From owner-freebsd-stable Sun Oct 5 09:14:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id JAA27395 for stable-outgoing; Sun, 5 Oct 1997 09:14:10 -0700 (PDT) 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 JAA27390 for ; Sun, 5 Oct 1997 09:14:05 -0700 (PDT) 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 LAA08841; Sun, 5 Oct 1997 11:13:54 -0500 (CDT) X-Sender: rkw@mail.dataplex.net Message-Id: In-Reply-To: <01bcd1a1$f152b340$4a53ab98@hetzels> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Sun, 5 Oct 1997 11:14:52 -0500 To: "Scot W. Hetzel" From: Richard Wackerbarth Subject: Re: CTM patch level added to newvers.sh Cc: stable@FreeBSD.ORG, nate@mt.sri.com Sender: owner-freebsd-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >>>One method I would like to see is that the .ctm_status file becomes a >>>permanant part of the src tree. (Why?) Because CTM deltas are created at >>>least once every day. >> >>Yes and that is exactly the problem with this solution. If I sup the CVS >>tree then some rcs deltas may (and will) be added to the tree that are not >>yet noticed by CTM. In other words, during one day a sup would give me >>different versions of the source tree, all with the same CTM delta version. >>And since people are building 'releases' from the CVS tree (instead of >>ftp'ing them) this is not good. >This suggest that CTM deltas are created every 6 Hours. Actually, that depends on which tree you are "sup"ping. The updates for the cvs tree are presently done every 8 hours. Source trees are on a slower interval. > And Thus if you sup >at 4 different times during the day, you will get at least 4 different >versions. Now, if you sup between the 6 hour delta creations, then you will >get different versions with the same CTM value. That all depends on what source you "sup". I believe that only those who can directly commit to the master tree are able to get a version without some latency. For the bulk of the users, they can only access a snapshot of the tree that is taken at some interval. >How fine, do you want it? Personally, I think that the granularity of ctm deltas is adequate to excessive for this purpose. Those of you who can commit are quite capable of cobbling pieces together and understand which pieces you got when and where. For the "unwashed masses", being able to identify the particular snapshot in an easy and understandable manner is important. However, their granularity is more likely to be weeks rather than minutes. See my other reply for additional comment. Richard Wackerbarth