Re: Backporting / stable

[Carl Sorensen]

On 4/30/11 4:51 PM, "address@hidden" <address@hidden> wrote:

> On Apr 30, 2011, at 8:03 AM, Graham Percival wrote:
> 
>> On Sat, Apr 30, 2011 at 08:52:42AM -0600, Carl Sorensen wrote:
>>> On 4/30/11 8:49 AM, "Graham Percival" <address@hidden> wrote:
>>> 
>>>> We don't need to be sure that we're clear.  All we need... or at
>>>> least, all I want before uploading the official 2.14.0... is for
>>>> one week without *known* Critical bugs.
>>> 
>>> Right, but we haven't had that happen with the beaming since
>>> beam-collision-engraver was added.
>> 
>> I'm pretty certain that we've had at least one 7-day period
>> without new Critical bugs -- the only probablem was that such
>> periods coincided with having old Critical bugs around.
>> 
>> I'll admit that our record so far is still less than 48 hours of
>> critical-bug-free-ness.  But once we *do* have some bugs in the
>> system, we can go for days and days without any new critical bugs
>> being reported.  If we'd been a bit faster in fixing some bugs, or
>> a bit luckier and having a Critical bug report a few days later...
>> or luckier in having critical bugs being reported earlier, and
>> then getting fixed earlier... then I think we could have had 2.14
>> by now.
>> 
>>>> Well, I'm on vacation for most of May, so we're looking at a June
>>>> release at the earliest, in any case.
>>> 
>>> So it seems that we've got some time to wait on this decision.  I'll just
>>> pretend I never raised the issue, and when you get back from vacation if
>>> we're still adding Critical beaming bugs, we'll have the discussion again.
>
> I have no problem commenting out the Beam_collision_engraver and putting a
> use-at-your-own risk bit in the docs along with a list of known problems.  In
> the devel version, the BCE could be a non-commented-out feature that people
> continue using/breaking so that we can perfect it over the next year or so.  I
> agree that issues keep cropping up, and I think that we won't be able to
> anticipate all of them without broad, continued use in a variety of
> circumstances.

I don't think we'd want to add that to the stable version.
Use-at-your-own-risk should be part of the development version.  But we've
got some time now, so let's see what happens in the next six weeks.

Thanks,

Carl