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

Re: Mailing lists



On Thu, Feb 19, 2004 at 05:52:25PM +0100, Mathieu Roy wrote:
> Wouter Verhelst <wouter@grep.be> wrote:
> > On Thu, Feb 19, 2004 at 02:49:12PM +0100, Goswin von Brederlow wrote:
> >> Thats a pretty hostile remark.
> >
> > Funny, this thread started with claims being made about people who spend
> > almost all of their free time to Debian, which should be replaced, as if
> > their contributions weren't considered useful.
> 
> I'm surprised to hear you saying that. Generally, you express more 
> subtle point of views:

Well, let's say he's hit the wrong button.

[...]
> How the hell did you reach the conclusion that doing much means being
> always right, never wrong?

Well, see, I didn't say that. But since you bring it up: if you're doing
a lot of work in a certain area -any area- then you'll, no doubt, earn a
lot of experience in that area. James and Ryan have been working with
wanna-build and buildd for a long time; much longer than I have, and
certainly much longer than you have[0]. So, it's fair to say that their
experience in this area is worth a lot more than mine, and certainly a
lot more than yours. If they say "I think it's best we do it this way",
I'll generally take their expert opinion for granted. If they tell me I
made a mistake, my first action would be to believe them, and to correct
my error. Even after having been the buildd admin of a few machines
since halfway 2001, that would still be what I'd do.

Does that mean they're always right and never wrong? No, certainly not.
Both James and Ryan are human beings, not computers (at least not to the
best of my knowledge), and unlike well-designed computers, human beings
are known to make mistakes.  However, if I think they're wrong, what
I'll do is make damn sure I'm right before claiming they've missed the
ball. Send them a private mail, pointing out the tidbits that I think
they might be wrong at, with a (short!) reasoning as to why I think
that's the case.  Not a long one, since I don't want to waste their
time[1]. 

What I won't do is start a flamewar on -devel, which will annoy the hell
out of everyone who reads it, including both Ryan and James who'll then
probably ignore me or give my requests lower priority if I mail them for
some other reason. After all, such flamewars don't really change
anything, even if the DPL does waste his time to explain what happened,
and why. Best case, it takes a month before everyone who was not
directly insulted will have forgotten all about it. Go ahead, try me. A
month from the last message in this flame, ask a random developer whom
you know is reading -devel, whether he remembers what that flame was
about again.

What I'm suggesting is that hostility towards people, even if they annoy
the hell out of you by not replying to your requests, does not buy you
anything but them turning away from you. If you want people to do
anything to you, be nice to them. Even if that means you have to wait
for a long time. Even if that means you have to outlive their attitude,
if you think it's an annoying one.

Because frankly, a sysadmin's job is a thankless job. If you're ever on
IRC, #debian-devel, and watch what happens when some nice guy who uses
the 'elmo' alias says something, you'll see what I mean: James is
stumped with requests. It usually takes a while for a request to be
handled, and they're *all* urgent. No matter what. If you think that
time is unacceptable, then I have a suggestion: help them out, instead
of requesting for them to be replaced. *Be* their (partial) replacement,
and have the flames be directed at you, instead of flaming James and/or
some of the other admin people. That's helpful; claiming that they're
doing a lousy job is not.

[0] Since, AFAIK, you're not even remotely involved with our
    autobuilder infrastructure. Do correct me if I'm wrong.
[1] Just for the record: I've never tought it necessary to do that yet.

> So, are you really trying to communicate here? Do you really think
> than Nathaniel and others are spending time to write mails just to be
> hostile to some DD? 
> 
> All this discussion about how hostile are Nathaniel and friends, how
> bad it is to question DD choices, it just looks like a way to avoid
> discussion the real issue raised by Nathaniel and friends: "Re: Debian
> needs more buildds. It has offers. They aren't beingaccepted.". It
> makes the issue looking even more strange, just like if there was
> little nasty secret to hide. Why is it so hard to get an official
> statement about why buildd offers aren't being accepted?

There has been one. Two days after the start of this flamage, the DPL
has posted a mail that outlined the reasons for the machines not being
accepted:
http://lists.debian.org/debian-devel/2004/debian-devel-200402/msg00463.html
That's with his DPL hat on ('From: leader@debian.org'), so I guess it
can't go any more official than that.

Yes, I know the answer has not been accepted by some. As someone who's
been maintaining a buildd machine for almost 3 years now, however, I'd
like to say that I understand Ryan's motives for not accepting the
buildd offer, even if I don't agree with all of them. His mail to me in
which he declined the offer was short, to the point, communicative, and
friendly; thus, in my experience, many of the accusations ("they're not
communicating!", etc) are simply wrong. And yes, both James and Ryan
seem to be ignoring this thread. Can't blaim them for that.

> So only solution to the problem you can come up with is to ignore the
> discussion with a killfile? Looks that there is a real problem out
> there.

If this were a productive discussion, then yes, there would be a
problem. However, as it is, not only have we got an endless, silly, and
heated flamewar, it is also highly off-topic here.

-- 
Wouter Verhelst
Debian GNU/Linux -- http://www.debian.org
Nederlandstalige Linux-documentatie -- http://nl.linux.org
"Stop breathing down my neck." "My breathing is merely a simulation."
"So is my neck, stop it anyway!"
  -- Voyager's EMH versus the Prometheus' EMH, stardate 51462.

Attachment: signature.asc
Description: Digital signature


Reply to: