launchpad milestones only for releases not rcs?

[Martin Pool]

2009/5/15 Bob Tanner <tanner at real-time.com>:
> On 2009-05-07 17:43:34 -0500, Martin Pool <mbp at canonical.com> said:
>
>> I was just wondering if there is much point having Launchpad
>> milestones (as bug targetting points) for release candidates as
>> opposed to the actual release.  If we're trying to fix something for
>> 1.14 then we'll be trying to get it in the rc, and if it misses rc1
>> we'll be trying for rc2 or the final release.  Having a shorter list
>> is simpler.
>>
>> There is the point that milestones have target dates, and if there is
>> only one for a release when would we put it?  On the other hand I
>> don't think we refer to the dates in Launchpad very often.
>
> Sorry to dumb up an old thread.....
>
> I'm starting the 1.15rc1 release process and part of my (unpublished)
> release docs state to create the next  Milestone after release.
>
> I've read the thread a couple times and I don't see a definiative answer.
>
> Just keeping 1.15 and not doing rc milestones so all bugs get targetted to
> 1.15?
>
> Keep doing rc milestones?

Well, you get a say in it too because you probably use milestones as
much as anyone else...

As I see it the thread said: the only useful reason to have targetting
at rcs rather than just releases is so that you can see when
specifically they were fixed.  I don't personally think this is very
compelling because we don't always reliably set the target when a bug
is fixed.  (And that's not really what the field is for, it's target
not fixed-in.)  Against this you have to count that seeing on one page
all the fixes needed before the final release seems useful, and that
(at least at the moment) manipulating the milestones is a bit more
work than it should be.

So I'm still inclined to say "just releases" for now, unless someone
specifically says -1 today

-- 
Martin <http://launchpad.net/~mbp/>