home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.object:4975 comp.software-eng:5397
- Newsgroups: comp.object,comp.software-eng
- Path: sparky!uunet!spool.mu.edu!enterpoop.mit.edu!bloom-picayune.mit.edu!space.mit.edu!berczuk
- From: berczuk@space.mit.edu (Steve Berczuk)
- Subject: Re: Why and how do organizations select O-O approach to S.E.
- Message-ID: <1993Jan21.172753.29161@athena.mit.edu>
- Sender: news@athena.mit.edu (News system)
- Nntp-Posting-Host: plantagenet.mit.edu
- Organization: MIT Center for Space Research
- References: <1jlic1INNvjn@emx.cc.utexas.edu>
- Date: Thu, 21 Jan 1993 17:27:53 GMT
- Lines: 41
-
- In article <1jlic1INNvjn@emx.cc.utexas.edu>, kalakota@emx.cc.utexas.edu (Ravi Kalakota) writes:
-
- |> 1. What are the factors that were considered and evaluated before you decided
- |> to select an object oriented approach to software engineering?
- |>
- In the 2 projects I have worked on so far, the primary factor in selecting a
- methodology was (yes, this is serious) what methodology the (novice, until the
- time of taking a course in OO) WHICH ONE THE DECISION MAKER SAW
- in the first course that he took on Object-orientation.
-
- Both times there were others in the project who had been building OO systems for at
- least a few years and were at least familiar with some of the OO methodologies, and
- these experienced people would find the choice to be wrong for various reasons.
-
- When I was working at Kodak, someone had taken a course that used teh Wirfs-Brock
- book -Designing object -oriented software, and the general opinion was (Please,
- let's not get into arguments about methodologies-- these are just examples) that
- while is was a great methodology for pedantic purposes, the notation, etc got in
- the way with the expression of design ideas. We still used wirfs brock ( Well, what
- we did was retro-fit our designs to use the right symbols when documents were
- due...)
-
- In another project elsehwhere, someone took a course that used Rumbaugh, (which is a better
- methodology, but one that tended to confuse those new to c++ on our project. It was
- easy to describe systems where the relationships between objects and their methods
- was fuzzy- ) and that is the one that all teh documents were written in.
-
- |> 2. What impact does the selection of object oriented methodology have on
- |> the project sucess, customer satisfaction and learning for future projects?
- |>
- Using the wrong one and being insistent on it can be painful. The methodology
- should be used to facilitate the DESIGN process as WELL as the documentation
- process. Probably the biggest problems arise from taking the phrase "We'll use the
- XXX methodology" to mean "We'll use the symbols in the XXX methodology" rather than
- using the methodology to explore the system.
-
-
- --
- Steve Berczuk berczuk@mit.edu
- MIT Center for Space Research (617) 253-3840
-
-