Diagramming Robust AI
billm at cygnus.com
Mon Mar 30 19:44:30 EST 1998
In article <351FC273.3006 at ix.netcom.com> "Phil Roberts, Jr." <philrob at ix.netcom.com> writes:
>> C++, however, wasn't supposed to be a successor to C -- it was just
>> a little hack job put atop C. The name is actually a geek-pun of classic
>> form, as C++ actually compiled to C, one could argue that the CC compiler
>> simply "incremented C and returns the old value".
>Hack job is right. It added tons of crap, most of which could have
>been handled by standard C structures. And worse, it didn't fix a
>lot of things that really needed fixing. Has anyone here found C++
>to be the godsend it was advertised to be?
I certainly haven't, but on the other hand I think you're being too
harsh on it (or at least harsh for the wrong reasons). Classes added
some much-needed modularity, even though they were implemented rather
poorly and are usually misused. Same goes with overloads. The auto-
matic initialization feature is quite useful. It's still just kludging
extra features onto a neanderthal language, though. It makes me think
of a turbocharged VW Bug.
I think we're as likely to see C++ replaced by something better in the
mainstream software industry as we are to see Intel processors replaced
by something better in the mainstream desktop market, though.
-- Bill Moyer
More information about the Neur-sci