C++ Threading Proposal

[Kevlin Henney]

In message <000401c4a6a6$e82160a0$eeb7490c at homefvra93mfsq>, Jim Rogers 
<jimmaureenrogers at att.net> writes
[...]
>If you do maintain a similarity with Java, then you must decide how
>you want to implement a Thread class, and whether or not you want to
>implement anything similar to a Runnable interface. My personal choice
>would be to ignore the Java Runnable interface, since that is simply a Java
>work-around for the lack of multiple inheritance.

This is not the only motivation for this design, although it does also 
satisfy that point of view. Delegation-based approaches (eg having a 
Thread run a Runnable) have found more favour in terms of their design 
strengths over recent years than inheritance-based approaches (eg 
inheriting from Thread).

I certainly would not wish to see a standard C++ threading library 
taking the retrograde step of adopting an inheritance-based model -- a 
generic model based on function objects has been considered the most 
natural approach for quite some time now.

Kevlin
-- 
____________________________________________________________

   Kevlin Henney                   phone:  +44 117 942 2990
   mailto:kevlin at curbralan.com     mobile: +44 7801 073 508
   http://www.curbralan.com        fax:    +44 870 052 2289
   Curbralan: Consultancy + Training + Development + Review
____________________________________________________________