Chris Stith wrote:
> And a huge patch should be implemented, taking hundreds of hours of
> someone's time, without the slightest idea of whether or not anyone
> else even like the implications of the patch? If I'm to spend that
> much of my time on something, I want to be reasonably sure it will
> at least get a look before being tossed aside.
You are right, it would be an enormous change. I don't know if you are
aware of the previous attempts to put threads into perl, which basically
took this route and failed abjectly. The whole point of the discussion is
to try preclude such wastage - a few hundred emails versus hundreds of hours
of work :-)
I also think you are doing people a disservice if you think it would be
tossed aside out of hand.
> Perhaps I should just unsubscribe and let p5p get on without me,
> since I think I could be helpful in implementing a patch but
> don't trust myself to make a major one to the necessary standards
> by myself.
No, please don't do that. Whatever the outcome, this discussion is
invaluable as it will be fed into the perl6 design. The other thing that we
should also remember is that there now appears to be consensus (gasp!) on
the threading architecture - one interpreter per thread. This was not the
case in the past. This has not been a waste of time, and in fact in
comparison to previous discussions has been remarkably polite and brief
(well, by p5p standards at least).
Alan Burlison
