From: Kevlin Henney (Kevlin.Henney_at_[hidden])
Date: 1999-08-04 01:21:19
Beman Dawes wrote:
>Kevlin, what did your template approach look like?
That strikes me as better than having a single type with
exact/least/fast/etc types embedded. Not sure exactly why, but my
gut feeling is it would extend better.
Yes, gut feelings count for a lot on this one. These are effectively
inverse traits, in the sense that a trait is being used to look up a
type (rather than the other way around, as for traits) and so naming the
inverse trait directly as the template name separately seemed a more
I also felt this was more extensible and more fine grained than offering
the user a basket of types, especially as the mechanisms employed for
each looked up type are quite different: simple specialisation vs a
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk