Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!decwrl!infopiz!lupine!rfg
From: rfg@NCD.COM (Ron Guilmette)
Newsgroups: comp.std.c++
Subject: Re: operator dot?
Message-ID: <4627@lupine.NCD.COM>
Date: 25 Mar 91 08:11:52 GMT
References: <70904@microsoft.UUCP> <4327@lupine.NCD.COM> <12036@pasteur.Berkeley.EDU>
Organization: Network Computing Devices, Inc., Mt. View, CA
Lines: 31

In article <12036@pasteur.Berkeley.EDU> jbuck@galileo.berkeley.edu (Joe Buck) writes:
+In article <4327@lupine.NCD.COM>, rfg@NCD.COM (Ron Guilmette) writes:
+|> In article <70904@microsoft.UUCP> jimad@microsoft.UUCP (Jim ADCOCK) writes:
+|> >Does anyone know what happened to consideration of overloadable operator dot?
+|> 
+|> We can only hope that people realized that it was a horrible idea, and
+|> that it then died a quiet and lonely death.
+
+Well, Ron, educate me.  Why is it a horrible idea?  I claim that, for
+many problems, a "smart reference" object is superior to a "smart pointer"
+object.

First off, the idea of overloading "." is horrible because "." is not
an operator.

Even if it were an operator (which it obviously isn't) there is no
evidence (that I have seen) that "smart references" add significantly
to the expressive power of the language.  I can imagine that they
might allow you to avoid having to type in the occasional `*' (to
dereference a "smart pointer") but that seems like an awfully small
gain when measured against the added complexity in the language.

Remember that each additional increment of complexity has to be paid for
many times over, both when people who are new to the language are
attempting to learn it, and when people who are implementing the
language are doing what they do.
-- 

// Ron ("Shoot From The Hip") Guilmette
// Internet: rfg@ncd.com      uucp: ...uunet!lupine!rfg
// New motto:  If it ain't broke, try using a bigger hammer.