> I claim that, in the big picture, the occasional "false wakeup"
> isn't going to have a measurable effect on performance.

	It seems to me that when things get busy, the writability signals that reading code gets and the readability signals that writing code gets could be a problem. But I'll hold this thought until I do benchmarks. :)

> I think I may have been overly influenced by the Xerox Mesa/Pilot take
> on semaphores, which is that a semaphore is signalled only as a hint
> that some condition has been fulfilled... Perhaps this is not the usual 
> way semaphores are taught in computer science these days, which 
> might explain some of our differences.

	Good point. That definitely wasn't the way I was taught to think about semaphores (in 1986, at UC Berkeley :). They were seen as bet-the-farm guarantees of specific conditions.


-C


--
Craig Latta
composer and computer scientist
craig.latta at netjam.org
www.netjam.org
latta at interval.com
Smalltalkers do: [:it | All with: Class, (And love: it)]