Re: [Gdbheads] Re: Feb's patch resolution rate

[Andrew Cagney]

> Well, what about
>     JimB
>     Elena
>     David
> ?

Or:

        Elena
        David
        JimB

?

Remember this lack of commitment has beeing going on for years and years 
and years.   I've shown increadible tollerance towards Jim here, I've 
even covered his ass on occasions (having received direct complaints 
that symtab patches were being intentionally ignored).  At the start of 
'03 (yes a year ago) as part of those private discussions I suggested:

        Elena
        JimB

as at least that way reflected reality.  He declined.  I pointed out 
that he wasn't pulling his weight publically.  Responses have varied 
from calling for Elena's resignation (jointly), calling for all 
maintainers to resign, through to pushing his current agenda.

> > > The symbol table is a critical piece of infrastructure and as such
> > > needs a steady hand, and a long term commitment by developers.  Both
> > > through overhaul, and through a disciplined and diligent review of
> > > patches. What the area does not need is random tugs in differing
> > > directions based on the wim of "maintainer of the day".

Sorry, I wasn't clear here.  I was not refering to Jim.

I was refering to a consequence of giving global maintainers carte
blanche override ability.  We need continuity, even if that does come at
the expense of a patch taking a little bit longer to get reviewed.

That is definitly not to say that a other maintainers have the capacity
to review a patch, rather that other maintainers need to be willing to
trust that the area lead is taking responsibility for the area.

For some area's this largely doesn't matter (Architecture specific code) 
for others it matters a great deal (the easiest example is the CLI where 
we desperately need consistency).

> But not all patches are the same.  Don't confuse being a gdb
> maintainer--somebody who can approve patches, including their
> own--with being responsible for ongoing architectural overhaul of gdb.
> Those responsibilities are not the same.  Different people can
> effectively contribute on both.

I don't think I am.  As part of patch review we need to be willing to be
pro-active.  Too often in such areas a simple patch isn't sufficient,
additional changes may be needed, that in turn requiring effort not only
from the original contributor but also the reviewer.

> > > > > So I didn't see how it was in GDB's
> > > > > interests to remove myself from the list of people able to approve
> > > > > symtab patches.
> >
> > > GDB needs people that _do_ review symtab patches, not people that are
> > > listed as being "able to approve" symtab patches.  There is a very
> > > real difference.
> > >
> > > Denying this is not being honest with either themselves or the GDB
> > > community, they are treating their position as one of power and glory,
> > > and not responsibility.
>
>
> That does not address Jim's issue.  Why is it in gdb's interest for
> Jim to step down as a maintainer?  Sure, there is a difference between
> people who review patches and people able to approve, but how does the
> existence of the latter affect the former?
>
> And what's wrong with a position of power and glory?  People need some
> motivation to work on gdb.  Power and glory is not the worst.

The power and glory need to be deserving.  I'm sorry but 25 patches in a 
year (of which very few were symtab) hardly cuts the cheese.  We have 
here a situtation where clearly (and no one disputes) that Elena is 
doing the work, yet Jim is the one basking in the sunshine.

Jim tries to point to all the work he has done, but none of that, as you 
note, is relevant.  It's all about pulling ones weight when it comes to 
patch review.

While glory can be a motivator, having someone less worthy steal it is 
one hell of a demotivator.  Why Elena even sticks this one out I don't know.

> What specific benefit will gdb gain if Jim steps down?  In what
> specific way will gdb become better?

If Jim is willing to step back, other more fitting reviewers will be
able bubble up and see their moments of glory.  If Jim doesn't step back 
he just demotivates those below him, eventually driving them away.

Projects such as this too easily run the risk of turning in to an old 
boys club (oops, too late).

> Here, I'll give you one for free: if Jim steps down, it would take the
> personality conflict between the two of you out of the realm of gdb,
> and could conceivably smooth relations between the remaining gdb
> maintainers.  Is that an argument you would be prepared to make?

Note that I'm talking here about symtab, not global.  Jim should step 
back as a symtab maintainer.

Andrew