Correctness concerns and, among other things, why they are resented

• Authors: Edsger W. Dijkstra
• Venue: Proceedings of the international conference on Reliable software
• Year: 1975

According to Webster's definition of a tutorial: “a paper and esp. a technical paper written to give practical information about a specific subject”, this paper is not worthy of the name “tutorial”, because I would never describe what I intend to do as “giving practical information”. On the contrary: I intend to give as little “practical information” as I possibly can. I am not going to enumerate facts, results and theories, for those you can find—in abundance, I am tempted to add—in the published literature. What I do hope to achieve, however, is helping you to understand and evaluate those facts, results and theories when you encounter them in the literature. I intend to do so by providing you with a historical perspective, be it—in more than one sense—a partial one. It is a lucky circumstance that the amateur historian, like myself, can always come away with very few facts: the fewer the facts, the greater our freedom of interpretation, and it is that freedom that I intend to enjoy. For a short while we have to go back to the scientific climate at the beginning of this century, for there we seem to find the roots of the philosophical opinions that prevailed a few decades later. And these philosophical opinions, in turn, seem to be the source of today's tacit assumptions. By being tacit, these assumptions tend to escape being challenged and that is bad, because they are third-generation off-spring of scientific hopes that, in the mean time, have been shown to have been unjustified, they express everyman's image of a goal that, in the mean time, has been proved to be unattainable.