This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: C++ PATCH: PR 20599 (1/3)
Joe Buck wrote:
> There are certainly risks if we proceed without a finalized standard.
I seem to recall in the days of the first C++ standardization process
that the two main competitors in the IBM PC-compatible marketplace had
divergent responses to a similar problem.
As I recall, Borland followed the cutting edge, resulting in some need
to backtrack on features, or continue to support 'obsolete' language
dialects.
Microsoft, on the other hand, famously lagged the language, insisting
that once they shipped a compiler to accept a given syntax, they were
committed to it for all eternity.
Perhaps some of the old-hands who recall more details can chime in
with their experience from that era?
(Comments about each vendor's subsequent divergence into proprietary
language extensions should be ignored as irrelevant.)