SCO
Sues IBM Over Linux Not Unix
Having first threatened to take Linux
users who use their UnixWare libraries to run UNIX software on Linux
without paying appropriate royalties (and garnered the anticipated
enmity of the Linux community), SCO is now looking for real money.
That means looking for someone with deep pockets to sue.
As my lawyer likes to tell me, in
America anyone can sue anyone, it doesn’t matter whether you
actually have a reason.
SCO has decided that since it’s
loosing its UNIX on Intel business to the rising Linux tide, it must
be IBM’s fault. The
fact that much of IBM’s Linux business is somewhere else (and most
of the Linux business isn’t IBM’s)seems to have escaped reason,
logic, and SCO.
Where should we start?
With the fact that Linux was created separately from UNIX and
not by IBM, but by a Finnish student and an international band of
free programmers? But,
of course, suing them would be a waste of time – they don’t have
any source of serious money.
How about with the fact that IBM does
not sell Linux, but merely partners with the major Linux
distributors, especially Red Hat and SuSE?
It turns out that what SCO is
complaining about in its formal complaint (since it could scarcely
start out by complaining that IBM had written Linux) is that IBM is
using the information it gained as a licensee of UNIX for its AIX (a
UNIX-variant) product to enrich Linux, at SCO’s expense.
I spoke with Mike Fay, who is acting as
the IBM spokesman in this matter.
He said that SCO did not approach IBM before suing them and
notes that the IBM license to UNIX is irrevocable and perpetual.
Mike believes the suit is not about Linux, but rather about
IBM (and its deep pockets).
Truth time.
In my ancient history, I used to write
the specifications for a lot of word processors and editors.
And there was a company in California called Interactive
Systems that was AT&T’s first licensee to offer a commercial
version of UNIX. IBM came to Interactive (c. 1981) and asked them to assist in
preparing the IBM port of UNIX, to be called AIX (IBM had procured
its own license from AT&T).
Interactive hired me to specify the editor IBM required.
Because even then, at the very start of its UNIX history, IBM
wanted to improve the product by providing a better editor than the
ones then offered.
IBM had already worked on a number of
operating systems. I won’t be able to list them accurately and in order but by
the time they acquired the rights to UNIX/AIX they had
already written and brought to market operating systems for
generations of mainframes and for their distributed processing
systems. They also had
some low-end products like the Series 1 and had just (I’m a little
sketchy on the exact dates here) or were about to bring out the PC
with Bill Gates’ DOS operating system, licensed from a Seattle
neighbor. And they
still had (shortly) the AS/400 and OS/2 ahead of them.
It’s a lot of operating systems skill, experience, and
expertise.
They also produced all the middleware
(infrastructure software) that worked with those operating systems
to provide the customers with the appropriate environments for their
custom and out-of-the-box applications.
I do believe this must mean they know a
great deal about how to write and refine operating systems without
peeking at the UNIX source code.
In fact, I suspect the AIX source code was probably enriched
by IBM’s knowledge of data center needs and mainframe operating
systems.
We could state some of the obvious
things, too.
-
It’s really easy to prove if a UNIX
program contains copied code.
Our friend the former VP of R&D at Interactive notes
that UNIX, unlike other operating systems has a function, DIFF,
designed to let you find duplications.
We’d assume SCO hasn’t found any of its code in
Linux.
-
If it did find any SCO code in Linux,
I’m not sure how it would prove that the source was IBM,
unless it was identical to AIX code and not to anything else.
SCO code has been around a long time and lots of people
have had access to it, not just IBM.
-
Much of the UNIX code and the
methodology that surrounds it has been open source for years (BSD,
Mach, et al) and IBM UNIX experts are just as likely to have
gotten their expertise from those sources, particularly during
their educations, as from AIX source code.
If you
would like to amuse yourselves with just how silly the American
legal system is, read the relevant paragraphs of the SCO complaint.
The complaint itself is on the SCO site at http://www.sco.com/scosource/complaint3.06.03.html.
The paragraphs that are most interesting, I think are 82-89.
First they accuse the Linux community of being a bunch of
incompetent hobbyists and then they accuse IBM of having created the
Linux market. You may
draw your own conclusions.
As you might imagine this has enraged
the Linux stalwarts since they are quite certain they wrote Linux
before IBM became interested in it and that it was pretty robust,
too. There are also a
number of Linux distributions who think they were responsible for
creating the Linux market.
There is a great article with a raft of
Linux developer comments about the above – some pro-IBM, some
anti, very few pro-SCO at http://lwn.net/Articles/24747/?format=printable
. You might enjoy it.
You will also find articles at all of the other Linux sites.
I have previously been
involved in discussing the earlier SCO legal situation (about the
use of the UnixWare libraries by Linux users) with Mike Angelo of
Mozillaqest. You can
find our discussions on their site www.mozillaquest.com.
Mike tells me more on SCO is coming.
(back to top)
