PP>> While it is true that C++ is *way* off from being a true
 PP>> object-oriented programming language,
 JC> No it isn't.  Saying otherwise shows only that you don't know what the
 JC> phrase actually means, or that you're ignoring what it means.
Hmm, if you refer to the difference (pointed out later (I'm not natively 
english-speaking, fault of language)) between "true" and "pure" OO language, 
I agree.
 JC> Let's true to keep terminology straight here: there's not necessarily
 JC> any correlation between being a true OO language, and being close to
 JC> paradise, or even a usable language at all.  If you wanted to, you
 JC> could create a language that was undeniably completely object
 JC> oriented, yet was completely unusable.
Agreed too...
 JC> Truly object oriented or not is a purely objective question, and has
 JC> virtually nothing to do with whether a language is particularly good,
 JC> bad or indifferent.  Opinions as to the quality of a particular
 JC> version of Pascal really don't belong here.
Yes, but I referred to Pascal because of a few things, like that the original 
author (the one I was replying to) was saying Borland Pascal was a better OO 
language (which is probably false), and a few other things, like C/C++ 
programmers looking down on Pascal because "nothing can be made of it" or 
something like that *AND* Borland Pascal programmers that think of this 
language as the greatest thing on Earth. I saw people here complain that the 
Borland VCL was probably weak and lacking many C++ "advantages" and 
"efficience" because it is written in Pascal. I agree on them that the fact 
of using a Pascal library with a C++ isn't a good idea because of difference 
of language, but another fact is that Borland has the same code generator for 
both their C++ and Delphi compiler and that in some cases the Delphi compiler 
showed almost magical optimization. But C++ is strong. But it is such a hell 
of a patch-up on C!
 JC> Objective C is a perfectly fine language, but has a large set of
 JC> problems of its own.  I'm not going to debate (nor allow others here
 JC> to debate) the relative merits of Objective C and C++ (or Eiffel,
 JC> Sather, C+@, or the myriad other OO languages based to some degree or
 JC> other on C) but I will point out one last time that the question of
 JC> whether a language is truly object oriented or not is open to purely
 JC> objective answers.  In the cases of each of the languages listed
 JC> above, there's absolutely NO question that the answer is YES, it is a
 JC> true OO language. That's completely independent of how good the
 JC> language is.  It's possible to define a pretty decent object based
 JC> language that's definitely NOT object oriented.  It's even possible to
 JC> define a pretty decent purely procedural language that bears nearly no
 JC> resemeblence to an object oriented language at all.
I was thinking more about *pure* OO than *true* OO. Pseudo-OO like the one in 
Visual dBase and Visual Basic is laughable at best. Around here we call this 
GOP (Graphically Oriented Programming)... (note the pun that says that "GOP 
is what makes our programs GUI" ;-) ) I also like to call it "widget-oriented 
programming"...
There is *many* questionable things about C++... Like for example, (regarding 
pure virtual functions) what happens if I put something else than 0 after the 
=??? Like in "virtual void fn() = 42"? Also, *not* having multiple 
constructors, *having* multiple inheritance...
 JC> Finally, there's no question that it's possible to define
 JC> languages that are absolutely unusuable, but still really and
 JC> truly object oriented.
Oh *yes*.
Pierre
... Documentation - The worst part of programming.
--- FMail/Win32 1.20
---------------