On Wed, 15 Dec 1999 09:36:51 PST, the world broke into rejoicing as
Rob Walker <[EMAIL PROTECTED]> said:
>
> >>>>> On Wed, 15 Dec 1999 23:14:42 +1100 (EST), Robert Graham Merkel
> >>>>> <[EMAIL PROTECTED]> said:
>
> <snip>
>
> Robert> Obviously, bug fixes and wart removals must also happen before
> Robert> a release.
>
> I disagree here. I would like to have way way more releases, so that
> we can have more people working on things. I don't feel comfortable
> having a current cvs tree around just to graft dave P's patches to, so
> that I can _begin_ working.
>
> At least more stuff in cvs so I can fell comfortable hacking on what
> is there. Why not a weekly sub-point release, which would only turn
> into a main-point release when we felt the stability was there? e.g.,
> 1.2.37 would be out now, with 1.2.38 and 1.2.39 every friday. And you
> know what? sometimes the sub-point releases wouldn't even compile.
> that is fine.
>
> As for getting out the bugs and warts before a major release, that is
> also counter-productive, for we are limiting our bug finding and
> debugging base needlessly. I agree that 1.3.0 should compile, but
> what better reason for 1.3.1 than 'Changelog: "The register should <do
> something simple"' ?
What is really needed is for there to be several developers that have
the ability to check things in to CVS.
At this point, it's only overloaded Linas who does this, which means
that sets of patches keep building up until he gets around to putting
them in.
I would think it a *very* good idea to try to arrange for there to be two
or three "trusted lieutenants" that could also submit patches into the
"official tree." That should *definitely* include Dave; Rob Browning
would probably be the second one "whose patches virtually always get
applied," and probably another one or two people. (Adding more people
causes some degree of "politicking" to raise its ugly head...)
This would provide the ability for changes to get into "official CVS"
on a more rapid basis, and diminish the need for things to be closely
examined by one fixed person.
There are enough people actively working on the code base that this
is starting to be truly necessary...
--
"...Deep Hack Mode--that mysterious and frightening state of
consciousness where Mortal Users fear to tread." -- Matt Welsh
[EMAIL PROTECTED] - <http://www.hex.net/~cbbrowne/lsf.html>
--
Gnucash Developer's List
To unsubscribe send empty email to: [EMAIL PROTECTED]