Return to BSD News archive
Path: sserve!manuel!munnari.oz.au!hp9000.csc.cuhk.hk!uakari.primate.wisc.edu!doug.cae.wisc.edu!zazen!decwrl!mips!swrinde!elroy.jpl.nasa.gov!ames!agate!soda.berkeley.edu!gwh From: gwh@soda.berkeley.edu (George William Herbert) Newsgroups: alt.suit.att-bsdi,comp.unix.bsd Subject: Re: UNIGRAM's article on the USL-BSDI suit Summary: Utter Bullshit! Message-ID: <15o75lINNfi9@agate.berkeley.edu> Date: 5 Aug 92 09:31:33 GMT References: <KANDALL.92Aug4161214@globalize.nsg.sgi.com> <5042.Aug412.31.0892@virtualnews.nyu.edu> <KANDALL.92Aug5175428@globalize.nsg.sgi.com> Sender: gwh@soda.berkeley.edu (George William Herbert) Organization: U.C. Berkeley, CS Undergraduate Association Lines: 59 NNTP-Posting-Host: soda.berkeley.edu In article <KANDALL.92Aug5175428@globalize.nsg.sgi.com> kandall@nsg.sgi.com (Michael Kandall) writes: >[...] >It may be legal to license stuff for a while, re-implement it, make >sure `diff' fails and cut the supplier out of the action, but I cannot >say I would call it moral, nor is it conducive to an industry based on >open-systems. > >BDSI's intentions are clear. Re-implement USL's UNIX System V (they >even advertised ITS-UNIX), claim it's USL-code-free, cut USL out >of their money. Uhh, what planet have you been on, Mike? UNIX is not equal to UNIX System V. Many many versions preceded or are in parallel to SV UNIX. BSD has been the "main competitor" in the overall architecture field, though Mach is now taking up much of that area. BSD UNIX started out based on early (mostly Version 7) UNIX code. It's incorporated some elements of later AT&T code (V32, if I recall right), but hasn't had any simmilarity with SV until AT&T made SVR4 which is mostly BSD compatable, in an attempt to attract people who like some BSD features in UNIX. Note that in that case, it was AT&T moving closer to BSD's standard, NOT the other way around 8-). BSDI (the company) took a version of BSD (the UNIX) that had had all origional traces of AT&T code cleansed out (but not replaced, so it wasn't a complete and functional OS) so it wasn't covered under AT&T/USL copyright per se (though it may be covered under "derived work", which is what the lawsuit is about) and wrote their own versions of the parts that had to be taken out. Thus, they had a complete OS whose origins clearly are in BSD UNIX, but that's clear of USL's code (we think... actual evidence to the contrary willingly accepted as correction by me & everyone else around here). BSDI didn't cut anyone out of anything; CSRG did if anyone did. Berkeley has said for years that they wanted to take the last AT&T code out and make an unencumbered version. That BSDI took a partial step in that process and completed it before CSRG did at Berkeley is entrepreneurial, not conspiratorial. If that whole process was unfair to AT&T/USL, then so be it, but it's been going on AND APPROVED OF IN GENERAL IN THE UNIX COMMUNITY for five years. Only now, as it's nearing completion in various forms (BSDI jumped the gun and made some money... darn, I hate it when people are smart and make money) do critics pop up. Is it bad for Open Systems? Depends. Lots of industry people like to define open systems as "We'll cross license with X, employ Y million programmers to do everything they do in our system, and keep it all proprietary.". I disagree. Open Systems are supposed to be one thing: Open. Where it's clear and agreed upon what the system does, and everyone can use the system. USL's UNIX is open but expensive. A free BSD would be the ultimate Open System. Mach is pretty good, as its popularity attests to. Open Systems is not an excuse to hire more programmers to keep everything the same. Open Systems is a whole way of OS life, not a buzzword. -george william herbert gwh@soda.berkeley.edu gwh@lurnix.com gwh@uchu.isu92.ac.jp until 28 aug