From: Charles D Hixson <charleshixsn@...>
Date: 2006-11-21T09:23:53+09:00
Subject: Re: Ruby vs Java vs c++

David Vallner wrote:
> Chilkat Software wrote:
>   
>> - multiple inheritance
>>     
>
> C++ mixins anyone? Multiple inheritance with data is evil, but as long
> as you only implement pure interfaces or mixins with it, it's
> acceptable. It's not like a language should restrict its users with
> regards to how horribly they can break the code (and brains) of other
> people reusing their creations, eh?
>
>   
When multiple inheritance is done properly, it's not terrible at all.  I 
point you to Eiffel circa 1995.  They had problems, but it WASN'T 
multiple inheritance.  That was one of their strong points.  (Not 
claiming that C++'s multiple inheritance is good.  I don't know it, and 
nobody seems to have anything good to say about it.)  Also consider 
Python's multiple inheritance.  No problems there!!
>> - operator overloading
>>     
>
> Except for contexts where the -semantics- of the operator isn't being
> overloaded, just implemented for a case the language doesn't natively
> supported. Matrix multiplication -is- multiplication and a fine use
> case, using, oh, shall we say, << for something that's not a bitshift,
> like pushing data into an array or a stream (which some C++ texts
> ironically discourage) is indeed bad.
>
> *runs for the hills - FAST*
>
>
> David Vallner
> Meditating upon Pots and Kettles
Operator redefinition can be seriously abused.  It's also something 
that's very powerful when done correctly.  I can see an argument for 
requiring that the *standard* operators not be redefined.  This doesn't 
mean that there shouldn't be user defined operators that CAN be treated 
just as any other (virtual) function.  Perhaps a requirement like "A 
user operator must begin with a '|' character and contain no whitespace 
in it's name.  It may contain any other non-control character, and must 
be delimited by whitespace." would suffice.