Re: [Gdbheads] proposed change to GDB maintainership rules

[David Carlton]

On 29 Jan 2004 14:58:48 -0500, Ian Lance Taylor <address@hidden> said:

> I just went back and re-read the long e-mail thread from a year ago
> which starts here:
>     http://sources.redhat.com/ml/gdb/2003-02/msg00277.html

> In that thread, the problem seems to be described as ``it takes too
> long for gdb patches to be approved.''

> That was a year ago, and I don't know if that is still the problem.
> If it is, then I would say that the answer is simple: gdb maintainers
> need to make handling patches a higher priority.

I don't think it's that simple.  We have a resource limitation: people
who can review changes don't do as frequently as others might like.
We can relieve this limitation in two ways:

1) Have each person review changes more frequently.

2) Increase the number of people who can review changes.

Probably both of these are good ideas.  Having said that, for now, I
favor 2, for the two reasons:

* Solution 1 basically boils down to "nag people more".  Which, while
  sometimes necessary, isn't my favorite way to manage personnel
  problems; one concrete problem with it is that somebody's response
  might be "fine, I'll stop being a maintainer", which will make the
  problem worse instead of better.

* It seems to me to be an obviously good idea that everybody competent
  to approve patches should be able to do so.  What's the benefit in
  further restricting the number of people who can approve patches?

> So, for those people who think there is a problem, please try to
> write it down.  If it takes you more than four sentences, then you
> aren't describing it well.  Take the time to say precisely what you
> mean.

1) Patches take too long to get approved.

1a) Some patches deservedly require approval, but nobody in a position
    to approve them takes the time to review them.

1b) Some people should be able to approve their own patches in areas
    of GDB where they are currently unable to do so.

2) In cases of conflict, the loudest voice wins, irrespective of the
   technical merits of the various arguments.

I am personally much much more affected by 1 than by 2, and by 1a much
more than 1b.

David Carlton
address@hidden