home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.object
- Path: sparky!uunet!news.univie.ac.at!scsing.switch.ch!univ-lyon1.fr!ghost.dsi.unimi.it!rpi!usc!howland.reston.ans.net!bogus.sura.net!darwin.sura.net!mlb.semi.harris.com!dw3g!smccoy
- From: smccoy@dw3g.ess.harris.com (Scott McCoy)
- Subject: Re: relation betweem OOM and IDEFo and EXPRESS language
- References: <1993Jan27.124431.16827@lut.ac.uk>
- Date: Wed, 27 Jan 1993 17:01:15 GMT
- Nntp-Posting-Host: dw3g.ess.harris.com
- Organization: Harris Corporation Information Systems Division
- Sender: smccoy@dw3g (Cheshire Cat)
- Message-ID: <1993Jan27.170115.264@mlb.semi.harris.com>
- Lines: 33
-
- In article <1993Jan27.124431.16827@lut.ac.uk>, A.H.Al_ashaab@lut.ac.uk (AH
- Al_ashaab) writes:
- |>
- |> OR Iam wrong and using any OOM is enough to do all our work from the
- |> beginning
- |>
-
- IMHO, you're wrong. :-) IDEF0 is a functionally oriented, hierarchically
- decomposed modeling notation. It typically is used to capture the
- business or manufacturing practices of an organization.
-
- I suppose you could use IDEF0 to capture/specify the sequential behavior
- of a software system, but since there are no data stores or the like
- defined in IDEF0, you might as well just use DFD's (yucko). So if the
- OOM of choice provides a behavioral specification notation, why bother
- with IDEF0 at that level of detail?
-
- IDEF1X does provide a good data modeling notation. But, if the OOM
- of your provides a data model technique, then why cloud the issue with
- yet-another-diagram?
-
- If you really want to mix IDEF0/1X and an OOM, you could:
-
- 1. Analyze the business practices using IDEF0
- 2. Build a corporate data model using IDEF1X
- 3. Determine which portion of the business you want to automate
- 4. Begin OOA/OOD, using the IDEF0/1X models for context
-
- Again, just MHO.
-
- --
- Scott McCoy Harris ISD Opinions expressed are my own.
- Staff Eng - SW Internet: smccoy@dw3g.ess.harris.com
-