[1046] in java-interest

home help back first fref pref prev next nref lref last post

Re: Overloading & New

daemon@ATHENA.MIT.EDU (vizbiz.com!green@vizbiz.uucp)
Thu Aug 17 18:35:23 1995

Date: Thu, 17 Aug 1995 14:55:03 -0400
From: vizbiz.com!green@vizbiz.uucp
To: parrt@parr-research.com
Cc: francois@coplanet.fr, gary@Intrepid.COM, java-interest@java.sun.com,
        johnm@alumni.EECS.Berkeley.EDU
In-Reply-To: <199508162007.NAA06322@lonewolf.parr-research.com> (message from
	Terence John Parr on Wed, 16 Aug 1995 16:07:32 -0400)

>>>>> "Terence" == Terence John Parr <parrt@parr-research.com> writes:

[Time to stop lurking!]

    Terence> Uh oh...not templates.  Templates are needed in C++ for
    Terence> really two reason:

    Terence> (i) lack of fully dynamic messages
    Terence> (ii) interfaces.

In practice, you may want to use templates for performance reasons. 
Dynamic messages and interfaces both incur runtime costs (which may be
eliminated in *some* cases with clever optimizations). Oh - and
they can be good for strong static type checking.

    Terence> PLEASE: All C++ people please go back to C++ and quit
    Terence> trying to turn Java into an interpreted C++.  Gawd!!!
    Terence> This is ridiculous!  Templates are a total nightmare to
    Terence> implement/get-right in C++.

True - but I still would like to see a simple parameterized type
mechanism added to Java. I have a lot of experience in this area and
have found them easy to implement and extremely useful. My
understanding is that the implementation side of Java arrays lay the
foundation for such a mechanism. 

    Terence> 	If you haven't learned your lesson with C++, you
    Terence> deserve each other.

Oh Ter - that hurts! :-)

Anthony

-- 
Anthony T. Green                                           Visible Decisions 
voice: 416-941-5776 fax: 416-864-3901                      Toronto, Canada
http://www.io.org/~vizbiz
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com

home help back first fref pref prev next nref lref last post