Re: [Gdbheads] proposed change to GDB maintainership rules

[Michael Snyder]

Andrew Cagney wrote:
> > Although it has not happened in a while, Andrew has done exactly that.
> > This was, in fact, when he first (to my recollection) spelled out his
> > position that a blanket write maintainer did not really have blanket
> > write privileges.  Here's the thread, which does not include some
> > rather heated telephone conversations:
>
>
> Michael,
>
> In choosing to immediatly revert the patch in that thread, rather than 
> simply pointing out the technical reasons why committing it was 
> incorrect, I recognized that I made an error of judgment.  At the time I 
> acknowledged this and appologized to you in front of (well on a phone 
> call with) all of Red Hat's GDB developers.
>
> I've since been very careful to not revert incorrect patches:
> http://sources.redhat.com/ml/gdb-patches/2003-07/msg00547.html

I recognize that, Andrew, and I was deliberate when I mentioned
that it had not happened more recently.  One reason for bringing
it up was what follows.

> GDB has, since 2000, had global, and not "blanket write" maintainers.

Or, to put it another way, in 2000 you decided that the blanket
write maintainers should be called global maintainers.  You
proposed it on the mailing list, but there was essentially no
discussion (pro or con).
http://sources.redhat.com/ml/gdb/2002-10/msg00185.html

> As
> I explained to david, it was to ensure that everyone, including me did 
> not have the ability to override someones decision.  I corrected the 
> wording to reflect the reality of this clause:
>
> "Note individuals who maintain parts of the debugger need approval to
>  check in changes outside of the immediate domain that they maintain.
>
>  If there is no maintainer for a given domain then the responsibility
>  falls to a global maintainer.
>
>  If there are several maintainers for a given domain then
>  responsibility falls to the first maintainer.  The first maintainer is
>  free to devolve that responsibility among the other maintainers."
>
> The first paragraph (which was added in 2000) is not consistent with 
> "blanket write privs".

I never took the first paragraph to be ABOUT "blanket write privs"
maintainers.  It says "individuals who maintain parts of the debugger".
My reading of that was "persons who are NOT blanket write maintainers".

Especially since the NEXT paragraph goes on to mention (what were
originally called) blanket write privs maintainers.  My reading was
that the two paragraphs were setting out the DIFFERENCE between
the two types of maintainer -- blanket, and not.

Your interpretation completely changes what I understood the
text to say.

> Also, here's the thread I found proposing the clarification on gdb@:
> http://sources.redhat.com/ml/gdb/2002-10/msg00186.html
> and there it sat ...
> http://sources.redhat.com/ml/gdb/2002-11/msg00075.html

There was essentially no response to your proposal.  Stan Shebs
proposed a couple of alternative names including "default
maintainer", and Daniel Berlin suggested "sucker".  No one else
said anything pro or con.

> And here I am posting, and then committing, the patch:
> http://sources.redhat.com/ml/gdb-patches/2003-01/msg00674.html

Well, and that sort of brings us to the current discussion.
Yesterday you posted a list showing that you had made almost
twice as many commits in 2003 as the next ten people *combined*.
You're a very prolific and hard-working engineer, Andrew,
but it's possible that that's not the only explanation
for this.