Question to Candidates
James Troup <james@nocrew.org> writes:
> But I never personally replied to Joey's mail about the next point
> release explicitly saying that fixing sudo was a pre-depends, and I
> apologise for that.
You're not a DPL candidate, and if this question is relevant at all,
it's relevant to DPL candidates. So don't think it's a personal
question, though I would appreciate hearing your thoughts on the
matter also.
Especially however, I want to hear what the candidates think.
I would appreciate the work done by others in Debian more if there
were some kind of rule, principle, or guideline that if a question is
not given a serious and thoughtful response, allowing for delays and
vacations and whatnot, that any debian developer has the right to go
ahead and do something themselves.
We already have this rule in the case of NMUs to fix release critical
bugs. We also have a procedure for replacing developers who go
totally MIA, and an informal procedure for hijacking packages from
vaguely-MIA maintainers.
But I feel frustrated when I ask a developer a question, and hear
nothing back for ages and ages. (And I admit that there have been
times when I have myself been guilty of this same problem.) However,
if there is an RC bug in a package, and the maintainer has not
responded, eventually it's allowed for me to just NMU the package
myself.
How about extending this to other areas of the project? One
difficulty, of course, is that NMUs are massively documented, and
fairly easily reversed. Other changes are not so, which may make this
too hard, in practice, to accomplish.
What about a related idea, in which everyone has some sort of
obligation to respond to emails? I appreciate James' apology above
for not responding to a message from Joey (or whatever the details
are; it's not something that particularly concerns *me* in this
case).
I have experienced that an email to debian-release nearly always gets
a response. I appreciate this, and it means that I can make progress,
or at least feel like I can. When I have a release-relevant question,
I get an answer, even if I write an over-hasty question, or whatnot.
Moreover, since it's a public mailing list, it doesn't matter if the
official release managers are overloaded: generally there is someone
else who can answer if they don't. And if a question is missed or
overlooked, it's fine to repeat it and ask again.
What about some sort of general expectation that other developers do
the same thing? I would find interactions with ftpmaster more, hrm,
comfortable? pleasant? if I could rely on getting some kind of
response back. However, when I mail ftpmaster, my recollection is
that most of the time the work implied by my question happens, or
there is a good reason for it not to, but I almost never have any
interaction with the person who did it (except in the case of NEW
queue processing, where I do hear back). That's only one example, but
perhaps there are others.
Thomas
Reply to: