On Thursday 26 July 2001 18:21, David Faure wrote:
> > The big disadvantage of branches is that you have to merge changes
> > between branches so we rather not use branches when it isn't absolutely
> > necassery.
> >
> > Compiling KDE is already bad enough, having to keep two branches up to
> > date would seriously reduce my productivity.
>
> I can only agree with Waldo here (heh, won't be the first time :).
>
> I'm about to import the future libkotext into a separate branch, its own
> "unstable" branch. Don't want it in the translations, is the main reason :)
>
> KDE is too big for having 2 branches of everything. Better branch only
> where necessary.

I agree here, with one suggestion though: _please_ standardize the name of 
branches to "UNSTABLE" everywhere. Now I have to fire up webcvs to find the 
name of the KSIRC_TEXT branch, I had to do this in the past for the KWord 
richttext branch, etc.

I'm not sure if it makes sense to have an UNSTABLE branch everywhere by 
default that is equal to CVS HEAD, so cvs co -rUNSTABLE would give the most 
cutting edge code for an entire module. Reason why I'm not sure:

- If CVS allows UNSTABLE to be an alias for HEAD unless specifically 
specified as branch in a given dir this is nice to have. Since I'm far from 
sure that CVS allows this, you would essentially end up with another branch, 
which was exactly what we wanted to avoid... If this problem is hypothetical 
(I don't know) then it might be a good idea to have though.

Even with a local UNSTABLE identifier (in the dirs where there actually is a 
branch), then at least you know the branch name without guessing if you know 
that there is an unstable branch there. This is much more likely than the 
chance that you additionally also know the name of the branch...

Martijn