home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.object
- Path: sparky!uunet!mnemosyne.cs.du.edu!nyx!khartig
- From: khartig@nyx.cs.du.edu (Kevin Hartig)
- Subject: Re: OO Analysis & Design Tool Review
- Message-ID: <1992Sep11.145354.7719@mnemosyne.cs.du.edu>
- Sender: usenet@mnemosyne.cs.du.edu (netnews admin account)
- Organization: University of Denver, Dept. of Math & Comp. Sci.
- References: <1992Sep10.192109.10371@projtech.com>
- Date: Fri, 11 Sep 92 14:53:54 GMT
- Lines: 70
-
-
-
- >A topic that was conspicuously missing from Kevin Hartig's posting
- >on OOA/D tool selection was the selection of the method.
- >
- >It is *extremely important* that the selection of the method drive the
- >selection of the tool, and not the other way around.
- >
- >What analysis/design method is this government project using? Was it
- >selected after tool selection?
- >
- >How about other tools supporting OOA/D? Why just these three?
- >Just curious....
- >-------------------------------------
- >Steve Mellor steve@projtech.com
- >Project Technology Training and Consulting using Shlaer-Mellor OOA/RD
- >Berkeley, CA Tel: (510) 845 1484
- >--------------------------------------
-
- Your points are well taken. I agree that the method should drive the tool
- selection. Given the current limited availability of object oriented tools
- which run on specific hardware platforms, software development projects of
- today have to either forgo using tools that support object-oriented methods
- or buy a tool and work with what the get.
-
- So far my experience and perceptions of how government and private industry
- have chosen specific object oriented methods for their projects is as much a
- matter of personal preference and familiarity as it is how a particular method
- applies to a specific problem domain.
-
- For large software projects no one method is sufficient throughout the entire
- analysis and design. It is best to apply appropriate methods to each part with
- an overall project methodology that ties all parts together. Paradigm Plus
- running on PCs support multiple methods and notation.
-
- As a contractor supporting the government, I recommended that the Booch
- method and notation be used for the top level design. Of the two tools
- which implement the Booch method (Rose and Paradigm Plus) I recommended
- Rose for the project. This was becuase of the development environment
- (SPARCstation, UNIX, MOTIF, C++) and the fact that ROSE interfaces directly
- with FrameMaker for documentation. FrameMaker is already being widely used
- at this government facility. As Rose exists today, it will not be sufficient
- for modeling the entire project. Other top down structured and data driven
- models will be used.
-
- The only other OOA/D tools that I know of are:
-
- OOTool - for PCs (shareware) which employ the Coad/Yourdon method and notation,
-
- Object-Oriented Environment - couldn't get a copy of this one
- Fuji Xerox ?
-
- Adagen - didn't evaluate this one either.
- Mark V Systems, Ltd
-
- Parc Place is also supposed to be coming out with an OOA/D tool.
-
- I'm sure there are others I don't know about. Also, many more will be introduced
- during the coming years.
-
- -Kevin
- Hartig@fsl.noaa.gov
-
-
-
-
-
-
-
-
-