Re: Bug#762194: Please consider declining (was: Initial draft of affirming transition to systemd as default for #762194)
- To: firstname.lastname@example.org
- Subject: Re: Bug#762194: Please consider declining (was: Initial draft of affirming transition to systemd as default for #762194)
- From: Josh Triplett <email@example.com>
- Date: Sun, 28 Dec 2014 21:45:14 -0800
- Message-id: <20141229054512.GA1690@thin>
- In-reply-to: <CANTw=MNvx8sn-y9aiKEWm_S4CwWPwVDAKLqYyoK0rKdXqP78firstname.lastname@example.org>
Michael Gilbert wrote:
> Abstention is likely the wisest option currently available; given that
> rolling the dice on another resolution could consume those scant
> remaining fumes. Even if the resolution itself sounds a lot like the
> status quo, there is still huge risk that it will annoy at least
> someone (consuming more capital of course and producing more
> negativity in the project) whereas abstention, as a signal to the
> project that further design by committee may be coming to an end, will
> So, with that said, I would like to suggest that the committee
> consider declining further init system arbitration acts, starting by
> closing the two currently open tech-ctte bugs as declined.
In principle, I think the technical committee should indeed hesitate
before ruling, and remember that declining to rule is a valid outcome.
If the only ruling thus far had been the original decision to make
systemd the default, I would agree with declining to issue any further
However, in this particular case, the technical committee *already*
issued a second, unfortunately-worded ruling, leaving the situation
rather open-ended ("After the result of the General Resolution is known,
we intend to formally resolve the question of automatic switching of
init systems."). We should not leave people concerned about when or if
that bludgeon will drop. Thus, in this case, it seems warranted to
*explicitly* acknowledge the currently implemented approach of switching
on upgrades and resolve the situation with finality.
Note, in particular, that the ruling explicitly "affirms the decision of
the init system package maintainers", rather than issuing an independent
ruling that just happens to roughly coincide. I'd like to see the TC
more often saying "we support this work already being done" and almost
never saying "we're going to go off and figure this out ourselves".
Also, as far as I can tell, the committee has only one open bug
remaining about systemd, not two.
- Josh Triplett