Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@httpd.apache.org
Reply-To: <rbb@covalent.net>
From: "Ryan Bloom" <rbb@covalent.net>
To: <dev@httpd.apache.org>
Subject: RE: Apache 2_0_31 is now rolled
Date: Sun, 3 Feb 2002 11:12:19 -0800
Organization: Covalent Technologies
Message-ID: <009f01c1ace6$b3852eb0$7f00000a@KOJ>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
In-Reply-To: <ELEDJONBOPPAEGANDEEICECGCBAA.joshua@slive.ca>
Importance: Normal


> > From: Justin Erenkrantz [mailto:jerenkrantz@ebuilt.com]
> 
> > Ian was hesitant to bump to 2.0.32 because he was under the
> > impression that it was not permitted to bump so close to a previous
> > tag.  He was the RM, so it was his call.
> 
> This argument has been had befor (ad naseum), but...
> 
> This is based on the false premise that version numbers are scarce.
> Tagging
> and rolling are fairly easy operations (with the fancy new script).
> Arguing
> for days about exactly what changes should justify a new tag is
difficult.

No that isn't what this is based on.  It is based on the fact that
tagging the tree with two different versions within two days discourages
people from testing.  If I roll a release every few days, why should
anybody test them, because they know that another release will be made,
which will obsolete what they are testing.

There is a balance that must be achieved between tagging often enough to
allow the code to develop, and tagging so often that it doesn't make any
sense to test what was tagged.

Not long after the current tag/roll procedure was developed, we had this
same situation, and Roy himself agreed that rolling more than once a
week discouraged people from testing the tarballs.

Yes, versions numbers don't mean anything, but people's time does.

Ryan