 |
|
3-Tier Client/Server at Work, by Jeri Edwards, describes n-tier project experiences, all notable large-scale efforts, from a perspective which the author describes as a "50,000-foot view." It's a book written for a hands-on, technologically astute member of the managerial caste, or for a career technologist who has project planning and vendor qualification responsibilities. It is one of a complex of books either authored or coauthored by Jeri Edwards and/or her habitual coauthors, Robert Orfali and Dan Harkey: Client/Server Survival Guide, The Essential Distributed Objects Survival Guide, Instant CORBA, Client/Server Programming with Java and CORBA, and the like.
You may remember the Orfali/Harkey/Edwards series even if you didn't read them, because in several of them the reader is guided through the subject material by Zog the Martian, a little green cartoon man often found on the covers. They are all breadth-first overviews cutting wide swaths across the high-end development landscape, focusing on topics of interest to developers and architects employed in data-intense environments of the sort which often turn to IBM for solutions.
Edwards's latest work focuses on n-tier conversions that succeeded. Named, authorized case histories include the United Kingdom Employment Service's Labour Market System, PeopleSoft's move into n-tier, Wells Fargo Internet banking, Apple Computer's SAP-backended online ordering, MCI data services 3M patient care, European Community unified automobile registration, and others satisfactorily explored.
These are running in the field servicing hundreds of millions of customers and clients. That millions of transaction are served by applications is not new; what's new, and the main thrust of 3-Tier Client/Server at Work, is the emergence of specialized tools within the extant (albeit still fresh) generalized framework of CORBA and other transports, to wit, Object Transaction Monitors (OTMs).
OTMs are the modern attempt to replace or enhance living fossils such as CICS and TPF, hoary but reliable transaction processing systems which underpin, for example, the planetary airline ticketing system. The push to OTM is analogous to the push from relational databases to object databases, and just as fraught.
OTM is a major focus of the book and the particular focus of the first of the three sections of the book. Edwards is an employee of BEA, producer of the Tuxedo OTM, which is treated in especial detail in the book. This is hardly a coincidence, as the author cheerfully and charmingly admits.
The second section of 3-Tier Client/Server at Work treats the case histories mentioned above.
The third section, Zen of 3-Tier, summarizes suggestions offered by the architects whom the author tracked and interviewed in the course of producing the case histories. The summary largely fails, partly because all the good and entertaining observations uttered by the architects were already sprinkled throughout the case histories, and partly because the stuff found in the third section smells like marketing. Either that or the Edwards couldn't leave us abruptly but had run out of news and was reduced to free advice.
In summary, 3-Tier Client/Server at Work, Revised Edition presents an exposition of 3-tier the focus of which is admittedly somewhat self-serving but the content of which is factual, interesting, and constitutes a valuable original contribution to the literature of n-tier implementation. If you are on the brink of converting a legacy enterprise transaction system to n-tier client/sever, you owe it to yourself to pause briefly and read Edward's slender yet acceptable techno-journalistic effort.
-- Jack Woehr
| Readability |
|
| Originality |
|
| Organization |
|
| Accuracy |
|
| Consistency |
|
| Depth |
|
| Timeliness |
|
| Editing |
|
| Design |
|
| Overall Value |
|
Explanation of ERCB rating scale: No stars = unacceptable, 1 Star = marginal, 2 Stars = average, 3 Stars = above average, 4 Stars = exceptional.