Newsgroups: comp.object,comp.lang.smalltalk,comp.lang.ada
Path: cantaloupe.srv.cs.cmu.edu!europa.chnt.gtegsc.com!howland.reston.ans.net!EU.net!news2.EUnet.fr!news.fnet.fr!ilog!news
From: davis@ilog.fr (Harley Davis)
Subject: Re: Has C++ had its day?
In-Reply-To: jcm@hgc.edu's message of Mon, 3 Jul 1995 14:11:02 GMT
Message-ID: <DAVIS.95Jul4100319@passy.ilog.fr>
Lines: 32
Sender: news@ilog.fr
Nntp-Posting-Host: passy
Organization: Ilog SA, Gentilly, France
References: <NEWTNews.804415994.23097.sellers@hgc.edu> <3t1j0h$4b4@Starbase.NeoSoft.COM>
	<1995Jul3.141102.7378@merlin.hgc.edu>
Date: 04 Jul 1995 08:03:19 GMT
Xref: glinda.oz.cs.cmu.edu comp.object:34265 comp.lang.smalltalk:25440 comp.lang.ada:32096


In article <1995Jul3.141102.7378@merlin.hgc.edu> jcm@hgc.edu (James McKim) writes:

   >	2.  There is a long list of successful projects using SI, or
   >	    SI+templates.  There is not a long list of industrial
   >	    projects where MI was commonly used (to my knowledge...I
   >	    suppose that's open to argument also :-) (And even if you

   Certainly every Eiffel project used MI. This is not a long list since
   Eiffel's penetration of the market is not (yet) deep. According to
   Robert Martin, nearly every C++ app uses MI. If true this would
   seem to argue that many more industrial projects use MI than not.

I have trouble believing this assertion.  Perhaps it is true in the
projects Martin has seen, but it certainly isn't true among our
customers.  Right now, our own C++ class libraries use single
inheritance exclusively, except in one case in which the MI is hidden
by a preprocessor.  As far as I know, our clients also stick to SI as
much as possible.  The fact is that MI in C++ can often lead to coding
errors, slight performance problems, portability problems, and
conceptual confusion for C++/OO newcomers.  Since most C++ programmers
are still newcomers, it is better to keep things simple for them.

-- Harley Davis
-- 

------------------------------------------------------------------------------
Harley Davis                            net: davis@ilog.fr
Ilog S.A.                               tel: +33 1 46 63 66 66
2 Avenue Gallini, BP 85                fax: +33 1 46 63 15 82
94253 Gentilly Cedex, France            url: http://www.ilog.com/

