Judged by the age of dedicated journals, Computing Science is now more than 25 years old. It would even be a mature science if the number of journals devoted to it were a reliable yardstick: by now this number is so large that the practising computing scientist can hardly suppress a shudder at the thought of yet one more journal to browse. Why, then, do I welcome the appearance of "Science of Computer Programming"? The shortest answer is that this one might be worth reading, but that is not the full story.
Being programmable is the automatic computer's outstanding characteristic and, accordingly, programming is computing's central activity: it is the core of the whole topic. Amazingly enough, for more than 25 years we have had no journal specifically devoted to it: no journal, no special interest group, no conference, no nothing! This absence would be defensible if programming were trivial and we could all do it well, but this is not the case. (Who doubts this statement should ask his colleagues to design a program for the Binary Search.) It is not hard to argue that the appearance of "Science of Computer Programming" is overdue.
But I don't think it could have appeared much earlier. Naur was the first to complain to me about the absence of "program literature", i.e. published
programs that could be read and studied like any other scientific article and that would set a standard; Hoare was the second one to voice that complaint to me. This was quite a few years ago, but already then a program of some sophistication was such a compact deposit of its author's intellectual labour, that its presentation in a way that would enlighten the reader was a major problem. Yet, mutual enlightenment is what forges scientific communities!
But a decade of work under the banner of "programming methodology" has changed the situation and justifies some confidence that we have reached the stage in which we can begin to communicate programs and their development for the enlightenment of our colleagues.
"Science of Computer Programming" also has a calling beyond the community of our closest colleagues. It should help ending the current state of affairs in which managers see programming primarily as a management problem because they cannot manage it, in which electronic engineers don't see the problem and in which mathematicians ignore it. "Science of Computer Programming" faces the challenge of showing the world at large that, concerning programming, there is far more than meets the eye. All by itself, that challenge would already be worth this new journal (which, I trust, will not adopt
the silly convention of switching to a smaller type font as soon as program text has to be printed).
For "Science of Computer Programming", the challenges are great; so are the opportunities. My best wishes accompany it: may it live up to our expectations!
| Nuenen | Edsger W. Dijkstra |
Last revised on Mon, 30 Jun 2003.