[Reviewers] 4.3 delayed: should we unfreeze on 8/1 anyway?

[Martin von Zweigbergk]

On Mon, Jul 31, 2017 at 8:13 PM, Sean Farley <sean at farley.io> wrote:
>
> Augie Fackler <raf at durin42.com> writes:
>
>>> On Jul 31, 2017, at 13:32, Martin von Zweigbergk <martinvonz at google.com> wrote:
>>>
>>>
>>>
>>> On Jul 31, 2017 10:09 AM, "Augie Fackler" <raf at durin42.com> wrote:
>>>
>>> > On Jul 31, 2017, at 12:36, Gregory Szorc <gregory.szorc at gmail.com> wrote:
>>> >
>>> >
>>> >
>>> >> On Jul 31, 2017, at 08:13, Augie Fackler <raf at durin42.com> wrote:
>>> >>
>>> >> We need to hold 4.3 for a few days (possibly a week?) because of some security fixes and coordinating with other tools.
>>> >>
>>> >> Should we unfreeze default on the first anyway, and cut the 4.3 release from stable once we've got the all-clear?
>>> >
>>> > Since we're actively trying to lower contribution friction and since freezes are harmful to that goal, my vote is to unfreeze as planned.
>>>
>>> Works for me.
>>>
>>> That works for me too.
>>>
>>>
>>> If anyone else has opinions (especially dissenting ones), please speak up today. Thanks!
>>>
>>> Reasons against unfreezing seem like they would likely be related to the reasons we have a freeze. IIUC, the reason for the freeze is to force people (mostly mercurial developers) to test the release candidate and not the default branch. If that's right, I guess the question is whether we think the RC has been tested enough.
>>
>> *nods* Barring any objection, I'll announce the unfreeze-but-no-release-yet tomorrow AM America/New_York.
>
> Just to add: it's hard to test a secret, security rc :-)

The point was that unfreezing means there may be exciting new things
on the default branch that will distract people from testing the RC
more.

> I'm fine with
> unfreezing, too (but will probably be focused on testing the stable
> stuff with our products internally).

As I said earlier, I'm still fine with unfreezing.