[std-interval] Suggestions for the 2006-09 draft

[Dr John Pryce]

Bill

At 22:18 20/09/06, you wrote:
>The 2006-09 (revision 1) draft is much better than earlier versions. 
>See: 
><http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2006/n2067.pdf> 
>The authors have done a lot of work and should be commended for it.

I liked your comments. I feel Guillaume and colleagues tend to see 
things too much from the implementation viewpoint and you have 
emphasized principles. In particular I am massively in favour of 
"opaqueness" of the data representation. But, being a C++ beginner, I 
can't comment on issues that depend on details of the language spec.

Opaqueness raises the question of how to tie the precision of 
interval<T> to that of T, which I posed in a posting one or two days 
ago. As their text stands, I believe that opaqueness would allow the 
implementations of interval<float>, interval<double> and 
interval<long double> to be all identical.

What do you think?

Regards

John

Dr John and Mrs Kate Pryce
142 Kingshill Rd
Swindon, Wiltshire SN1 4LW
UK
Tel (+44)1793-331062