The ACM and standardization

T. B., Jr Steel

1/01/2005
(Original Publish Date: 1964)

The complexities of standardization procedures and the interlocking multiplicity of Sectional Committees, Subcommittees, and Working Groups comprising the formal standardization organizations are undoubtedly bewildering to the uninitiated. The author can assert from experience that it requires about a year of heavy involvement in the process before one begins to have a clear understanding of how it all works. Upon observing this, one is tempted to view the whole complex as a classical example of Parkinsonism; concurrently one develops the wish to cut the structure apart and proceed toward attaining standardization goals in a simple and direct manner. Let me assure you that the obvious, straightforward approach will not work. There are too many interrelated questions to answer and too many conflicts of interest to resolve. It is not clear that the standardization structure which has evolved is the optimum one but, just as in the case of the Congress - which may not be the optimum legislative configuration - it is the best one we know. Further, it seems to work. American Standards in information processing are beginning to appear and, once published, they have not caused any violent or widespread dismay. Despite all this, the computing man in the street generally feels that his opinions are not considered and his technical problems given little, if any, weight. There i s substantial justification for this view. To take one example, the preliminary work in the preparation of proposed American Standards in the area of common programming languages, such as ALGOL, COBOL and FORTRAN, is being done almost entirely by people whose principal occupation is compiler writing. The Subcommittee whose function it is to determine when the work is ready to enter the real standardization mill is largely constituted of people who have not written computer programs on a day-to-day basis for some time. Finally, the Board that ultimately decides whether the community Interest is being served through the promulgation of an American Standard has few members who have ever written a program for a stored program computer. One cannot help but notice the absence of those really affected - the users of common programming languages.