Slashdot Mirror
SCO Announces Final Termination of IBM's Licence
ickle_matt writes "SCO have announced the final termination of IBM's UNIX license, despite Novell telling them they can't. Interestingly enough there's a new set of "stolen code" figures in the release - 'approximately 148 files of direct Sequent UNIX code to the Linux 2.4 and 2.5 kernels, containing 168,276 lines of code. This Sequent code is critical NUMA and RCU multi-processor code previously lacking in Linux. Sequent-IBM has also contributed significant UNIX-based development methods to Linux in addition to the direct lines of code specified above.' "
Well, one wonders if SCO trusts any of the operating systems it sells, or has sold, in recent times. Take a look at the http response string for the web server used in this announcement:
$ HEAD http://ir.sco.com
200 OK
Connection: close
Date: Wed, 13 Aug 2003 14:42:15 GMT
Server: Microsoft-IIS/5.0
Content-Type: text/html
Client-Date: Wed, 13 Aug 2003 14:42:16 GMT
Client-Response-Num: 1
Page-Completion-Status: Normal
Page-Completion-Status: Normal
"...was terminated for improper transfer of Sequent's UNIX source code and development methods into Linux." (emphasis mine)
You heard it here first, people. SCO owns the UNIX development methods too. That means that producers of just about any software (because who hasn't been influenced by UNIX development methods?) will have to pay off SCO. What a bunch of bull.
"Customers may not acquire a license in Dynix/ptx from today's date forward."
In a departure from their standard MO, SCO doesn't threaten IBM customers who currently hold Dynix/ptx licenses, and instead just claims that no new licenses may be issued.
Maybe they are starting to worry about gettting nailed on extortion charges?
Oh, now the line is that there is "Sequent-UNIX" code in the kernel, with no mention of Sys V code?
So, what we're really talking about is IBM code. No more BS weasle-words, they're talking about code they NEVER owned in the first place.
Well, that tears it: SCO's suing IBM for contributing their own code to the kernel. Yeah, that's totally gonna hold up in court.
So, if the "stolen" code is in RCU and NUMA, why are they asking money from single processor systems? AFAIK this RCU and NUMA stuff is all SMP related, and not even compiled in single processor systems.
IBM owns those patents. You can't say BTW since we license something to you, we now own your patents. IF, big IF, that were the case, IBM would have made them sign a cross-licensing agreement allowing IBM to keep control of thier patents.
IBM has owned those patents since '92 when they bough sequent, sequent had those patents since the late 80's so either way SCO SOL.
So Long and Thanks for all the Fish.
IBM's System V license seems to state that when they add code to their SysV (==AIX) that code has to be treated as the rest of the SCO owned SysV code.
However, IBM (and Sequent before them) has made very sure that they wrote up a general outline of how the process works, and then made an implementation on AIX (and Dynix/ptx). They then made a very similar implementation on Linux. As long as they can show that both implementations came from the general outline, there should be no problem.
Heck, they should be able to argue that they can copy their stuff from AIX to Unix as long as there is a general outline. They may be able to argue that as long as they wrote it, and nothing of Unix comes out to the public, that they're in the clear.
"Giving money and power to government is like giving whiskey and car keys to teenage boys" P. J. O'Rourke
This is INCORRECT. As I have written in a paper available here, IBM's contract with AT&T (and now SCO) explicitly states that code developed by IBM is the property of IBM and is NOT a derivative workof AT&T (now SCO).
:)
HOWEVER, the contract between Sequent and AT&T has NOT been made public by SCO nor IBM and so it is not clear to third parties whether or not the Sequent-AT&T agreements give Sequent the same rights as were given to IBM, though one may speculate based on the format of the IBM-AT&T agreements (see my article for full details).
Fundamentally, this is a contract law case in which the status of the Sequent-AT&T and IBM-AT&T agreements are defined. In other words, when IBM bought Sequent, what happened to the Sequent-AT&T agreement? Were it made null and all dealings between Sequent and AT&T now under the stipulations of the IBM-AT&T agreement? Or is it the case that code developed by Sequent is still bound by the original Sequent-AT&T contract?
Someone more familiar with contract law will have to respond; however, I believe that if in fact ANY code developed by Sequent BEFORE it was bought by IBM AND that code was placed in a SysV derivative (in this case probably Dynix/ptx) AND that code was placed in the Linux kernel, then it may very well be the case that SCO is standing of firm legal ground on that issue.
There are, of course, other details to this situation which may invalidate SCO's claims to right of the code, such as the GPL, and unclean hands doctrine (both arguments which IBM has included in its rebuttal) but I suppose we'll have to wait and see what happens. Also, read my article.
OK, maybe it is time to put my tinfoil hat on and start looking out the window for black helicopters, but what if this is all a nod-and-wink kind of scheme.
IBM and SCO make it known that IBM is thinking of buying SCO. Instead, SCO sues IBM. SCO's stock price goes up. SCO's owners dump stock (getting rich) and SCO uses the inflated stock price to buy up small companies. Once SCO has all the pieces in place, IBM suddenly forces things into court and SCO is blown away. SCO stock plummets. *THEN* IBM buys them for a basement bargin price.
This does multiple things for IBM. (1) Publicity. (2) Good PR with the Linux crowd as IBM "saves the day". (3) They get more stuff in the SCO buyout. (4) It places IBM's competitors in the Linux arena on shaking ground while all this is happeneing, possibly forcing some out of business.
Suddenly, in the end, IBM has a Linux distro, AIX for the high-end, lots of other IP, and a rep as being the big champion for Linux that is willing to put its money where its mouth is.
Right about that time, IBM would be ready to take on Microsoft again for the servers and desktops. IBM gives away the desktop OS just to have an in and starts recapturing the mid-level servers. As IBM becomes the "leader" of Linux, why would you not go with IBM and AIX on the big iron since IBM makes it one smooth continuum of *nix?
Man, I have got to stop drinking 10 diet cherry cokes while on medicine for a head cold before I post. =)
What constitutes a derived work is going to be the real copyright issue here. Outside the software arena, the notion of a derived work has been very broadly interpreted. Movies based on novels have been held to be derived works even when the connection between the two included little more than the title and the name of the lead character.
IBM has retained Cravath, Swaine, and Moore, probably the strongest litigation firm for difficult cases in the world. The Cravath approach on big cases is to put an army of lawyers on the problem and litigate everything to death. We'll probably see a detail-oriented litigation, rather than one based on broad principles. Following a Cravath lawsuit tends to be a mind-numbing experience for all concerned.