We didn't really have "code freeze" clearly defined so no one should have
to be "guilty" of anything :)  but that's what i'm trying to sort out here.


if we feel good that we will all be responsible to not have some gnarly,
half-cocked PR coming in on code freeze week that chews up the whole week
when we could(should?) be focused on:

+ documentation
+ testing - recall that we integration tests take hours and hours to
execute - and thus each change that affects those force those to be
re-executed
+ having a stable environment for the release manager to validate docs,
builds, etc.
+ retrospective on process/procedure
+ planning next release cycles

then we don't have an issue with code freeze week being about "merging
remaining PRs".

I suppose that in code review we could -1 for merge if it was
half-cocked....it would just stink if that was a "critical" priority issue
for that release for some reason.  of course, if we were being responsible
we would have identified that as a critical issue much earlier in the
development cycle and it shouldn't end up coming in on Code Freeze Eve.
Seems like a longshot for this to be a problem i suppose....



On Mon, Nov 9, 2015 at 8:40 AM, Daniel Kuppitz <[email protected]> wrote:

> I thought of it as a week to merge remaining PR's (and do minor cleanups,
> like you're planning to do for the tutorials). Code freeze means (to me),
> that we don't add new things / PR's, but care more about the things that
> were already done.
>
> Cheers,
> Daniel
>
>
> On Mon, Nov 9, 2015 at 2:21 PM, Stephen Mallette <[email protected]>
> wrote:
>
> > We agreed on having a 1 week code freeze - yet i'm now merging PRs on
> code
> > freeze day.  bad?
> >
> > I'm not sure it is.  I guess it *could* have been bad if someone tried to
> > jam in a massive change during code freeze week and the PR was full of
> > holes and risks.  Seems like we were smart enough to mitigate that for
> this
> > release as we carefully watched the "high-risk" JIRA tickets on the way
> > into code freeze.  If we continue to do that, I imagine we'd be fine, but
> > all committers will have to be responsible in their choices as it
> pertains
> > to PRs.
> >
> > Anyway, just thought I'd bring it up, in case anyone had any thoughts
> they
> > wanted to share on procedural changes in this area.
> >
>

Reply via email to