Re: [boost] Review: Config system

From: George A. Heintzelman (georgeh_at_[hidden])
Date: 2001-08-29 14:33:16

> From: "John Maddock" <John_Maddock_at_[hidden]>
> > I think that's about 2:1 in favour of a more pessimistic approach so far
> > (assuming that future compiler versions have all the defects of the
> current
> > version).
> 
> Well then let me cast a vote for "optimism".

> If we assume that future versions have all the same defects, how will we
> know
> that defects have been fixed?  We'll carry on using the workarounds for
> ever.

I also vote for "optimism". Historically, I've seen pessimism cause 
exactly this effect in major packages, such as the Sun CC 5.0 compiler 
vis-a-vis its Roguewave standard library implementation...

Besides, this approach might help encourage faster fixes to 
compiler/library defects, as users shout "<newest version unnamed 
compiler here> *still* doesn't support <feature that other compilers 
have supported for multiple years>"; if we retain the workaround 
support, they might not notice that lack of support.

George Heintzelman
georgeh_at_[hidden]