[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
RE: [cross-project-issues-dev] DSDP-TM requests delaying our M6 dropby 4 days
|
Hi Dave,
I disagree because I think it all depends on the
consumers.
As far as I know, we don't currently have any consumers who
require
strict train date. But we do have consumers who much rather
download
a Milestone for their personal use and adoption, rather
than an I-build
or an M6 and later M6b ... that's all extra work for lots
of people (me
as release engineer as well as adopters, testers and
downloaders)
which I'd like to avoid if
I can.
If it turns out that there is ANY consumer who DEPENDS on
the
train schedule, I'll stick with the schedule because that's
what we
committed and what everyone needs to be dependable on.
But as long as nobody depends on it, I'd rather make life
easier
for my other consumers, placing a clear message: THIS is
the
download you should use as "milestone 6" because it
fulfills
the promises we have given (availability date, API,
quality) to
the greatest possible extent.
Makes sense?
Cheers,
--
Martin Oberhuber, Senior Member of Technical
Staff, Wind River
Target Management Project
Lead, DSDP PMC Member
Martin,
Lack of testing is no excuse for slipping dates ... as you can see from
other typical eclipse project's behavior :)
My advise is to stick with the train metaphor ... if
you leave the station late, you never know what other train you might run into
later on, down the tracks.
While
it does greatly depend on the wishes of your adopters and committers, I
recommend you declare M6 with whatever you have, with whatever amount of
testing you can do, but (as you've done here) warn people it hasn't gotten as
much testing as you would have liked, due to the important API work, so if
anyone runs into a blocking or critical issue, you would consider providing
them with an M6b upon request.
Besides ... once one project slips, it'll change our mindset and you
know who else will start asking for a slip too. :)
I think there can be good reasons to slip a schedule,
but not so much for "unknown, potential problems" ... only for "we have a
known problem that will prevent adoption of the milestone".
Make sense?
Thanks,
