NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/object/5112

Internet Message Format | 1993-01-28 | 1.7 KB

Xref: sparky comp.object:5112 comp.software-eng:5506
Newsgroups: comp.object,comp.software-eng
Path: sparky!uunet!rational.com!thor!rmartin
From: [e-mail] (Bob Martin)
Subject: Re: How About Using Bootch for Analysis?
Message-ID: <rmartin.728267948@thor>
Sender: [e-mail]
Organization: Rational
References: <[e-mail]>  <[e-mail]>
Date: Fri, 29 Jan 1993 00:39:08 GMT
Lines: 27

[e-mail] (G.K. Khalsa) writes:

>The one other area where Booch's notation may be weak is in the
>ability to develop high level views of the model.  For large systems,
>it is not practical to represent a specification as a flat, huge
>set of classes/objects and their interactions.  Coad/Yourdon and
>Embley, Kurtz and Woodfield are two texts that describe ways to
>present a model in manageable chunks.

I beg to differ.  Booch's notation has a very powerful method for
dividing a design into "chunks".  In fact, it is the richest
mechanisms I have seen to date.

Booch recommends that you break your design up into aggregates of
classes called "Class Categories".  The notation supports these
entities and relationships between them.  Class categories may be
related by "Visibility" which means that some of the classes in one
category make use of some of the classes in the other.  Booch's
notation also allows you to specify which classes are visible outside
a category and which are private to the category.  This is powerful
stuff.

--
Robert Martin       | Design Consulting    | Training courses offered:
R. C. M. Consulting | [e-mail] |  Object Oriented Analysis
2080 Cranbrook Rd.  | Tel: [phone]  |  Object Oriented Design
Green Oaks, Il 60048| Fax: [phone]  |  C++