httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@engelschall.com (Ralf S. Engelschall)
Subject Re: Proposed ANNOUNCE for 1.2.2 (II)
Date Mon, 11 Aug 1997 20:06:59 GMT

In article <97081111392891@decus.org> you wrote:
> From the fingers of Marc Slemko flowed the following:

> >Traditionally, test tarballs have been test tarballes; not release ready
> >and a new one has to be rolled.  Once the new one deemed to be the final
> >version is rolled, then it is left for a day or two and people test it
> >even more before being released.  By the time we got to rolling the final
> >tarball, it was almost certain that either that version would go out and
> >the version was incrementend to the next -dev right away.
> >
> >You are doing things differently.  It did not appear clearly to me that
> >you were doing things that way.  That is the biggest cause of my concern.
> >
> >I prefer doing it the old way, but that is just my preference.  If people
> >like this way, thats fine.  Just make it clear...

>     I prefer the old way, too: every tarball is a test one until
>     everyone says, "okey, let's go for it."  Then:

>     1. the version number is bumped (well, the "-dev" is removed),
>     2. the tree is tagged,
>     3. the *final* tarball is rolled, and
>     4. the version number bumped to "{n+1}-dev".

>     I like that system; less chance of confusion about where things
>     stand.

Ok, no problem. Sounds ok. But please add this procedure to how_to_release so
we won't have the same confusion the next time someone rolls a tarball.

                                       Ralf S. Engelschall
                                       rse@engelschall.com
                                       www.engelschall.com

Mime
View raw message