Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 20 Nov 2003 04:53:47 -0800
From:      Terry Lambert <tlambert2@mindspring.com>
To:        John Baldwin <jhb@FreeBSD.org>
Cc:        Arjan van Leeuwen <avleeuwen@piwebs.com>
Subject:   Re: SCO goes after BSD?
Message-ID:  <3FBCB95B.44193DF2@mindspring.com>
References:  <XFMail.20031119135153.jhb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin wrote:
> On 18-Nov-2003 Michal Pasternak wrote:
> > Arjan van Leeuwen [Tue, Nov 18, 2003 at 10:19:07PM +0100]:
> >> What to think of this?
> >
> > So, who would be attacked by SCO in case they decide to run against BSD
> > systems? Which one of big-bucks-worldwide-famous corporations would it be?
> 
> Apple.

Unlikely.

A more likely target would be Cisco Systems or practically any
company using TCP/IP, given SCO's theory of what constitutes a
derivative work.

For the most part, however, the 1994 settlement agreement is
unassailable from a lot of different perspectives:

1)	It's a settlement agreement which both parties agreed
	to be legally binding.  An attempt to overturn it would
	open them to a Contempt of Court charge, at a minimum.

2)	USL was in violation of UCB Copyright on many printed
	materials; reopening this would make SCO subject to
	the counterclaim of copyright infringement.  If they
	lost, they would b liable for collecting every scrap of
	paper on which the material or derivative works have
	been printed.  How many Ultrix manuals did DEC print?

3)	Much of the code in SVR4.x was imported from the Net/2
	sources out of Berkeley.  Almost all of the ntworking.
	They will have a hard time proving provenance of their
	code.

4)	Part of the counterclaim's cause of action was Copyright
	and license violation by USL, by virtue of removal of the
	Copyright and license statements in the header files.

5)	Much of the code that makes up the SVR4 networking code
	was developed under contract to DARPA.  Despite the recent
	slapping of a GPL on things developed with public funds,
	things developed with public funds are technically requied
	to be in the public domain (i.e. slapping a license on top
	of it before releasing it is not allowed).

6)	One of the contributing factors to the settlement was the
	judge effectively telling USL "I think you have a very
	weak case, and will probably rule against you".

7)	SCO is an assign of the rights in the UNIX source code, and
	those rights were specifically limited by the settlement
	agreement.  SCO is therefore a priori bound by that agreement.

8)	USL's primary legal theory at the time was "trade secret
	disclosure"; however, trade secret law states that no matter
	how a secret is disclosed, once it is disclosed, it is no
	longer a secret.  This is generally useful in this case,
	since SCO can only go after the disclosing party for damages,
	and can not limit further propagation of the trade secret as
	if it were still secret (this is what they attempted to do);
	one of the judge's arguments was that they were attempting
	to obtain the moral quivalent of patent protection without
	disclosure, and that this attempt was unconstituional.

9)	The FreeBSD and NetBSD projects, at least, have auditable
	records of every line of code added since the 4.4 BSD-Lite
	code was imported into the tree.  For them to come after
	FreeBSD, as an example, they would need to overturn the
	settlement agrement, refile and win the case against UCB,
	and then prove that their trade secrets are still secret
	after having been published for over a decade and a half.

10)	The UCB license was the old Western Electric license, which
	did not have a non-disclosure clause in its original form;
	hence the Lyon's book.

11)	When UNIX was invented and first published, USL was a part
	of AT&T, and AT&T was specifically enjoined from making a
	profit of any kind off of software -- including a paper
	profit in the form of the accumulation of intellectual
	property -- as a result of the 1956 consent decree, under
	which they were legally acknowledged to be a monopoloy, and
	thereafter had to operate as a regulated monopoly.  It's
	not clear that their sale of USL would permit USL to later
	claim intellectual property from conversion of illegally
	accumulated assets.

12)	If SCO's theory of derivation is correct, then SVR4 is a
	derivative work of BSD UNIX and publically funded work.

During the original case, a number of well known people offered to
testify a witnesses on behalf of UCB; among these were Dennis Ritchie
and Ken Thompson, as well as other prominent computer scientists with
an involvement in UNIX since its inception.

Another interesting thing that happened was that MIT offered to fund
the defense, and offered their patent portfolio as ammunition (I
still get annoyed at UCB turning down this offer).

FWIW, I'm personally willing to testify as an expert witness as a
former Novell/USG employee (Novell/USG was the UNIX Systems Group
that was formed after the Novell acquisition of USL).  I personally
camped out in Mike DeFazio's (then Novell VP over Novell/USG, and
the man who eventally dropped the lawsuit) office with a number of
other Novell/USG employees to get 386BSD, FreeBSD, and NetBSD the
same deal that USL was giving BSDI.  Originally they sent a cease
and desist order to everyone they could find, Jordan included, and
there was no grace period for continuing to ship code (like BSDI
was being allowed) until the 4.4-Lite code was made available. I'm
pretty sure Jim Freeman and others would be similarly inclined.

Finally, remember that civil cases are won or lost on the basis of
a preponderance of evidence.  It is much easier for thousands of
angry engineers who know the code to produce such evidence than it
is for lawyers who don't to manufacture it.  Going by number of
reams of paper alone, ther's no way SCO could win, if it came down
to it.

In summary, the legal case against any SCO claim against UCB or
claim on BSD code is very, very strong.

-- Terry



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3FBCB95B.44193DF2>