home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gatech!psuvax1!psuvm!mkk2
- Organization: Penn State University
- Date: Wed, 22 Jul 1992 11:53:08 EDT
- From: <MKK2@psuvm.psu.edu>
- Message-ID: <92204.115308MKK2@psuvm.psu.edu>
- Newsgroups: comp.object
- Subject: Re: A question of notation
- References: <graham.711774150@galois>
- Lines: 32
-
- >In a "standard" OO language to invoke method M in object
- >O one writes something like,
-
- > a = O.M( .. )
-
- >Now for historical reasons this kind of notation would not
- >fit well with our existing language. The following alternative
- >has been proposed
-
- > a = M( O, ... )
-
- Your "standard" is C++ like (and some other languages). In fact, the
- compiler or pre-processor converts it into something like your second
- alternative.
-
- a = sendmessage(M, O, ...)
-
- to be a little more precise. It shouldn't make any difference, it seems
- to me.
-
- There are other possibilities, too. Smalltalk and ObjC use similar
- notations, something like
-
- a = [O M:...];
-
- You could make up anything that implies messages being sent. How 'bout
-
- a = M(...) => O
-
-
-
- have fun
-