The following is an imaginary cross-examination of an SCO spokesperson such as, for example, Chris Sontag. It is, of course, a work of rhetoric, not of law, and a real lawyer will probably blanch at my amateurish portrayal of his art.
Cross-examination is a lot easier when you can control the answers as well as the questions. Nevertheless I hope to present some relevant facts and explore some serious legal issues.
Does SCO own copyrights to the code for JFS?
No.
Who does?
IBM.
Who developed JFS?
IBM.
Then what claim does SCO have on JFS?
SCO has control rights on JFS, because JFS is a derivative of System V.
Does System V contain JFS?
No.
On what operating system was JFS first developed?
On AIX, which is a derivative of System V.
Did IBM contribute that version of JFS to Linux?
No. That version was abandoned.
Then where did the Linux version come from?
It was developed on OS/2, and later ported to AIX and to Linux.
This is a tangled tale. The original version of JFS was closely tied to the internal details of AIX. IBM then re-implemented JFS on OS/2, in a cleaner, more portable fashion. Later it was introduced to Linux and, in its new form, re-introduced to AIX.
Is OS/2 a derivative of System V?
Not so far as I know.
Then, JFS is a derivative of OS/2?
No, it's a derivative of System V.
Even though it came from OS/2?
By being incorporated into AIX, which is a derivative of System V, it takes on the same contractual status as the original System V code.
By this point the imaginary cross-examination has descended into pure fantasy. I have no idea what kind of legal theory SCO will come up with. I'm giving them the least incoherent theory I can think of.
JFS is an addition to System V, insofar as AIX is a version of System V. Can we agree on that much?
Of course.
And it's also an addition to Linux?
Yes, but an improper addition.
And why is it improper?
Because JFS is now a derivative of System V. IBM can't contribute it to Linux without SCO's authorization.
Was IBM permitted to make additions or enhancements to System V?
Yes, but the additions and enhancements must be treated the same as the AT&T work. That means that they can't be disclosed to anyone who doesn't have a System V source license.
Is that what the side letter says?
Not explicitly. It's implied.
In other words, that's not what the letter says?
SCO's lawyer: Objection, your honor!
I withdraw the question. Doesn't the side letter of February 1985 say that IBM's additions belong to IBM?
Yes. IBM owns JFS. But that doesn't mean that IBM can do anything it likes with it.
IBM owns JFS, but doesn't control it?
Yes.
How is that different from not owning it?
IBM owns the copyright on it.
So IBM can make copies of the source code for JFS?
Yes. But it can't show the copies to anyone who doesn't have a System V source code license.
When IBM distributes AIX, it makes copies of it, including copies of material from System V, correct?
IBM is not entitled to distribute AIX any more. SCO terminated their System V license.
Before SCO terminated the license, IBM could make and distribute copies of material from System V?
Of course.
And for internal use in development, IBM could make copies of the source code for System V?
Yes.
So IBM had the right to make copies of System V, subject to certain restrictions.
Correct.
Likewise, IBM had the right to make copies of the source code for JFS, but not the right to show the source code to anyone without a System V source license. Is that your contention?
Correct.
Aren't those exactly the same rights that IBM would have had if AT&T had owned the code for JFS?
Yes.
Does IBM's nominal ownership JFS give them any rights that they wouldn't have if AT&T had owned it?
IBM can show the source code to someone else who has a System V source code license, subject to the usual terms of non-disclosure.
Wouldn't IBM have had the same right to do so, if the JFS had been included in the original System V?
Yes.
So IBM has no more rights to JFS than if JFS had been included in the original System V?
Correct.
Why does the side letter even include the ownership clause, if not to expand IBM's rights as a licensee?
I can't say. I wasn't there.
So you have no answer to that question?
No.
Neither do I. I believe that will be all, your honor.
Last updated: 2003/10/11 01:1647 PM CDT.
Copyright 2003 by Scott McKellar (mck9@swbell.net). You may reproduce this document in its entirety in any medium without restriction, provided that you preserve this notice.