home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!convex!darwin.sura.net!mips!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!rutgers!igor.rutgers.edu!dumas.rutgers.edu!hedrick
- From: hedrick@dumas.rutgers.edu (Charles Hedrick)
- Newsgroups: comp.os.linux
- Subject: Re: BSD Unix aka Freedom is a myth....
- Message-ID: <Jul.26.14.29.23.1992.1031@dumas.rutgers.edu>
- Date: 26 Jul 92 18:29:24 GMT
- References: <23992@castle.ed.ac.uk> <1992Jul21.112705.6276@cs.cornell.edu> <veit.711728954@du9ds3>
- Distribution: comp
- Organization: Rutgers Univ., New Brunswick, N.J.
- Lines: 71
-
- There's nothing illegal about using "Unix ideas". You can be sued
- only if you violate some specific intellectual property right. These
- can be created by copyright, patent, trade secret, or special
- contractual arrangement. Pieces of Unix are protected by each of
- these. Most of the new source code is copyrighted. I believe there
- is at least one patent (the setuid bit), which however I think may be
- either expired or not enforced. But the primary issue in Berkeley's
- case is trade secret and licenses that involve "methods and concepts".
- Berkeley presumably agreed in their license agreement to regard all of
- the methods and concepts of Unix as trade secrets. If Berkeley has
- really recoded everything so that there is no actual Unix code, then
- there should be no copyright issues. However the claim would be that
- the code still used the same concepts and methods as the original
- code. Indeed in theory even the networking code, which is completely
- original to Berkeley, might also involve general Unix methods and
- concepts. However things could get complex, because ATT has allowed
- publication of books by their own people on Unix internals. Trade
- secrets clauses always say that if information because available to
- the general public, through no fault of yours, then it is no longer a
- trade secret. So one would have to show that either (1) Berkeley's
- code had been written by people who had not seen the ATT code, or (2)
- the code used only those concepts and methods that had become public.
-
- BSDI may be in a different situation, because they have signed no
- trade secret agreements with ATT, and have no access to ATT trade
- secrets, except from Berkeley. I don't know enough about trade secret
- law to be sure whether they are responsible for violations of
- Berkeley's license by Berkeley (if any such violations occured). This
- is a situation where as a non-lawyer I don't even want to speculate.
- But I'm pretty sure BSDI is in a different legal situation than
- Berkeley. ATT may be trying to finesse this issue by using false
- advertising rather than a direct suit for violation of trade secret.
- I think the theory is that if Berkeley has violated their agreements,
- then BSDI's claim that their code is ATT-free is false, even if BSDI
- themselves can't be sued for that violation.
-
- However the point of all this is to talk about the implications for
- Linux. I'm assuming that Linus has not had access to any ATT trade
- secrets. Code written from scratch by someone with no access to ATT
- information is clearly not in violation of copyrights, trade secrets,
- or concepts and methods. It could be in violation of patents if there
- were any patents and those patents were valid. My guess is that there
- are no relevant patents, but that's just a guess. As far as I know,
- Linux, Hurd, Coherent, and other Unix variants done by people without
- access to ATT code are safe. In order to maintain that status, people
- with access to ATT code should not contribute any substantial code to
- the Linux kernel. (People at MIT may be an exception, if they really
- an ATT license that doesn't have the trade secret clause. I frankly
- find that hard to believe.) I assume that bug fixes and other minor
- tweaks are not a problem.
-
- I've read the ATT claims and BSDI counter-claims. ATT is making only
- the most general claims. They do not give any specific examples, nor
- do they specific exactly which type of intellectual property right was
- violated. BSDI's response compares it to a Xerox suit against Apple,
- which was apparently very similar. Like ATT's, they did not give any
- specific violation of intellectual property right. Instead, they sued
- for false advertising, as ATT has done here. (The alleged falsehood
- is BSDI's claim that the code is free from ATT licensing.) BSDI says
- that you can't make a claim that this is false without showing that
- some specific intellectual property claim is violated. In effect they
- say that the suit is too vague to be meaningful.
-
- The question of the use of the Unix name seems more clear-cut. It is
- well-known to be an ATT trademark. Unless someone wants to try to
- prove that it has become a generic name, it should not be used without
- ATT's permission. It appears that BSDI has settled on this issue.
-
- As usual, before taking actions that depend critically on these
- issues, consult a lawyer who is familiar with intellectual property
- issue.
-