chris@mimsy.UUCP (06/19/87)
In article <670@vixie.UUCP> paul@vixie.UUCP (Paul Vixie Esq) writes: >I have heard both Chris' and Ed's positions before. I think the entire UNIX >community would benefit from some resolution of this issue. Can we have some >more discussion? What I said was: It is my impression (note key word) that Berkeley does not care how much their stuff gets redistributed as long as .... This was rather careless; by `their stuff', I meant code that was clearly not derived from AT&T code. For instance, Mike Karels has said that the Berkeley TCP may be reused so long as their copyright is acknowledged. (But check with the lawyers first anyway!) As Ed Gould pointed out, The license between the Regents of the University of California and the party to which they supply a 4BSD distribution does not distinguish anything as to its origin. What this really means is: `Our license that you signed says that if you give away some code and a lawyer from AT&T comes knocking at our door, we need not worry; we can just point him at you.' That, at least, is clearly the *purpose* of the clause(s) in which The UC license states that the licensee will treat *all* of the material received under that license *as if* it had been received from AT&T, under the terms of the licensee's AT&T source license. It could still get you into twice the trouble. (Legalese is sharp: Use care when juggling it.) The problem with taking something like tip and giving it away is that there is no telling whether tip might happen to have something torn out of, e.g., uucp. >Can we have a little more discussion on this? There are other >examples.... What about Sendmail? I **know** there is no AT&T >code in it; Alas, knowing it and proving it in a court of law are two different things. Given a different state of affairs, including a more permissive license (such as the one I think CSRG would have preferred to offer), the UC Regents could conceivably be held responsible in part for anything you redistribute. Consider, e.g., if someone at UCB were accidentally to mark /usr/src/lib/cpp as `not AT&T derived'. >Sendmail is (was?) available for public FTP a while back, so that >all those ARPA sites could get at the MX-record version. Was this >proper, if BSD code has to be treated as AT&T code ("all reasonable >protections...") ?? Certainly. UC makes their own rules, so they may change them at any time---they did not have to sign their own license. Whoever made sendmail available was as sure as you that there is no AT&T code in it. I believe it is possible for others to get redistribution rights on a program-by-program basis as well. (This is known as `sticking your neck out just a bit at a time'. :-) ) >According to an article from John Gilmore a long while back, CSRG >distributed a list of AT&T vs. non-AT&T files with the 4.3 release. >I havn't seen this list, but it sounds like they (CSRG) are trying >to move away from the "keep everything under the AT&T license" position. It is hard to say. As I understand things, John Gilmore is in large part the reason that list exists (though this is merely a guess based on vague memories of a moment's conversation last January): He took care of organising its creation, and if so, he may someday be held liable for its contents. This is a risk, but one he thinks is worthwhile. Obviously CSRG would not have distributed it if they did not also consider it worthwhile. Ultimately, though, all this says is that you should check with the lawyers before you do anything. And I think this is where we came in. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) Domain: chris@mimsy.umd.edu Path: seismo!mimsy!chris