TITLE: Solid and Relevant AUTHOR: Eugene Wallingford DATE: May 13, 2008 9:15 AM DESC: ----- BODY: I notice a common rhetorical device in many academic arguments. It goes like this. One person makes a claim and offers some evidence. Often, the claim involves doing something new or seeing something in a new way. The next person rebuts the argument with a claim that the old way of doing or seeing things is more "fundamental" -- it is the foundation on which other ways of doing and seeing are built. Oftentimes, the rebuttal comes with no particular supporting evidence, with the claimant relying on many in the discussion to accept the claim prima facie. We might call this The Fundamental Imperative. This device is standard issue in the CS curriculum discussions about object-oriented programming and structured programming in first-year courses. I recently noticed its use on the SIGCSE mailing list, in a discussion of what mathematics courses should be required as part of a CS major. After several folks observed that calculus was being de-emphasized in some CS majors, in favor of more discrete mathematics, one frequent poster declared:

(In a word, computer science is no longer to be considered a hard science.) If we know [the applicants'] school well we may decide to treat them as having solid and relevant math backgrounds, but we will no longer automatically make that assumption.

Often, the conversation ends there; folks don't want to argue against what is accepted as basic, fundamental, good, and true. But someone in this thread had the courage to call out the emperor:

If you want good physicists, then hire people who have calculus. If you want good computer scientists, then hire people who have discrete structures, theory of computation, and program verification. I don't believe that people who are doing computer science are not doing "hard science" just because it is not physics. The world is bigger than that. ... You say "solid and relevant" when you really should be saying "relevant". The math that CS majors take is solid. It may not be immediately relevant to problems [at your company]. That doesn't mean it is not "solid" or "hard science".

I sent this poster a private "thank you". For some reason, people who drop the The Fundamental Imperative into an argument seem to think that it is true absolutely, regardless of context. Sure, there may be students who would benefit from learning to program using a "back to the basics" approach, and there may be CS students for whom calculus will be an essential skill in their professional toolkits. But that's probably not true of all students, and it may well be that the world has changed enough that most students would benefit from different preparation. "The Fundamental Imperative" is a nice formal name for this technique, but I tend to think of it as "if it was good enough for me...", because so often it comes down to old fogies like me projecting our experience onto the future. Both parties in such discussions would do well not to fall victim to their own storytelling. -----