Dijkstra on YAGNI

Via Douglas Crockford,

E.W. Dijkstra talks about "superfluous features" while reviewing the IBM 1620 Data Processing System:

Before going on I should like to explain why I may have objections to "superfluous features".  Suppose that a machine contains a certain feature and that I can show, for instance, that it is impossible to use it intelligently or that its use gives rise to undesirable programming conventions; suppose furthermore that the defender of the design agrees to my objections but defends the feature by pointing out that, if I do not like the feature, I do not need to use it, implying that no harm can be done by something "extra".  In that stage of the discussion I shall stress that the design would have been better without the feature under discussion.  If it is impossible to use it intelligently every effort to do so is spoilt and the programmer would have been better of without it.  If its use gives rise to undesirable programming conventions, also in that case the programmer had better ignore the feature completely.

There's also quite a funny line in the concluding remarks (emphasis mine):

As the reader will understand, my recent study of the IBM 1620 has been a shocking experience: I knew that it was a rather small machine but I had never suspected that it would embody so many basic blunders.  Personally, I cannot undergo such an experience without asking myself what its morals are.