Copyright Notice

The following manuscript

EWD 639: The introduction of MAES®

is held in copyright by Springer-Verlag New York.

The manuscript was published as pages 331–333 of

Edsger W. Dijkstra, Selected Writings on Computing: A Personal Perspective, Springer-Verlag, 1982. ISBN 0–387–90652–5.

Reproduced with permission from Springer-Verlag New York. Any further reproduction is strictly prohibited.


EWD 639

The introduction of MAES®.

“Mathematics Inc. proudly announces MAES®, its knowledge-based Mathematical Articles Evaluation System. Developed and tested for internal usage by the World’s Leading Manufacturer of first-class mathematical products, MAES® is an indispensable quality control tool for professional producers and consumers of Twentieth Century Mathematics. Being adopted by most of the International Mathematical Journals to replace their subjective, error-prone, labour-intensive and time-consuming refereeing process, MAES® will be welcomed by the Scientific Community as an enlightening new standard. In order to assist your mathematicians in the writing of the significant articles that give recognition to your Institute, Mathematics Inc. has acquiesced to give in all the leading Scientific Centers of the World our special three-day course “How to increase the MAES®-grade of your publications.” Courses in London, Philadelphia, Moscow, Amsterdam, Grenoble and Djakarta have already been planned, courses in Cambridge, Austin, Oxford, Brussels, Munich, Oslo, New York, Honk Kong and Loempia are in a stage of active preparation. For the larger Scientific Institutes and Universities the friendly specialists of Mathematics Inc. will be happy to give a personalized in-house Course, fully tuned and adapted to the special needs of your organization.”

*         *         *

With the above text MAES® was announced to the World. Its announcement on the 1st of August 1976 gave exactly the stir and response we had anticipated. We could have done so earlier, but the date was carefully chosen in the wake of the Bicentennial 4th of July, Ulster’s 12th of July and the Quatorze Juillet in France, during the summer when the papers have little politics to report on and during which we could take the Universities by surprise (in their Summer Sleep, so to speak). MAES® was the immediate success it deserved to be.

For the coming year institutional customers will be the main target of our promotion campaign; as soon as the Program Committees of the most important conferences and the Editorial Boards of the leading journals have adopted it, the individual career-conscious mathematician is expected to provide for the next market extension. Mathematics Inc. is considering MAES®-grading service bureaus in the world’s major scientific centers. It is still an open question whether they will be run by Mathematics Inc. itself or by one of our educational daughters, such as Instant Inspiration Company or Methodology Mechanics.

I must make a note to ask our PR-man for his advice, although it is not purely a PR-question, it has technical aspects as well. One of the distinctive features of MAES® is the knowledge-based determination of the metrics for the Exceptional Feature Attribute. KOD’s (= Keyword Occurrence Densities) are determined and compared to the statistical average, from which the Exceptional Feature Attribute favours deviations in various chosen directions. By being adopted MAES® will, therefore, not only influence the average style of mathematical publications but also —potentially at least— one of its own evaluation standards.

Within Mathematics Inc. there is a hot debate how to organize the ongoing MAES® adaptation. We have —like in every world-wide organization— the centralists who argue that only the Mathematics Inc. HQ in the Hosanna Building can coordinate and fully synchronize these successive releases; they furthermore argue that this synchronization is absolutely essential if MAES® is going to be accepted as an ISO Standard. On the other hand we have the decentralists who argue that, in view of the drastic differences in publication delays in the various parts of the world, some markets will be more successfully penetrated by a more slowly evaluating quality standard for mathematical publications, while other markets require a much more aggressively adapting grading system. Also they have, undoubtedly, a strong point. It is a hard battle, the issue is undecided yet and as Chairman of the Board I suppose that it will remain so until I have cut the Gordian knot.

Inside the Company we hope that in the meantime the MAES® concept of the Conceptual Paragraphs will somewhat ease the problem in that it makes the eventual MAES®-grade less sensitive to local changes in the preferred KOD deviations. Personally I think —more than its widely advertized Knowledge Base— the concept of the Conceptual Paragraphs will determine the MAES® success. After all, the introduction of Conceptual Paragraphing —which I suggested during its development— has in our internal usage already shown to be the most revolutionary enhancement of our grading techniques. It is so simple! And we have the patents, all over the world, so now I can divulge the secret! Not necessarily contiguous (!) sentences are grouped together in the same Conceptual Paragraph if, by doing so, we can increase the inter and decrease the intra cross reference correlation coefficients of the —of course: normalized— KOD’s. It is as simple as that! By maximizing the one and minimizing the other, MAES® Conceptual Paragraphing automatically arrives at the optimum retrieval modularization which is the basis for all our Quality Attribute Metrics. Besides its intrinsic significance for the destructuralization of the flow of the SSR (= Symbol/Significance Ratio), MAES® Conceptual Paragraphing has the added advantage that the place of the optimum boundaries between the Conceptual Paragraphs is —as a result of the refined averaging procedures— statistically speaking less sensitive to fashions, as may be reflected in the Knowledge Base.

*         *         *

To give you some idea of the effectiveness of MAES®, let me report to you some of our in-house experience during the period 15th April 1976 – 14th of June 1976, during which nearly 4000 of our mathematical articles were subjected to MAES®-grading on an experimental basis. (Our market research people have guaranteed that those 4000 papers were a representative sample from our spring production.) The MAES®-grades given ranged from 0.632 to 0.944, with the exception of two articles with MAES®-grades under 0.3, one from our Loempia subsidiary, and one from the Grenoble one. (The latter exception is the more remarkable because on automatic evaluation systems particularly the French products tend to score very high.) The statistics collected by MAES® during that period are quite interesting. We detected significant metrics for quite a few new negative Quality Attributes.

Production of course material in Hebrew and in Japanese has been held up, pending the decision how it should be affected by the recent discovery of our Department of Mathematical Psychology, that the notion of deconceptualization should be represented by arrows that point backwards. In all other languages, I am happy to say, course material production is working full blast.

And it is high time too! Academic mathematicians are terribly slow in the uptake, and most of them not understood yet that the era of soundly engineered mathematics is already there, today! They fool themselves —and worse: their students!— by closing their eyes to modern scientific techniques for the development and controlled growth of mathematics, they continue teaching their old, ad-hoc ways of doing research. But the wholesale introduction of MAES® will teach the reactionary bastards! At last their private, hobbyist norm will evaporate, for MAES® will force them to adopt the standards of the mathematical industry. Even the departments of pure math will now be forced to produce soundly engineered articles! This, no more and no less, is the Great Service that the introduction of MAES® will render to the World. Thanks to Mathematics Inc. Semper floreat et crescat!


Hosanna Building
Plataanstraat 5
The Netherlands
prof.dr.Edsger W.Dijkstra
Chairman of the Board of
Mathematics Inc.