Letters
To The Editor: SCO And
The Case vs. IBM
In
our March 16 and April
26 issues we wrote about SCO’s case vs. IBM in which it claims
(among other things) that IBM is illegally giving SCO’s
intellectual property, gained as part of the UNIX license that
underlies IBM’s AIX, to the Linux community.
I think you could summarize what we wrote previously by
saying, “We doubt that.” We
believe that Linux itself was written separately from UNIX and that
a firm like IBM, which is intensely careful of its IP, is unlikely
to casually give to the Open Source community code, which it
doesn’t know the origin of.
However,
we were not at all surprised when SCO’s PR folks asked if we would
speak with its CEO, Darl McBride.
Of course, we did that, although our experience in
interviewing folks involved in legal cases, is that they are rarely
in a position to tell us anything more than what is in the documents
they’re filing.
I
think that’s the case here. Darl
McBride mainly repeated what we had read in the legal documents SCO
had presented to the court – and which were the basis of our
previous articles.
Darl
reiterated that the basis of SCO’s suit against IBM centers on the
point that IBM created a contract with AT&T (the then-owner of
UNIX) in the mid-80’s, which then became the basis for AIX.
I don’t think that’s in question.
IBM,
McBride says, has to protect and keep in confidence that code base
and its derivative works. McBride
also claims (as does SCO in its legal filing) that it is the Methods
and Concepts that IBM uses in writing AIX that it isn’t allowed to
use (or teach others to use) for Linux.
I
think we’re going to have a very interesting court case that ends
up centering on the meaning of “derivative works” and “methods
and concepts.” After all, many of the things programmers learn
have little or nothing to do with particular operating systems,
languages or tools, but rather with good technique.
Saying they belong to one (and only one) of these spaces is
going to look pretty foolish when any experienced technologist can
point to their use across dozens of environments.
And separating the mainframe methods and concepts of IBM,
which may have been at least as important here, from UNIX concepts,
should be really interesting, particularly since SCO intends to
demand a jury trial. (Technical
cases, because of the difficulty of explaining the issues involved,
are almost always presented to a judge.)
I
seem to recall that I first learned to vigorously diagram sentences
to pursue precise grammar in first year Latin, but I doubt that
anyone would argue that it is a method that can therefore only be
used in Latin. It is,
of course, used in many languages, which have complex grammars,
often by people who have never learned Latin at all.
McBride
also wanted to point out that a major issue for them is the fact
that SCO thinks IBM somehow colluded with Intel to get SCO involved
in the Monterrey project and to learn about details of SCO UNIX on
Intel so that they could (a) damage SCO by diverting its resources
on a project which they knew would be abandoned and (b) learn
information that they would not otherwise have been entitled to.
I
tried to point out to McBride that in that case a whole lot of
industry analysts, myself included, had been hornswoggled, since we
had all attended joint IBM-Intel briefings on Monterrey and wasted
lots of time trying to understand the technical and marketing
strategy for the project.
I
believe it is unfortunate when the progress of markets (in this case
Linux) makes the IP of a company like SCO much less valuable, but
that is scarcely the fault of the participants in that market.
It is simply the way markets operate.
SCO itself is a participant in the Linux market and I can’t
wait to hear SCO explain to the court how it avoided doing anything
illegal in the Linux Open Source Community (where everything is
shared) while suing its colleague IBM.
McBride told me that since they will give a license to any
UNIX code in their own version of Linux, their version will be
legal. Aha!
So they are going to try to have a BETTER (that is more
legal) version of Linux? I’d
bet the Linux coders will code around that problem (and avoid the
SCO version), just on general principles, on the theory that it goes
against the grain of the GPL and Open Source.
But
as I said in may last article, we’ll see.
That’s what courts are for.
(back to top)
