Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!cs.utexas.edu!uunet!bfmny0!tneff
From: tneff@bfmny0.UUCP (Tom Neff)
Newsgroups: news.software.b
Subject: Re: patch tracking in C News
Message-ID: <14416@bfmny0.UUCP>
Date: 25 Jun 89 09:12:31 GMT
References: <9476@b-tech.ann-arbor.mi.us> <1989Jun23.194724.2047@utzoo.uucp> <9486@b-tech.ann-arbor.mi.us> <1989Jun24.210926.25443@utzoo.uucp>
Reply-To: tneff@bfmny0.UUCP (Tom Neff)
Organization: ^
Lines: 18

The problem with date-based patch levels, and I think that Henry has all
but admitted this but I want to get it on the record explicitly, is that
you never know whether you've missed an interim patch.  If I am at
patchlevel 3 and I receive patchlevel 6, I know something's wrong.  If I
am at patchlevel 3/2/90 and I receive patchlevel 9/17/90, who knows.

Conversely one might say that the problem with numeric patch levels is
that you don't know when they date from.  However if the patch is
distributed as a new style context diff a la 'gdiff -C', then you get
everything you need.

If the cnews authors want state of the art, their version strings
should be formatted to include both a raw number and the revision date.
-- 
You may not redistribute this article for profit without written permission.
--
Tom Neff				UUCP:     ...!uunet!bfmny0!tneff
    "Truisms aren't everything."	Internet: tneff@bfmny0.UU.NET