From: Philip Nash (philip.nash_at_[hidden])
Date: 2001-11-20 12:16:03
> > They may be, but even if they are, I'd still rather keep separate
> > in separate parts of library.
> I'm not sure on how to address this conundrum :o(. So far, I wrote an awk
> script that automates the name convention transition and puts namespace
> boost where it has to be. But it would not be elegant at all if
> Loki as part
> of boost would define and use facilities similar to boost's. If anyone has
> ideas on how to proceed, please let me know.
I would support your inclination to assimilate Loki(tus) into the boost
collective natively :-)
After all, Loki will always have some existence outside of boost as Loki. If
it is to become part of boost it wouldn't make sense for it to remain an
individual entity (there's that borg precedent again!).
Note that this is a different issue from keeping libraries autonomous
(libraries in the sense here of individual classes or heirarchies, like
smart_ptr, for example). Although there is some overlap (ie smaller classes
that might be part of the private implementation of Loki).
My 2 groats,
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk