RE: [boost] future of boost.threads

From: Jeff Garland (jeff_at_[hidden])
Date: 2004-02-05 11:35:37

On Thu, 5 Feb 2004 09:20:09 -0500, Bennett, Patrick wrote
> Honestly, I'd suggest you just use ACE for threading/synchronization
> primitives.  It's available for a large number of platforms and is
> heavily used and frequently updated.

All true.

> I like Boost a lot, but I think some people are trying to make Boost
> more than (IMO) it should be.  People discussing efficient io 
> dispatch techniques for example would be better served to just use 
> ACE, which already does many of these things extremely well.
> (http://www.cs.wustl.edu/~schmidt/ACE.html)

I say that it depends.  If you need networking, threading, event loops, etc, 
in C++ right now then yes, by all means you need to look at ACE.  That said, 
ACE is a huge library that is reasonably difficult to learn and that suffers 
from it's own legacy.  That legacy includes a long list of platforms, 
compilers, and standard libraries that don't even come close to standard 
C++.  As a result ACE's code base is a macro nightmare and it doesn't utilize 
the standard library as much as I'd like to see (it was around pre-standard 
library). Right now boost threads is a minimalist solution without all the 
legacy.  So if I was trying to do some basic thread management I'd be looking 
at Boost first and ACE second.

Further, even though I have used ACE sucessfully for years, I'd like to see 
Boost libraries for networking and threading continue to move into ground 
where I currently use ACE.  The reason is simply that I believe choice and 
diversity is good.  And if the standards committee is ever going to look at a 
networking library I'd like to have a Boost, more minimalist solution on the 
table for discussion because I don't see how ACE can fill that bill....

Jeff