the idea that the md5 sum scheme is a loophole that would let us legally see the "copied" lines.
I hope there is a way for us all to see the "copied" lines ASAP. I just don't think that this is it. Please don't mod me down just because you dislike my conclusion.
What I don't understand about the SCO/IBM case, is why IBM isn't taking action to immediately stop SCO from doing what they are doing.
The burden of proof is on whoever launches the law suit. It will be much easier for IBM to win if they wait for the SCO suit. That way SCO will have to prove their case. It would be much harder for IBM to prove SCO is wrong before they (IBM) get a chance to see SCO's "proof".
Slashdot Mirror
Sorry for shouting but there are so many messages here chastizing FSF for not backing up. They were backing up but they were compromised in March '03 and they can't trust backups that were made since the intrusion.
The FSF site was cracked back in March so it is likely that a lot of the FSF software in the Gentoo system is also compromised. You'd have much better luck with Debian because its stable tree is much less up-to-date then Gentoo's.
It is really a wonderful meller-drama. In this world filled with ambiguity, here we have a story where good and evil are well defined and unambiguous. SCO is playing up their role to the hilt twirling their mustacios. We are all falling for it but in return we get to be on a great emotional rollercoaster ride.
When SCO lose to IBM they (SCO) will have no money and no assets left. Anyone who pays money to SCO (to buy a license or to buy stock) will never get any of it back after the suit.
The theory is that SCO can't complain about somebody distributing these hashes because you can't get the source code from it.
I agree with you that the md5 scheme is clever but the problem is that you can get the SCO source code from it since the lines in the Linux files (which everyone has access to) are identical to the lines in the SCO files.
To take an extreme case, what if I did an md5 sum, not on every block of 5 lines but instead on every character? If I compared the SCO source with a file that contains all the ascii characters then the hash results would disclose the entire SCO source.
Don't get me wrong. SCO totally sucks for doing what they are doing. I believe that McBride has already made reservations for himself in one of the hottest corners of hell. I also think that the md5 scheme is a brilliant technique for finding overlapping code. I just disagree with the idea that the md5 sum scheme is a loophole that would let us legally see the "copied" lines.
I hope there is a way for us all to see the "copied" lines ASAP. I just don't think that this is it. Please don't mod me down just because you dislike my conclusion.
What I don't understand about the SCO/IBM case, is why IBM isn't taking action to immediately stop SCO from doing what they are doing.
The burden of proof is on whoever launches the law suit. It will be much easier for IBM to win if they wait for the SCO suit. That way SCO will have to prove their case. It would be much harder for IBM to prove SCO is wrong before they (IBM) get a chance to see SCO's "proof".