Return to BSD News archive
Newsgroups: comp.unix.bsd Path: sserve!manuel!munnari.oz.au!news.hawaii.edu!ames!elroy.jpl.nasa.gov!sdd.hp.com!wupost!gumby!destroyer!uunet!indetech!cirrus!dhesi From: dhesi@cirrus.com (Rahul Dhesi) Subject: Re: AT&T sues BSDI & Our Retaliation Message-ID: <1992Aug5.224337.6733@cirrus.com> Sender: news@cirrus.com Organization: Cirrus Logic Inc. References: <9207231306.AA06854@GRANNY.CS.NYU.EDU> <4442@hq.hq.af.mil> <UEKLEINF.92Jul29092927@queen.mcs.Drexel.edu> <1992Aug5.072411@eklektix.com> Date: Wed, 5 Aug 1992 22:43:37 GMT Lines: 31 Can authors of freely-redistributable software retroactively prevent AT&T from using it? Could the Regents of UCB do this, for example? In <1992Aug5.072411@eklektix.com> rcd@raven.eklektix.com (Dick Dunn) writes: >Assuming you mean "freely redistributable" (e.g., with a BSD-style copy- >right notice), you still can't do much. You can re-issue the software with >a new, more restrictive notice...but that doesn't let you retroactively >change the copyright on what's already out there. My understanding is that nothing in US law prevents an author from retroactively changing the copyright terms, provided they are consistent with copyright law and no enforcible contract is breached. There is no enforcible contract unless there is "consideration" (i.e., payment). This means that, for example, if UCB releases some copyrighted code and permits others to copy it under certain conditions, UCB can retroactively change its terms, provided it does not try to retroactively and unilaterally change a valid contract. The question, then, is whether AT&T has a valid and irrevocable contract with UCB that allows AT&T to use UCB code. I believe 4.2BSD, 4.3BSD, and 4.3BSD-tahoe (and probably 4.3BSD-renoe too) were distributed subject to signed license agreements, and consideration was required; so if AT&T signed such licenses, it can probably use the code from those releases and UCB may not be able to retroactively stop this. But if AT&T took any Net/2 code and no license was signed and paid for, chances are that no enforcible contract exists. If this is so, and if any Net/2 code found its way into the SVR4 kernel, then there could be some legal fireworks. -- Rahul Dhesi <dhesi@cirrus.com> also: dhesi@rahul.net