Thanks for pushing out those thoughts to the mailing list. I'm not
much on IRC lately, so I likely missed a lot of the discussions
leading up to this.

On Fri, Jan 14, 2011 at 10:24 PM, Elizabeth Leddy <ele...@umich.edu> wrote:
> In general, let's move away from the fixed timeline of announcing,
> gathering, reviewing plips and towards a continuous integration of new
> features.

I very much like this. I think our deadline approach doesn't work out
that well for us. It currently seems that no deadline we set is
actually taking seriously and we happen to postpone the real ones for
months afterwards. So I'm not all too motivated anymore to actually do
things for a "fake" deadline anymore.

It currently looks to me like we aren't able to ship the actual
finished work for 4.1, because we are waiting and waiting for some
work that isn't quite ready yet. I'd rather have earlier, faster and
smaller releases for the 4.x series again. The minor feature releases
ideally should be more time based than feature based.

For the major releases like 4.0 and 5.0 we need to handle them
differently, as certain types of features can only land in them for
backwards compatibility concerns. Having a feature almost ready and
then postponing it to the next major release means not shipping it for
two to three years. That's a lot different from not shipping some
small feature for another minor release and six months.

Your notes on the actual process sound good to me in general, I trust
the FWT to organize themselves ;)

Framework-Team mailing list

Reply via email to