locke@rnms1.paradyne.com (Richard Locke) (08/24/89)
In article <11801@athertn.Atherton.COM> jimb@athertn.UUCP (Jim Burke) writes: >In article <530@ajpo.sei.cmu.edu> eberard@ajpo.sei.cmu.edu (Edward Berard) writes: [this followup directed to OO lang groups, in addition to comp.software-eng] >> Object-Oriented Requirements Analysis >...My personal opinion is that the [Shlaer and Mellor] book [OO Systems >Analysis: Modelling the Wold In Data] appeared to be a repackaging of >the Structured Analysis and Design book previously published by Yourdon, >but with OO buzz words instead. Beware! There seems to be a lot of this going on. I attended an "Object-Oriented Symposium" in June, at which a number of speakers plugged inappropriate old ideas (IMHO) dressed up with the new buzz words. Grady Booch, Bertrand Meyer, and Burt Rubenstein were the exceptions to this!. [disclaimer: I didn't attend every presentation, many were concurrent!] Indeed, too much of the literature suffers from the "OO is almost the same as [old method]" syndrome. Paul Ward's article in IEEE Software (March 89), for example, purports to show the "there is no conflict" between "structured analysis and design" and the use of OO languages. I certainly disagree. [But hey, don't trust me! Decide for yourself! 8^)] (To be fair, many "standard" software engineering concepts apply equally well to OO development and traditional development. However, I feel that successful OO design & implementation requires a different mindset.) Always be aware of an individual's influences. People with a strong investment in structured A & D (i.e., they've written books, sell tools, etc) are the ones likely to tell you that OO techniques are the same as structured techniques. They are salesmen. (Yes, you can still use my CASE tool for OO stuff!) >[The Cox] book has its shortcomings ... he tends to look at the world >as one big smalltalk and/or Objective-C environment. Still, I am hoping >for better, more in depth books on oo design and analysis than the one >mentioned above. OO is old enough now for better books, methodologies, >and case tools. Where are they? I hope you've read Meyer's "Object-oriented Software Construction". Yes his book is influenced by Eiffel, just as the Cox book is language- influenced. Meyer claims it is naive to think OO concepts can be discussed in a completely language independent manner. Despite this, his book does a fine job with generic OO (and generic software engineering) concepts, showing how OO supports modularity, reusability, etc. As for case tools, I'm not convinced there are any *good* case tools for plain old 'C' stuff, much less for OO development 8^) Correct me if I'm wrong. <my opinions only> -- Dick Locke AT&T Paradyne Corporation {uunet,peora}!pdn!locke Mail stop LG-133 Phone: (813) 530-8241 P.O. Box 2826 Largo, FL 34649-2826 USA