home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.object
- Path: sparky!uunet!walter!porthos!dancer!haim
- From: haim@dancer.cc.bellcore.com (kilov,haim)
- Subject: Re: Entity Relationships -> Objects
- Organization: Bellcore, Livingston, NJ
- Date: Mon, 23 Nov 92 17:59:33 GMT
- Message-ID: <1992Nov23.175933.14875@porthos.cc.bellcore.com>
- References: <BxyzJL.K9L@fig.citib.com>
- Sender: netnews@porthos.cc.bellcore.com (USENET System Software)
- Lines: 21
-
- There exists no inherent contradiction between ER and OO approaches. The
- building block library that is reused in information modeling certainly
- includes associations. These associations are defined by operations jointly
- owned by their components. Naturally, there exist different kinds of such
- associations that may be defined precisely and explicitly, in an
- implementation-independent manner (OO). They are reused in various applications.
- Certainly, various extended ER models provide a good starting point for
- formalizing the informal descriptions of these associations.
-
- We are doing this in Bellcore. This has been described in some of our
- documents (I had mentioned in comp.object about "The framework: a disciplined
- approach to analysis", for example), and also in outside publications
- (TOOLS '91, Workshop 4 at OOPSLA-92, OO(D)A Workshop at OOPSLA-91, DSOM '92,
- TINA '92, etc., to name a few). Some of these ideas have been included also
- in the General Relationship Model draft ISO standard for management
- information base.
-
- Hope this helps.
-
- -Haim Kilov
- haim@bcr.cc.bellcore.com
-