Nimble Computer Corporation
16231 Meadow Ridge Way
Encino, CA 91436
(818) 986-1436
FAX: (818) 986-1360

October 15, 1991

ACM Forum
Association for Computing Machinery
11 West 42nd St.
New York, NY 10036

Dear ACM Forum:

I had great difficulty in controlling my mirth while I read the self-congratulatory article "Database Systems: Achievements and Opportunities" in the October, 1991, issue of the Communications, because its authors consider relational databases to be one of the three major achievements of the past two decades. As a designer of commercial manufacturing applications on IBM mainframes in the late 1960's and early 1970's, I can categorically state that relational databases set the commercial data processing industry back at least ten years and wasted many of the billions of dollars that were spent on data processing. With the recent arrival of object-oriented databases, the industry may finally achieve some of the promises which were made 20 years ago about the capabilities of computers to automate and improve organizations.

Biological systems follow the rule "ontogeny recapitulates phylogeny", which states that every higher-level organism goes through a developmental history which mirrors the evolutionary development of the species itself. Data processing systems seem to have followed the same rule in perpetuating the Procrustean bed of the "unit record". Virtually all commercial applications in the 1960's were based on files of fixed-length records of multiple fields, which were selected and merged. Codd's relational theory dressed up these concepts with the trappings of mathematics (wow, we lowly Cobol programmers are now mathematicians!) by calling files relations, records rows, fields domains, and merges joins. To a close approximation, established data processing practise became database theory by simply renaming all of the concepts. Because "algebraic relation theory" was much more respectible than "data processing", database theoreticians could now get tenure at respectible schools whose names did not sound like the "Control Data Institute".

Unfortunately, relational databases performed a task that didn't need doing; e.g., these databases were orders of magnitude slower than the "flat files" they replaced, and they could not begin to handle the requirements of real-time transaction systems. In mathematical parlance, they made trivial problems obviously trivial, but did nothing to solve the really hard data processing problems. In fact, the advent of relational databases made the hard problems harder, because the application engineer now had to convince his non-technical management that the relational database had no clothes.

Why were relational databases such a Procrustean bed? Because organizations, budgets, products, etc., are hierarchical; hierarchies require transitive closures for their "explosions"; and transitive closures cannot be expressed within the classical Codd model using only a finite number of joins (I wrote a paper in 1971 discussing this problem). Perhaps this sounds like 20-20 hindsight, but most manufacturing databases of the late 1960's were of the "Bill of Materials" type, which today would be characterized as "object-oriented". Parts "explosions" and budgets "explosions" were the norm, and these databases could easily handle the complexity of large amounts of CAD-equivalent data. These databases could also respond quickly to "real-time" requests for information, because the data was readily accessible through pointers and hash tables--without performing "joins".

I shudder to think about the large number of man-years that were devoted during the 1970's and 1980's to "optimizing" relational databases to the point where they could remotely compete in the marketplace. It is also a tribute to the power of the universities, that by teaching only relational databases, they could convince an entire generation of computer scientists that relational databases were more appropriate than "ad hoc" databases such as flat files and Bills of Materials.

Computing history will consider the past 20 years as a kind of Dark Ages of commercial data processing in which the religious zealots of the Church of Relationalism managed to hold back progress until a Renaissance rediscovered the Greece and Rome of pointer-based databases. Database research has produced a number of good results, but the relational database is not one of them.


Henry G. Baker, Ph.D.