[Python-Dev] 1324 bugs in the tracker

[Georg Brandl]

Christian Heimes schrieb:
> Dear fellow Python developers!
> 
> The Python bug tracker contains more than 1,300 bugs and it's growing.

Not speaking of the 432 bugs that weren't migrated from SourceForge
(though I don't know how many of them were open).

> And growing ... and growing. I'm picking a bug once in a while or
> tossing some invalid bugs away but it's a helpless cause. The bugs are
> augmenting with anybody stopping them.

This is unfortunately not an easy problem. I've had some thoughts about it
myself, and done some forays through the tracker closing "easy" bugs, but
it has become more difficult to find such lately.

> Well, I'm exaggerating a bit but you probably get my point. The core
> developers can't keep up with new bugs and check old bugs at the same
> time. The resources are already stretched thin. But Brett gave me an
> idea how we could solve the problem when he posted the link to
> http://www.djangoproject.com/documentation/contributing/#ticket-triage.
 >
> What do you think about inviting some trustworthy and experienced Python
> users to join the cause? They don't need to solve every problem and they
> won't need developer access to the svn. Instead their task is cleaning
> up the tracker, categorizing bugs and checking patches. The tracker sure
> contains a lot of outdated junk and already fixed bugs.
> 
> A group of five to ten highly motivated people could squall through the
> tracker smashing all those ugly bugs like the infantry in Starship
> Troopers -  but hopefully with less loss on our side. :]

Many old issues in the bug tracker fall in one of these categories:

* Odd problem on a non-mainstream platform
* Minor problem that is hard to fix
* Minor problem that is in theory easy to fix, but there is no consensus
   on how to fix it
* Minor problem that is easy but very tedious to fix
* Major problem that is very hard to fix
* Problem that one developer is more qualified to fix than all the others,
   but he is retired
* Behavior about which even developers cannot consent whether it is a problem
* Patch for a minor problem that is correct in theory, but will cause
   backwards compatibility issues
* Patch for a minor problem that is not correct, but points in the right
   direction

And of course, many RFEs that don't have a chance of being implemented
without someone writing a patch and making a good cause on python-dev,
which probably nobody except the OP will have an interest in.

None of these categories are issues that you can just close, especially if
you are new to Python core development, if we want to have a satisfactory
resolution to each closed bug (which was the policy, as far as I was
concerned, but it can be changed of course).

On the other hand, there are kinds of issues than can be dealt with quickly:

* Duplicated issues which were bug/patch pairs at SF
* Issues where developers requested OP feedback but got none

Georg

-- 
Thus spake the Lord: Thou shalt indent with four spaces. No more, no less.
Four shall be the number of spaces thou shalt indent, and the number of thy
indenting shall be four. Eight shalt thou not indent, nor either indent thou
two, excepting that thou then proceed to four. Tabs are right out.