Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!husc6!rutgers!cmcl2!brl-adm!brl-smoke!gwyn From: gwyn@brl-smoke.ARPA (Doug Gwyn ) Newsgroups: comp.unix.wizards Subject: Re: CALL FOR UNAME OUTPUT FROM EVERYWHERE! Message-ID: <8272@brl-smoke.ARPA> Date: 30 Jul 88 05:00:01 GMT References: Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) ) Organization: Ballistic Research Lab (BRL), APG, MD. Lines: 18 In article eric@snark.UUCP (Eric S. Raymond) writes: >On AT&T and other POSIX-conforming systems I want to use `uname' as a way to >get processor/os/version in a reasonably standard format. To do this, I need >to know what variations in uname output to expect. Forget it -- years ago I tried to get the vendors to cooperate on standardizing the use of the various uname fields, which AT&T neglected to do. Unfortunately many of them made what appear to be random choices in their use of the fields. It wasn't helped any by the fact that vendor OS version numbers didn't match the closest UNIX System V version numbers, and the latter have been so disorganized that very few people can correctly interpret one. E.g. "UNIX System V Release 2 Version 2 for DEC VAX", not to be confused with SVR2.2 for DEC VAX, nor with SVR2V2 for 3B20, nor with DEC's Ultrix-32 for the VAX (any release). If you want such a facility, you or 1003.2 will have to design it yourself. uname is hopelessly broken as it now exists.