From: Aleksey Gurtovoy (agurtovoy_at_[hidden])
Date: 2002-10-30 21:18:46
David Abrahams wrote:
> > It's in the August 2002 issue, (not on line) and as I recall didn't
> > really work for lambda. I'll get my copy tonight and post a critique
> > if anyone really cares.
>
> I think there's a lot we'd want to do differently (e.g. use MPL), even
> if it does work for lambda (which is not clear because of the
> reference-stripping behavior). However, the basic technique looks
> pretty powerful.
I believe the typeof implementation John R. Bandela posted quite long time
ago (1 year?) does pretty much the same thing. From what I remember, it
represents composite types using a type sequence instead of encoding them as
numbers, and that's probably the most significant difference.
Aleksey