The jakarta commons threading proposal seems useful.

I've been stunned and wondering for years why threading problems are not 
seriously tackled at the reusable component level. Perhaps because its hard 
to write high level synchro primitives that are at the same time correct, 
fast & reusable.

Here a suggestion. So far I've been using the public domain util.concurrent 
package

http://gee.cs.oswego.edu/dl/classes/EDU/oswego/cs/dl/util/concurrent/intro.html
http://gee.cs.oswego.edu/dl/cpjslides/util.pdf

I found it highly reliefing having all these countless synchro subtleties 
(just look at the src) encapsulated into correct, efficient and 
easy-to-reuse abstractions.

IMHO it would be useful to have the util.concurrent codebase move over to 
jakarta, in one way or another! Provided [EMAIL PROTECTED], the author, 
would accept this offer, of course. I'm not an authority, but perhaps a 
jakarta commiter may want to try following up on this one.

BTW, Doug Lea has also written "Concurrent Programming in Java: Design 
Principles and Patterns" and is on the expert group of JSR-51 "New I/O APIs 
for the Java Platform".

Reply via email to