[NTLUG:Discuss] Evil GCC-2.96

[cbbrowne@ntlug.org]

Will Senn wrote
> Christopher wrote:
>      To be a wag, the kernel isn't written in C++, nor is XFree86,
>      nor Emacs, nor TeX, nor Perl, nor Python, nor Apache, nor GLIBC.
>      If none of those important Linux-related programs are written in C++,
>      why would anyone consider C++ to be of any importance?  :-)
> 
> Sheesh! How provincial :)

I'm from Ontario, which is a province.  How could I be other than
provincial...  :-)

> I once had a professor that told me that all object oriented programming
> was just syntactical sugar.  He said, "Oh, yeah. We've been following
> Object Oriented Programming practices for years.  We use typedefs,
> ifdefines,
> structs and various other mechanisms to achieve exactly the same results."
> When I put forth the idea that Procedural Programming and high level
> languages
> in general were just syntactical sugar, for some reason he objected.  I then
> explained that we Assembly Language hacks had been following the same
> rigorous
> procedural principals for years...  Then something hit me - what if one were
> to
> program in machine code - could they manage reuseable object oriented
> programming?
> Absolutely!  It would probably be easier with a tiny bit of syntactical
> sugar though.
> C++ may not be as little used as you have suggested, but it is not as wide
> spread
> as it should be either. OOP is a very powerful paradigm, I think that it has
> the
> potential to replace most non-oop methodologies.  C++ is too complicated to
> be
> the vehicle though.  Java - too slow!  Smalltalk, hmmm - may still come out
> on
> top.  Maybe a language is too small a medium for OO to shine.  Think Zope.
> A system
> of objects, where the language is a core part of a larger system of objects.
> 
> Anyway, just thinkin' out loud. To your point of why anyone would consider
> C++ to
> be of any importance?  Retorical though it was, C++ was the biggest OO play
> in the
> genesis of OO, a revolution in programming culture.

The _big_ problem (to my mind) with C++ is that the standard document
is considerably longer than that for other standardized OO languages
like Ada or Common Lisp, even though there are substantial things that
those standards include which C++ does not.  [e.g. - stuff like STL
must be _added_ to the already daunting complexity in order to get
similar functionality to the other guys...]

If the language definition was _so_ controversial as to require so much
description, that means it's got _big_ hairy edges, and the "fighting"
over which version of G++ should get released when, and the previous
longstanding serious deficiencies of G++ all come as a consequence of
this.
-- 
Christopher Browne
<http://www.ntlug.org/~cbbrowne/resume.html>
cbbrowne at acm.org
(613) 225-3689