On Mon, Oct 08, 2018 at 06:37:02PM -0400, Kevin O'Connor wrote:
On Fri, Oct 05, 2018 at 01:46:53PM +0200, Gerd
I'm wondering whenever it makes sense to switch seabios to time-based
releases, like many other projects do meanwhile.
For major releases one release per year looks reasonable to me, given
the low rate of changes we have. 1.11 was tagged in November 2017.
So maybe target 1.12 for November 2018 ?
For stable releases we could plan to create a release every one or two
months. Skip in case there are no patches queued. Or stick to the
current model which kind-of syncs stable releases to qemu releases
(which is roughly every four months).
A release in November sounds good. How about we target November 16th
with a feature freeze on October 21st.
Sounds good to me.
We can document a yearly release if that helps. I
think we'd need to
be able to make a release out of cycle though if a situation warrants
Sure, we can make exceptions if needed. The release model used to be
something along the lines "when we have enough changes that a new major
release makes sense". But with the low rate of changes we have right
now "every November" looks more useful to me.