Re: [Gdbheads] proposed change to GDB maintainership rules

[Eli Zaretskii]

> From: Ian Lance Taylor <address@hidden>
> Date: 30 Jan 2004 19:23:25 -0500
> 
> If Andrew is not the head maintainer, if all the maintainers are
> equal, then why do people feel that he has so much power?
> 
> Earlier, Eli said:
> 
> > It is true that the global maintainers could in theory solve the
> > problem, but approval of a patch which Andrew objects is not something
> > I'd expect to see, nor something I myself would like to do.  I'm sure
> > you understand why.
> 
> Well, in the light of the above, I guess I don't understand why.

Because, as far as I'm concerned, Andrew _is_ the head maintainer.
Like you, I think that tyranny (I call it ``a voluntary monarchy'', a
term coined by DJ Delorie for the way the DJGPP project is maintained)
is a better arrangement than voting for technical projects.  For me,
Andrew has been the de-facto head maintainer since the day Stan Shebs
stepped down.  For me, it was only natural that this should be the
case, since I don't believe in technical projects without a leader.  I
was surprised to learn (very recently) that other maintainers objected
to Andrew being in this role, and cited half-spoken agreements from
discussions I wasn't part of that seemed to imply that wasn't the
case.

In my assessment, for all practical purposes, Andrew behaves like a
head maintainer for quite some time.  That is why I wrote that
approving a patch over Andrew's objections is not something I'd expect
to happen a lot.  If others disagree with this, please speak up.

> For the binutils, Nick Clifton is the acknowledged head maintainer,
> and there are seven other people with blanket write privileges, of
> whom I am one.  If Nick objected to a patch of mine, I would not check
> it in.  But if any of the other people with blanket write privileges
> objected to a patch of mine, I would consider their arguments, and, if
> I were sure that I was right, I would check it in anyhow.  For that
> matter, people have checked in patches to the binutils over my
> objections.

That is also a picture I'd like to see in GDB.  I thought we had such
an arrangement, but then I learned that I was wrong.