Newsgroups: comp.lang.dylan
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!satisfied.apocalypse.org!news.mathworks.com!uunet!sytex!smcl
From: smcl@sytex.com (Scott McLoughlin)
Subject: Re: QuickSort function?
Message-ID: <s9ig2c1w165w@sytex.com>
Sender: bbs@sytex.com
Organization: Sytex Access Ltd.
References: <199503041449.JAA16121@pecan.chestnut.com>
Date: Sat, 4 Mar 1995 21:08:27 GMT
Lines: 68

dmitry@chestnut.com (Dmitry Nizhegorodov) writes:

> 
> >    >this function will work for any vector containing objects that
> >    > support the comparison operator <.
> >    > Hah!  Have fun doing _that_ in C++!  ;-)
> >    Actually, you can do that in C++ using the Standard Template Library
> > My unstated first point is that you have to invoke an "advanced"
> > feature of C++ to do something straightforward.
> 
> heterogeneous containers and polymorphic operators are "advanced"
> features of C++? I know why the mainstreamers do not read this mailing
> list...
> 

Howdy,

This is a a mischaracterization of what the original poster
meant by "advanced feature of C++" (e.g., uninformed crap).

STL is a an advanced feature of C++ insofar as it is a 
new feature of the standard library.  It is not yet widely
available as a supported component of commercially 
distributed C++ environments. (I believe that a PD version
is available that compiles under some C++ implementations
)  I have read a number of articles about STL (C++ report,
Dr. Dobb's, etc.)  As I understand it, STL _DOES_ use 
"advanced features" of the C++ templates mechanism; features
not yet widely used in practice or supported by many 
compilers.

I've been "out and about" a great deal recently, talking 
to many different programming shops that use C++ in 
the Washington, DC area.  Some, not all, but some were
singing the same tune: Gosh, we thought inheritance +
polymorphic latebinding functions (e.g., virtual functions)
was "the way to go" -- you know, the "OOP" revolution and
all that.  What's all this template stuff for?

They are, to put it simply, disillusioned, angry and 
confused. Alot of this is purely "emotive" reaction --
they staked their reputations, careers and projects
on a much overhyped and immature language, design
methodology and "programming religion".  They suffered
through painful retraining (and much hiring/firing),
tools migration (suffering lots of buggy compilers),
etc.  Now they are NOT HAPPY having "the industry"
pundits and marketing gurus throwing a new bag of
tricks at them.  And this bag of tricks is, to my
eye (and I _like_ programming languages in general),
MUCH MORE COMPLICATED than the classes/inheritance/
virtual functions/objects stuff.

Anyway, I _LIKE_ C and C++. I tend to shy away, though,
from building huge towers of abstractions in C++. I 
use it first as a better C and use additional features
of the language where they are unobtrusive and their
is an obvious and clean match of problem to language
feature.

For the C++ zealots out there, though, get ready for
another rough and painful ride as new and improved
language features come online.

=============================================
Scott McLoughlin
Conscious Computing
=============================================
