AIX status

[Mark Mitchell]

Ulrich --

  I agree with most of what you say.

  Please refrain from personal attacks on David, or on anyone else.
It's just not helpful.

  I don't care *how* the work gets done.  It doesn't matter to me
whether things go in #ifdefs or in separate files -- that's something
for the maintainer of the code to decide.  I think it's reasonable to
ask David to contribute (since he is most interested in AIX), and to
ask Benjamin to review his work.  And obviously if AIX really can't
support some functionality, then that's just tough luck for the AIX
folks.  But let's see what we can do.

  I fully understand the risk to other systems if the AIX changes were
to be extremely pervasive.  It's reasonable for Benjamin to ask David
to bear that in mind.  

  All I want to ensure is that David gets a chance to do his work, and
that people review his patches just as they would any other patch.

  I'm sorry that everyone has gotten a little bit upset about this.
It would help if Benjamin, David, and Ulrich could all take a deep
breath, and start from scratch.  Don't assume a lot of history, drop
any perception of perceived slights, don't get personal, and go from
there.  It's not going to help to pick on AIX, or on libstdc++'s
non-support for Solaris, or on the time it takes for anyone to respond
to email, or antyhing else.

  David, if you explain the problems in detail, that will help.  I'd
appreciate it if Benjamin and Ulrich would then provide constructive
suggestions, and David can implement them.

  Obviously, I would prefer that y'all work it out without my help,
but I am happy to facilitate if that is helpful in some way.  I am
even willing to have all email sent *only* to me, and to redistribute
it among the other people after appropriate editing if that is
absolutely necessary.

  Thank you,

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com