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

Re: Debian needs more buildds. It has offers. They aren't being accepted.

On Mon, Feb 23, 2004 at 06:27:32PM +0000, Will Newton wrote:
> My opinion may not matter to you or any other person in the project, but I 
> don't think your attitude is constructive. 

And I don't think yours is. Imagine that.

If you want to be a constructive influence on the direction of the
project, *do the work*. If you want to put forward some opinions, well,
you're welcome to do that, but you need to realise that most of the
time they're going to be dismissed out of hand or ignored if you're not
willing or able to actually back it up by putting in the effort to make
your ideas actually work. 

Every single person (or near enough that there's no difference) who's
developing Debian or using Debian has opinions on things we do, and ideas
about how we could do them better. That's what you get when you have
intelligent and informed users. But expecting people to deal seriously
with every one of those opinions is utterly impractical.

> Having a pop at people for being 
> non-DDs or not doing "the work" from your position seems to me ...

What it should seem to be, and indeed what it is, is some advice on what
you should realistically expect as a result of expressing your opinion,
however intelligent and informed it is.

> ... indicative of 
> the attitude of a small number of people in this project who see their 
> position as a stick to beat people with rather than a privilege.

How do you think my position is _in any way_ a privilege?

Why do you appear to think my comments above don't apply to my opinions
too? I do near enough to nothing wrt the buildds, so while I have
opinions about them and have things I want implemented (like support
for experimental autobuilding, or different handling of build-depends,
or historically autobuilding support for testing updates and testing
security updates), I don't imagine that my opinion holds any particular
weight unless and until I can explain it in a way that convinces the
people who actually do the work.

I think my judgement on how to discuss things in Debian and how to get
things done has some merit because I've actually managed to discuss
things and get them done in the past. Not everything I want, certainly,
but quite a bit more than other people. Contrary to what people seem to
assume, I didn't win some sort of n-m lottery, or find a gold ticket in
a candy bar with "Admit 1 to Debian Cabal" written on it.

> I presented my opinion in an attempt to engage in a reasoned discussion. I 
> don't really see what's so wrong with that.

If you want to engage in productive discussion, you need to realise that
your opinion simply doesn't count for squat.

> > Working out who to accept and reject is an obviously controversial task,
> > it's also one Debian and our users rely on rather heavily considering the
> > damage that could be done if we do this wrong. As such, it's reasonable
> > and appropriate to expect that anyone who wants to suggest changes or
> > set new policies demonstrate a significantly higher level of competence
> > and trustworthyness than your average developer. The most obvious way to
> Your opinion of the average developer is telling.

My opinion of the average developer's (or user's) intelligence or skills
has absolutely no influence on my comments above. My comments apply to
people who are above average just as much to those who are below average:
the only difference is the above average developers tend to actually put
the work in so the end up saitsfying the criteria for having opinions
that matter.

I've no idea why you and others are so eager to make this personal,
or imagine that it already is. It's not.

> > demonstrate the appropriate level of both those attributes is to work on
> > n-m as an AM. Personally, I think the fact that you haven't done this, and
> > are evidently unwilling to do this, means your opinions aren't worth much.
> I strongly disagree. Opinions should be evaluated on merit, 

No, in general opinions shouldn't be evaluated at all. It's your right to
have whatever opinion you like on whatever subject you like; although some
opinions (like "Debian sucks" or "Free software sucks") should exclude
you from involvement in Debian. It's not opinions that matter at all;
it's useful ideas about how to improve Debian. And those should be and
are evaluated by the people who actually implement them, and as a rare,
final resort, by the tech ctte or Debian as a whole.

You have ideas about how to improve the NM process. That's great. It
shows you're thinking about things, it shows you care about Debian,
and it shows you've got some creativity. But the only way to be sure
it's a good idea is to test it against experience, which either means
you need to get the experience, or someone else who's already gotten the
experience needs to think the idea's valuable enough to spend some time
on. If nobody with experience is enthused enough to do the latter, then
you've got no one to blame but yourself for not making the former happen.

> That is prejudice, and small-minded attitudes like that mean 
> that good ideas are dismissed for the wrong reasons. 

If they're good ideas, they'll come up again; and in the meantime the
things that the people with the experience to accurately judge these
things will have had the time and opportunity to focus on the more
important and more urgent tasks.

> Maybe good developers 
> are also rejected for the wrong reasons. I don't think we should base 
> decisions on reputation rather than reasoned arguments.

Arguments count for nothing without evidence, and it's very rare for all
the relevant evidence to be readily available and obvious to people who
aren't significantly involved in the topic. If you try to make decisions
based on arguments that don't deal with all the facts, you'll usually
come up with a bad solution. One way of avoiding that, that doesn't have
the problems associated with information overload and wasted time trying
to make every scrap of information available on every topic available
to everybody is delegating tasks to individuals or smaller groups, and
trusting their judgement. Micromanaging them and second guessing every
decision that makes anyone unhappy is vastly wasteful and unproductive.


Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.

             Linux.conf.au 2004 -- Because we could.
           http://conf.linux.org.au/ -- Jan 12-17, 2004

Attachment: signature.asc
Description: Digital signature

Reply to: