Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!brl-tgr!gwyn From: gwyn@brl-tgr.ARPA (Doug Gwyn) Newsgroups: net.unix,net.unix-wizards,net.micro Subject: Re: Binary Compatibility 80286 Message-ID: <2464@brl-tgr.ARPA> Date: Sat, 26-Oct-85 08:30:26 EST Article-I.D.: brl-tgr.2464 Posted: Sat Oct 26 08:30:26 1985 Date-Received: Sun, 27-Oct-85 07:33:24 EST References: <248@omen.UUCP> <10764@ucbvax.ARPA> <2143@amdahl.UUCP> Organization: Ballistic Research Lab Lines: 15 Xref: watmath net.unix:6049 net.unix-wizards:15470 net.micro:12503 > Uh, excuse me, but Mr. Forsberg neglected the *approved* version > of UNIX System V/286; the one that has passed port acceptance. See, that's the problem with trying to establish binary standards: who sets the standard? In this case, the only two logical choices are AT&T and Intel, and in spite of AT&T's efforts, the package vendors did their own thing(s). I don't know all the details, but I suspect the AT&T system used COFF and the Xenix system used Microsoft's proposed object file format; the two are of course different. (Incidentally, after working with COFF for a while, I found it to be insufficiently portable. I haven't evaluated Microsoft's proposal.) And common object file format is only a small part of establishing binary compatibility.