home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.object
- Path: sparky!uunet!walter!porthos!dancer!haim
- From: haim@dancer.uucp (24103-kilov)
- Subject: Re: A question of notation
- Reply-To: haim@dancer.UUCP ()
- Organization: Bellcore, Livingston, NJ
- Date: Fri, 24 Jul 92 17:02:19 GMT
- Message-ID: <1992Jul24.170219.28020@porthos.cc.bellcore.com>
- References: <MOSS.92Jul22103537@ibis.cs.umass.edu> <1992Jul23.182929.7473@dec254.uucp>
- Sender: netnews@porthos.cc.bellcore.com (USENET System Software)
- Lines: 42
-
- In article <1992Jul23.182929.7473@dec254.uucp> hall@dec254.uucp (Phil Hall) writes:
- >
- >I might add some comments about these concerns. While the difference between
- >O.M (...) and M (O ...) is really only syntactic, in practicality, the
- >two notations often have very different interpretations.
- >
- >The O.M (...) notation obviously distinguishes the object O. This form
- >clearly implies that the method M must belong to the class or subclass of
- >O. This is quite different from just passing O as an argument, which
- >does not suggest M associates with any class at all (at least not in
- >a class definition).
- >
- >This distinction seems to be pretty significant in a persons perception
- >of what object-oriented-ness is. It is for me.
- >
- >>
- >> J. Eliot B. Moss, Associate Professor
- >
- >Phil
- >-------------
- >Philip W. Hall Internet: hallpwcd@ucunix.san.uc.edu
- >Cincinnati Milacron, Dept. 82G
- >Advanced Systems Division
- >4701 Marburg Avenue, Cincinnati, Ohio 45209
-
-
- Exactly!!!
-
- The Object Data Management Reference Model (an ANSI OODBTG document)
- specifically stresses that there exists a "classical" object model, with
- operations that do have distinguished recipients, and a "generalized"
- object model with operations that do not have distinguished recipients.
- In information modeling, operations often (almost always) do not have
- distinguished recipients, i.e., are jointly owned by several associated
- objects (see our recent publication "The Framework: a disciplined approach
- to analysis" and other publications from this source).
-
- Hope this helps.
-
- -Haim Kilov
- haim@bcr.cc.bellcore.com or
- haim@dancer.cc.bellcore.com
-