Date: Sun, 7 Jan 2001 09:01:33 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Tim Van Holder <tim DOT van DOT holder AT pandora DOT be>
cc: djgpp-workers AT delorie DOT com
Subject: RE: DJGPP CVS users
In-Reply-To: <NEBBIOJNGMKPNOBKHCGHOEGCCAAA.tim.van.holder@pandora.be>
Message-ID: <Pine.SUN.3.91.1010107085247.21289D-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Sun, 7 Jan 2001, Tim Van Holder wrote:

> > We've never done that, and starting now would be tricky.  We use cvs
> > logs and the knowledge base for that.
> I don't see how tricky it can be - all it takes is a single run of
> rcs2log to create the initial version, and then some minor initial
> editing to clean it up.

It's tricky, believe me.  As an evidence, please look at standards.texi, 
the GNU coding standards document: it has a whole section on how to write 
good ChangeLog entries (and that section just got larger in the last 
version as I  asked Richard Stallman to add some more there, because some 
of his requirements were never documented before).

It is even trickier for people who don't use Emacs, and a very recent 
version of it, because then they need to watch out for all kinds of 
small gotchas and fix them by hand.

So if we ask for ChangeLog entries, we will need to have a ChangeLog 
Police to enforce the standards.  That's more workload on those who 
participate in peer reviews.  Those people are already overworked, so 
please let's not add anything that isn't absolutely necessary.

(And, btw, rcs2log doesn't DTRT in many cases, if you want the standard 
ChangeLog entries.)