% fortune -ae paul murphy

Suicide by victory: more on SCO

About the middle of last week I got an email with the subject line "SCO - Too Soon to Comment?" that said:

Come on "Paul", you need to step up someday and admit that you have been wrong all along about SCO and their scheme against Linux. You can do it, I know you can!

This led to an interesting exchange - interesting because when I responded I got back an intelligent question instead of the personalized denunciation various groklaw fans have taught me to expect from an exchange on this subject.

Here's my immediate response:

Well I was clearly wrong about the jury verdict - still find that hard to believe - but not, I think about the facts of the case.

One of those facts being that SCO did not have a scheme against Linux - their lawyers tried to license Linux as a way of valuing the claim against IBM, nothing more (or less either since that was a very bad idea.)

Two notes about this:

  1. the judge's ruling essentially just extends the jury's verdict to the ancillary issues: the surprise would have been had he decided not to do this - and his decision to do the safe thing seems, like the jury's, to have been more of a responses to some of the witnesses than to their testimony.

  2. much of the stuff on SCO I wrote for LinuxWorld and CIOToday doesn't seem to be there anymore - but you can see some of it here - and the second essay on that list, from October of 2003, tries to debunk many of the myths surrounding this "attack on Linux" idea.

The response this drew was:

Forgetting the copyright issue for a minute, do you really think there are "tens of thousands" of lines of Unix V copied into Linux and if so why didn't SCO disclose any of them?

And, because that's a pretty good question I wrote a long response:

Sort of - and I can only guess

It's my belief that the official System 390 Linux port done in Germany and released to the community via SuSe was done with the assistance of people who worked, or had worked, on AIX maintenance and development - and who therefore brought a lot of that system specific knowledge and access along with them.

(It didn't help that the project was an emergency funded, ego driven, put down response to a much better and cleaner but highly unofficial Linux 390 port by a guy who'd managed to offend his entire management chain within IBM.)

I think the author of the advanced unix programming book [Marc Rochkind] testified at length about the impact this had on the code base - but two things seem to have happened:

  1. first, because it's not line for line copying the comments about what they were trying to do or what the mechanism was are often nearly the same between the two (AIX 3.X and SuSe 7.X/PPC), but the code itself is sufficiently changed to deny that thousands of lines business; and,

  2. neither SCO's lawyers nor their own top management understood the technical issues, and kept issuing releases based on their highly simplified view of the consequences of what happened rather than what actually happened.

The worst part of this, of course, is that all the more probable futures are worse for Linux than they were before the jury did its thing.

In exchange for which, the guy fed me another straight line:

OK, you have peaked my curiosity and have my attention. How would the Linux community come ahead if SCO had actually won?

to which I said:

1) the copyrights are a distraction - this would have left the way clear for the original lawsuit to go ahead and it didn't have to have any impact on Linux.

2) it would have left SuSe as a viable market competitor not up for grabs by the next MBA with an itch to sue

3) now that copyrights have been made an issue, someone capturing Novel can most probably really do what SCO's lawyers only thought they could do: issue real Linux licenses and make them stick.

The basic problem here is that it's hard to get clarity on this thing - in my opinion because the underlying claim was a slamdunk from the beginning, but the attempt to attach a big dollar value to it never made much technical, financial, or legal sense to anyone other than the lawyers involved.

And, of course, this kind of furball generates its own weird spin-offs - as the guy's next question gives me a chance to illustrate:

Do you believe that Novell retained the right to control SYSV contracts (I believe this was in section 4.16 of the APA) and could have prevented the SCO - IBM lawsuit?

No - whoever holds the AT&T contract is contractually obliged to enforce it - i.e. if Novell held the rights, they also held the obligation. This is one area where a new owner could sue because IBM's $50M raises all kinds of arguable issues. (Note: I have no idea how right/wrong this would be - too many unknowns.)

This will go down in legal history as an astonishing hail mary that actually worked. Amazing really.

And that brings us to one of many consequences arising from the jury verdict: because if the deal didn't transfer the rights, it's reasonable to conclude that the obligations didn't transfer either. Given that, I'd guess (and this is pure speculation: actually working out the details would take legal expertise and a lot of time) that it should now be possible to argue that the verdict, along with Novell's testimony leading to it, negates the entire sales agreement - and because Novell's own CEO and chief negotiator didn't know what they weren't selling, an allegation of fraud isn't any crazier than what actually happened with the copyrights - possibly leaving Novell on the hook for all of the costs and losses incurred by SCO since this started.

Overall this is a case in which the next surprise has almost always seemed a red herring to those judging on the basis of the underlying issues - and red meat to those to those using any available means or information to attack SCO. Right now, I have no idea about what's going to happen next: imaginary scenarios run from IBM trying for cover by directly or indirectly acquiring Novell, to Microsoft buying them out and ultimately making Linux either its own core OS or impractically expensive - conceivably we could even get an outbreak of sanity - but one thing seems clear: with this much money on the table somebody's going to do something, and just about all the longer term scenarios look net negative for the Linux and open source communities.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.