home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!darwin.sura.net!mips!swrinde!network.ucsd.edu!mvb.saic.com!unogate!beckman.com!dn73!j_ohearn
- From: j_ohearn@dsg4.dse.beckman.com
- Newsgroups: comp.object
- Subject: Re: Help: OOA and OOD Methodologies
- Message-ID: <j_ohearn.712424550@dn73>
- Date: 29 Jul 92 15:42:30 GMT
- References: <1992Jul28.003545.3594@ntmtv>
- Lines: 56
- Nntp-Posting-Host: dn73.dse.beckman.com
-
-
- Just thought I would add my two-cents worth to this thread. My company has
- decided to cautiously test the OO waters. Its inspiration has been a small
- group of software engineers who initiated a grass-roots investigation into OO.
- This object-technology group quickly realized that some formal training was
- required. So we asked for training. We were told that in order for management
- to justify the training costs, we needed to write a proposal. So we did. The
- proposal was crafted and distributed along with a wonderful book-- "Object-
- Oriented Technology: A Manager's Guide" by Dr. David Taylor (anyone and everyone
- interested in OO should read this book!). The distribution list was composed
- of all the senior software department personnel (our "steering committee"). We
- then made a presentation to this group where we gave them a chance to voice their
- concerns. Oddly enough, the majority of concerns either had to do with our soft-
- ware development process or any new technology in general. Approximately one-
- third of the concerns raised had to do with OO in particular. We then were asked
- to draft a written response to these concerns. We did-- the responses were well
- received. Let me give you a quick idea of the contents of the proposal: intro-
- duction, candidate projects, integration issues, risk evaluation (very important!)
- and technical appendices. The technical appendices provided a historical back-
- ground of OO, an overview of methodologies & languages, a list of consultants (for
- training), a reference list (i.e. those companies using OO) and a glossary. If
- I get enough requests, I will post the proposal to the comp.object newsgroup.
-
- Anyways, during the proposal phase, we had to take a close look at the various OO
- methodologies. We studied Booch, Coad-Yourdan, Rumbaugh and Shlaer-Mellor. I
- spent a great deal of time monitoring the comp.object newsgroup looking for opin-
- ions on the subject. I also made contact with a software development manager at
- TI when I saw his project mentioned in "American Programmer" (he recommended using
- a synthesis of Rumbaugh and Booch). We also utilized the monthly newsletter
- "Object-Oriented Strategies", Vol. II, Number 4 and 5 (Cutter Information Corp.);
- object-oriented methodologies were profiled and evaluated in these two issues.
- OOS came to the conclusion that the Rumbaugh method was "the most powerful and
- internally consistent OOM" reviewed. Furthermore, another division of my company
- had already chosen Rumbaugh and were pleased with their choice.
-
- It is probably apparent that we selected the Rumbaugh methodology. However, we plan
- to supplement Rumbaugh with Booch-- we just find the Rumbaugh method easier to read
- and interpret; we also appreciate the sense of scope. We recognize that Grady Booch
- possesses a great deal of meaningful experience and I have found his text to be
- filled with many "pearls" of OO engineering wisdom that we cannot afford to ignore.
- Furthermore, Grady has a great deal to offer software developers who work with real-
- time embedded systems. Our primary emphasis will be Rumbaugh but eventually the
- methodology we will be using will be a synthesis of Booch and Rumbaugh. I strongly
- recommend this approach for everyone since all the OO methodologies are somewhat
- immature.
-
- Feel free to send me email if you have questions about issues pertaining to the
- adoption of OO technology. I will be glad to share my experiences with you.
-
- Thanks for listening.
-
- Jim O'Hearn, Software Engineer
- Beckman Instruments, Inc., Diagnostics Systems Group
- j_ohearn@dsg4.dse.beckman.com
-
- Character, not color, is what matters...
-