From: David Abrahams (dave_at_[hidden])
Date: 2005-04-24 00:16:54
Edward Diener <eddielee_at_[hidden]> writes:
> I heavily agree on this last remark. Mr. Ramey has done a fantastic job
> with serialization, even with some really broken compilers, and it works
> very well with largely conforming C++ compilers. C++ really needs a
> standard serialization library, especially because IMO serialization is
> an absolute necessity of C++ RAD programming environments, and I hope to
> see many more C++ IDE's supporting RAD programming in the future once
> all the pieces are in place for doing it purely with standard C++.
No argument from me, except that how a library does with broken
compilers has zero bearing on its standardizability. For the
standard, you wouldn't want to make any design compromises to
accomodate nonconforming implementations.
-- Dave Abrahams Boost Consulting www.boost-consulting.com