On Mon, 2002-07-08 at 13:37, Peter Donald wrote:
> On Mon, 8 Jul 2002 20:19, Leo Simons wrote:
> > > > Which will lead to forking, competition, the community falling
> > > > apart
> > >
> > > IMHO, there's absolutely nothing wrong with forking if forkers don't
> > > abuse the Avalon brand, and enough information is provided to let users
> > > choose. The healthiest approach is to encourage diversity on one hand,
> > > and facilitate darwinian selection on the other. If you're worried about
> > > forks hurting the Avalon brand, then create a fork that unifies the
> > > other forks, and see if that flies ;)
> >
> > hmm. We've not had a fork before.
> 
> We have had massive numbers of forks before.

okay, I definately misphrased some stuff here. Let me try and explain
what I ment.

We've not had forks of the core framework before. I consider the
metainfo stuff currently under development something that should move
into framework once stable. As long as we have competing implementations
that will be impossible.

> One thing we have never 
> discouraged is forks. Berin will probably be the one who remembers it best - 
> at one stage we had 3 forks of framework cooking, we also had 4 different 
> pooling frameworks (3 from different people and one merged version).
> 
> This does not even mention the number of different container forks or 
> revolutions we have had and continue to have (remember Merlin1 was a fork of 
> phoenix and I encouraged Stephen to work on it and eventually submit it to 
> excalibur).
> 
> Forks are fine and healthy. Forced adoptions of untested code is never a good 
> idea and has always resulted in crapola that we later regret and spend a year 
> deprecating and replacing.

very true. I am quite okay with the current divergence of efforts into
multiple forks. What I am very concerned about are statements that seem
to indicate there will not be convergence later, or that it will be
impossible.

cheers,

- Leo Simons



--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to