Re: roadmap status update/update request



On Mon, 2005-03-07 at 11:36 -0600, Matt T. Proud wrote:
> I do not want to belabor this issue and make any more of you angry, but
> I do think that what you have posed here is still less-than-ideal, for I
> have, on countless occasions, submitted and seen other people's patches
> go neglected for eons in bugzilla. 
> 
> It is very discouraging. Whose fault is it?
> 
> Please, do not tell me this is a non-issue. How do you reconcile one's
> help and dedication with the reality that patches go ignored for an
> unreasonable period--or more often permanently ignored?
> 

It's a real problem, but I don't think it's anyone's "fault" per se.
It's just the reality of a finite number of developers and a huge number
of things those developers could be doing or are asked to do.

Elijah had some good suggestions. The Linux kernel guys are using
Bitkeeper to solve this problem. Firefox was just on slashdot today with
apparently the same problem (I don't know if it's true or not). It's not
new.

Most of the developers would rather write code themselves than spend all
their time on patch review, and this is even the right choice in many
cases; lots of times the core developers will be doing a large new
feature which is more important than lots of small tweaks. Sometimes
people just don't like patch review, or a module doesn't really have a
maintainer.

You can only put so much through a pipe of fixed size. Comes down to
physics in the end. But I'd certainly support efforts to try and make
the pipe larger.

One very good way to help is bug triage. Elijah has shown that with
metacity; basically the reason I have trouble with patch review is that
it takes too long to find/remember the patches to review, and then
actually put them in CVS. Elijah's help reduces my job to just looking
at the patch itself, and even though I'm a useless managerial type these
days, I can find time to do that. Of course Elijah is also writing a lot
of the patches, which doesn't hurt.

Havoc





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