home *** CD-ROM | disk | FTP | other *** search
- From: antles@hpspkla.spk.hp.com (Tom L. Antles)
- Date: Tue, 28 Jul 1992 21:49:48 GMT
- Subject: large OOA models vs. SA/SD models
- Message-ID: <2700005@hpspkla.spk.hp.com>
- Organization: Hewlett Packard Company, Spokane, Wa.
- Path: sparky!uunet!usc!sdd.hp.com!hpscdc!cupnews0.cup.hp.com!news1.boi.hp.com!hp-pcd!hpspkla!antles
- Newsgroups: comp.object
- Lines: 25
-
- I recently took a university class on OOAD in which the technique taught
- was the Rumbaugh method.
-
- The book was very detailed and thorough on the modeling technique itself
- but it falied to address at least one issue in my mind that I am hoping
- someone can shed some light on.
-
- The issue the book seemed to avoid is how to deal with the complexity of
- very large projects
- which result in very large models. For example, in the SA/SD techniques
- there is a decomposition/hierarchy approach used to manage complexity.
- In Rumbaugh's technique it seemed only to be a 'flat' model resulting in
- the whole model being at the same level.
-
- Of course there is the aggregation association, but it doesn't seem the same
- as heirarchy approach in SA/SD. Rumbaugh also uses 'sheets' to segregate
- large models into smaller pieces, is this his only approach to managing
- complexity?
-
- Can anyone comment on the appropriateness of OOAD (and especially
- Rumbaugh's method) to acommodate large and complex models compared to the
- SA/SD technique of hierarchy/decomposition?
-
-
-
-