home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.databases.theory:426 comp.databases:6420
- Newsgroups: comp.databases.theory,comp.databases
- Path: sparky!uunet!ide!biesty
- From: biesty@ide.com (Bill Biesty)
- Subject: Re: Different representation of entity-relation diagrams
- Message-ID: <1992Aug28.025207.545@ide.com>
- Sender: bert@ide.com (Bert Beaton)
- Organization: IDE, San Francisco
- References: <1992Aug27.154114.23295@eng.umd.edu>
- Distribution: usa
- Date: Fri, 28 Aug 92 02:52:07 GMT
- Lines: 33
-
- In article <1992Aug27.154114.23295@eng.umd.edu> hyu@eng.umd.edu (Henry K. Yu) writes:
- >Recently, another method was presented to me. And, I don't understand how it
- >works. It specifies the type of relations between each entity and relation,
- >not between entity and entity.
- >
- > +----------+ XXXX /----------\ XXXX +----------+
- > | Entity_1 | -------- | Relation | -------- | Entity_2 |
- > +----------+ \----------/ +----------+
- >
- >Where 'XXXX' can be one of 0:0, 0:1, 1:0, 0:N, N:0, 1:1, 1:N, N:1, N:N.
- >
- >Supposely, someone from MIT came up with this method. Can someone explain
- >the advantage of this method? How about ways to map from the former to the
- >latter method and vice versa?
-
- 0:0? That's productive. Basically this is a general scheme that is useful
- for n-ary relationships. It doesn't buy you anything with binaries. Unless
- the notation is really just a binary notation like Martin and Bachman's
- old notation and the relationship symbol is just semantic info. In that
- case you've got the same situation you described with out the relationship
- diamond/roundtangle/tritangle/whatever.
-
- That's if this is just cardinality. Since there are zeros me thinks that
- this is just introducing existence or optionality. The valid values are:
- 0:1, 0:N, 1:1, and 1:M. A zero means that the entity optionally participates
- in the relationship, a one means it's required to participate. I've only
- seen this reading as existence:cardinality. But the notation may be that
- existence is the closest/farthest to the entity. I.e., On the E_1 side
- you would have 0:1 and on the E2 side 1:0. Not the way I would do it.
-
- --
- Bill Biesty <biestie@aol.com> (I don't read this too often)
- <biesty@ide.com> These are not the opinions of my employer.
-