From owner-freebsd-stable Mon Oct 6 05:53:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id FAA25921 for stable-outgoing; Mon, 6 Oct 1997 05:53:54 -0700 (PDT) (envelope-from owner-freebsd-stable) Received: from shrimp.dataplex.net (shrimp.dataplex.net [208.2.87.3]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id FAA25911 for ; Mon, 6 Oct 1997 05:53:50 -0700 (PDT) (envelope-from rkw@dataplex.net) Received: from [208.2.87.4] (user4.dataplex.net [208.2.87.4]) by shrimp.dataplex.net (8.8.5/8.8.5) with ESMTP id HAA29628; Mon, 6 Oct 1997 07:53:46 -0500 (CDT) X-Sender: rkw@mail.dataplex.net Message-Id: In-Reply-To: References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 6 Oct 1997 07:54:54 -0500 To: Chris Dillon From: Richard Wackerbarth Subject: Re: Fwd: CVSup release identity Cc: stable@FreeBSD.ORG Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >I have been following this debate since it started, and my own idea at the >beginning of it was very similar to the above. A time stamp is about as >fine-grained as you can get, and is not dependant on how either CTM or >cvsup work. After all, this has been how we have been describing just how >far along we are on the stable tree for a long time now. "And you cvsupped >when? 19:45 10/4/97? Oh, thats just before I made those >changes." I have no objection to using timestamps to identify "progress" along a branch. However, you need to realize that the stamp must be applied at the time that the snapshot is taken from the master tree. We used to look at the timestamp placed on the kernel at compile time. However, I don't care when you compile the source. (Although I can conclude that you are missing later changes.) What I want to know is WHICH version of the master source you are using. I also do not care when YOU got the source from someone else. Since you do not have direct access to the master source by either ctm or cvsup, we need to stamp the sources at the time that they are extracted from the master source by the primary distributor. Anything less increases the uncertainty interval. The more latency in the distribution mechanism, the higher the uncertainty. The only objection that I have to the use of timestamps as the identifier is that they may not be "user friendly". A user who gets a release CD knows that he has release "5.1.9". He doesn't realize that it was generated on October 31 at 4:26:37 PM PDT. Similarly, users of CTM distributions know that they have updated through delta 4726. Richard Wackerbarth