E.W.Dijkstra Archive: To a new member of The Tuesday Afternoon Club (EWD 683)

When you were admitted to The Tuesday Afternoon Club I wrote to you that I trusted that you would be quick enough in discovering "the rules of the game". After two Tuesdays I feel that we all could benefit from an effort at making those rules more explicit.

Let me first state its purpose, which, right from its inception, has been two-fold. Firstly it is a technique of preventing our area of interest from shrinking, of remaining at least aware and possibly abreast of major developments in computing science. Secondly it is aimed at the teaching and further development of the art of conducting one's research.

For this purpose we conduct mainly two types of activities: the scanning, evaluation, and sometimes the study in depth of selected articles and reports written by others, and the thinking together about the possible solution of what seem seminal problems. (As you will discover, the first activity is, despite the selection, often disappointing, second activity, although not always successful, is often exciting; but in view of our purpose, it would be a serious mistake to drop the first one!) In order that all this works (at least most of the time) the following should be observed and remembered.

The average person prefers to talk about 40% of the time. (Hence the saying: "Two is company, three is a crowd.") We are close to ten people, hence the first thing to learn is to keep your mouth shut most of the time. (For some people this is very difficult, but in that case the training is the more valuable.)

Let the fact that the discussion is ideally open, "free-and-easy" etc. not blur the fact that I started these meetings and am responsible for their continuation and the way in which they are conducted: I am the master of ceremonies, the topic selected and their level of discussion are primarily my responsibility.

In view of the vast amount of printed material that reaches our desks, a major ability that needs to be developed is the fast discovery that something need not be studied in depth, and that an overall picture suffices, a picture of the author's aims and methods, of the (in)adequacy of his use of language, of the nature and amount of the formal labour involved, etc. By its very speed and intrinsic "superficiality" this sifting process, when you see it in action for the first time, may strike you as arbitrary, crude or even unjust. There is, however, some system behind the seeming madness, and when feelings such as "But ho do they dare to discard that article so rapidly?" assail you, realize that in many cases --not always: I am sure that we do make mistakes-- that only means that by now the Club as a whole sifts much more effectively than you yourself are used to.

When trying to solve a problem, most people just try to solve that problem. When The Tuesday Afternoon Club tries to solve a problem, simultaneously two things happen. The problem and its solution, you see, are as a rule not our major concern, as the exercise only serves as a carrier for "the teaching and further development of the art of conducting one's research". In a almost schizophrenic manner we are thinking about the problem and simultaneously observing our ways of thinking and evaluating their effectiveness. I haven't made the count, but but would not be amazed if half of the interjections don't refer to the problem, but to someone's way of thinking or expressing himself. In order to teach each other we correct each other whenever there is reason to do so. (quite a few of my colleagues, I have learned, are such careless language users that they don't appreciate it when I correct their written or spoken Dutch. I can only conclude that they don't care.) When, for instance, when someone shows an example that kills a conjecture, this is accepted, for the counterexample is very effective and convincing tools. When someone --in an effort at what other people call "inductive reasoning"-- shows an example as a representative element of a set, while the set itself is what we should be dealing with, his wisdom is almost certain to be challenged (if he succeeds in showing the example at all). Don't feel a "winner" when having corrected, don't feel a "loser" when having been corrected, and banish all feelings of mutual competition. Never compete and enjoy each improvement, no matter who made it, because mutual competition would destroy the atmosphere in which we can freely correct each other without being offensive. As the Chinese proverb goes "The talented ones talk, the stupid ones argue, but the wise keep silent."

All this is carried by a profound conviction that we should change our thinking habits when more effective ones have been discovered, and that we have tho change our ways of expressing ourselves when more adequate ones are available. That is what we regard as sharpening our tools, knowing that it is impossible to carve a gem with a blunt instrument. Hence our great awareness of patterns of thinking. (When a few weeks ago Martin Rem again blended two concerns fifteen minutes after they had been carefully separated, all of us noticed and thought "It is clear that he has been away for more than a year." and all of us observed --with pleasure-- that half a word still sufficed.) Hence our great linguistical awareness. (You will notice that we have always dictionaries within reach: we really need them!)