DBMS, April 1997

Abuse of RAD
David Linthicum's article, "The Good, the RAD, and the Ugly" (see DBMS, February 1997, page 22), gives readers the impression that the RAD process is the root of all problems in developing mission-critical, scalable, and reusable applications. But the RAD process isn't the problem; corporate usage of RAD as a term for "quick and dirty" building of systems to meet a time schedule is. Don't blame the technique -- blame the people who are trying to "make a deadline."
When used correctly, RAD provides a set of techniques that result in a strict analysis and design (either object or traditional) that can be used to build and reuse components as part of a large scalable business system. Corporations that train their development staffs to use the RAD techniques in developing software have been successful for years.
RAD is a highly structured systems-building approach that relies on breaking the analysis and design into related components (or objects) and building the messages and objects/components independent of each other. This development is performed within several managed timeboxes and iterations to ensure that each component is being focused on during the development process, as opposed to part of the giant picture. Today's development world is intended to build and buy components and to integrate them into a scalable and reusable system.
Timothy J. McBreen
Technology Management Group Inc.
tmcbreen@compuserve.com
If you read the column closely, you'll see that I agree with most of what you say. RAD is a useful part of the system development life cycle, and we should not "throw out the baby with the bath water." My message was that we can't iterate our way through complex systems development efforts. I'm not at all blaming the technique, only the abuse of it.
-- Dave Linthicum
Is HHCODE New?
In Tom Spitzer's article, "A Database Perspective on GIS, Part 1" (see DBMS, November 1996, page 95), he seems to be crediting Oracle with a "new" invention: Helical Hyperspatial CODE, which, according to descriptions found on Oracle's Web site at www.oracle.com.sg/products/oracle7/Oracle_Universal_Server/press/html/bloor_report.html, sounds a lot like a Quad Tree representation. Quad Trees have been discussed in the computational geometry circles for years and have been used for storing spatial information in databases since at least 1989. In 1990, I used them in my own GIS engine, but the database I was working with did not have the requisite fundamental support, and I found that it was faster to use two indices for x,y than Quad Trees. Therefore, Oracle's real improvement is not the representation (with its snazzy new name) but the fact that the company has added fundamental support for the representation to its database.
Kevin English
kenglish@intelliquest.com
Subscribe to DBMS and Internet Systems -- It's free for qualified readers in the United States
April 1997 Table of Contents | Other Contents | Article Index | Search | Site Index | Home
DBMS and Internet Systems (http://www.dbmsmag.com)
Copyright © 1997 Miller Freeman, Inc. ALL RIGHTS RESERVED
Redistribution without permission is prohibited.
Please send questions or comments to dbms@mfi.com
Updated Tuesday, March 18, 1997