Re: triggers wishlist
On Sun March 30 2008 23:01:40 Christian Perrier wrote:
> You're apparently living in that past, where debian-devel was a
> playground for trolls and flames. Welcome to 2008: conflicts, even the
> most complicated ones like the recent one about dpkg, are now solved
> by discussion and constructive approach. This can take time, this can
> be more complicated than a few patches to C code....
You are mistaken. There are at least three dpkg problems: (a) that the
triggers feature has been unnecessarily delayed by eight to ten months,
(b) that bugs in some key packages are not being fixed fast enough, and
(c) that serious software engineers are discouraged from working on Debian.
None of these problems have been fixed, whether by discussion or otherwise.
The fact that none of these problems have been resolved is itself a problem.
> From your own claims, you apparently have a great knowledge about
> programming and wisdom about development projects management. Could
> you consider sharing some parts of that Wisdom with your fellow
> colleagues and also develop the same wisdom in human relations?
I just did. People like Ian who work hard to improve Debian despite all
the roadblocks thrown in their way deserve recognition. Equally, people
who abuse their power to delay Debian features by eight to ten months
deserve a different kind of recognition.
You have written of the difficulty of getting bugs fixed in key packages.
Debian has laudable goals but the execution is often farsical. The more
that Debian allows incompetents to abuse their power the more that
experienced software engineers will choose to spend their time elsewhere.
> If you have no such intent, I suggest you really reconsider your
> involvment in this project and your subscription to debian-devel. A
> subscription to debian-curiosa would be more appropriate for mails
> liek the one I'm answering to currently.
Debian has many mailing lists but none devoted to project management
because Debian has no project management. A few people have limited
areas of responsibility, and the DPL has a bully pulpit, but there is
no Torvalds or Stallman. Not even a PHB. And so marginal programmers
who have managed to obtain control of a key package are allowed to
abuse their power, block important features for eight to ten months,
and effectively guarantee that no serious software engineer will fix
dpkg bugs because it just isn't worth the hassle of dealing with petty
And I'm sorry, I don't have a solution to offer. I wish I did because
Debian is such a valuable endeavour. But I can tell you the first step
towards finding a solution, and that is to recognise the problem.
It doesn't have to be like dpkg. Take a look at the TeX maintainers.
They do a wonderful job despite a very challenging upstream. Maybe we
can find a clue to a solution by examining the teams that work well.