[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: RFS: ripit (updated package)

On Sun, 26 Apr 2009 17:16:30 +0200
Adeodato Simó <dato@net.com.org.es> wrote:

> I am writing to let you and the list and all sponsorees who may be
> reading this thread know that I don’t agree with most of what you’re
> saying, and that I don’t like the tone in which you are writing your
> thoughts, and that I think you are a harmful influence regarding these
> matters in particular.

Well, maybe. OTOH sometimes it is better to explain why RFS emails get
ignored, otherwise the ping requests get more and more pointless.

Is it really better to routinely ignore RFS emails for months?

There are too few sponsors, that has been an ongoing problem and I
don't have a magic solution either, but not providing the information
that sponsors need isn't going to make it easier for those who can
sponsor to do more sponsoring. Pinging the list with no extra
information doesn't justify taking time from other roles to obtain the
data that the maintainer could have included first time around.

> You are treating sponsorees as if you’re doing a grand favour to them,
> when (in my opinion) they’re just doing us an equally sized favour by
> spending their time in contributing to our project and wanting to
> learn to do it better. It is true that, sadly, I don’t get to do much
> real mentoring myself nowadays (and note how I use the word
> “mentoring” instead of “sponsoring”), and that’s a pity, because
> together with being AM, they are two of the most important tasks in
> Debian, because in their hands lies Debian’s future.

I wouldn't mind if those maintainers had the same expectations of
potential sponsors, but quite often, that is distinctly lacking.
Maintainers need to follow existing guidelines in the FAQ and
sponsoring requirements. The main section I quoted was from the FAQ on
mentors.debian.net - the RFS in question failed to expand on the bare
template or follow the advice of the FAQ.

There has been a very welcome improvement in RFS content recently - more
written content and less boilerplate - so it was sad to find one
maintainer who not only used the boilerplate without addition but also
followed up without improving things.

I do have time for sponsoring and I do invest time in one-to-one
assistance for those maintainers who I sponsor. It annoys me when other
maintainers expect the same without any consideration that sponsors are
few in number and all are overworked volunteers who continue to try to
help anyway.

Sponsoring a package is a considerable investment - it usually takes me
the best part of 4-5 hours to review a package that I haven't looked
at before. Maybe I'm too picky or too thorough but I have a strong
aversion to wasting my time locating information that the maintainer
has already been asked to provide via the FAQ.

Even for a package like ripit, I would have expected to spend at least
2-3 hours testing and reviewing the package before uploading. That is
why information on the language used by the package is so important to
me. I cannot afford to waste time finding out that a package is python
or PHP or KDE - I need to know in advance. The long description is
similarly important for even the most basic triage of incoming RFS

> Recently, an old sponsoree of mine who became DM mailed me to ask some
> questions regarding library packaging. I found it most pleasing to
> write a very detailed mail to him tailored to his specific questions,
> rather than telling him “Go read your answers scattered in these three
> documents and mailing list threads”. I was most glad for the
> opportunity to make a difference in his learning process, and felt
> proud in contributing if only a teensy bit in somebody excelling in
> their future work as a developer.

For reference, I spend a much larger amount of time on the
debian-embedded mailing list and spend time explaining and assisting
with various issues to do with those topics. It can be rewarding, it is
most certainly useful to the content of the manpages and documentation
for the code itself. However, on the mentors list the guidelines and
docs are routinely ignored. I don't see this on other lists, I don't
see that it reflects badly on mentors.debian.net, although docs can
always be improved, it reflects more about the maintainers not
following existing guidance and treating mentors.debian.net as if it
was just a queue.

> If you, Neil, don’t find it in you to see it that way or, even worse,
> think sponsorees should be spending their time striving to please you
> and your rules, and think it’s okay to go patronizing them around,
> then you should consider stopping. Or, at least, you should be aware
> that some of us (or at least me) who lurk in this list, are not in
> agreement at all with you.

I don't intend to be patronising but I do try to show where people are
making obvious errors that contribute to requests being ignored. I
don't like ignoring requests - it annoys me when I cannot make any
progress with an RFS because the maintainer has failed to follow the
existing instructions, especially if I might have actually been able
to help if the information was at hand.

I don't want to take up more of your time on this, I understand your
point of view and I don't intend to come across quite so abruptly but
the problem is not confined to a lack of sponsors.

> I do realize there are way more requests on this list than people
> willing to sponsor, and I have no magic solution to that. I only know
> that this thread deeply disturbed me, and I decided to let that be
> known.

I'm sorry that my comments came across that way and I will take your
comments on board but please, maintainers, provide the information
requested first time around in your RFS, don't make things difficult
for over-committed sponsors who are trying to help you.


Neil Williams

Attachment: pgpowxUUE4mjN.pgp
Description: PGP signature

Reply to: