The fainting surgeon

This morning I came across a funny comment.

On a post that highlights a subtle implication of the use of Java and as a teaching tool (picked up on another article, this one about Scheme), a guy named Stephen Fraser dropped this: "As I see it, a software engineer who hasn’t worked through SICP with Scheme, a basic editor and command line, is like a surgeon who has never dissected a frog and faints at the sight of blood."

I happen to agree with that. It's not as much a virtue of Scheme and SICP (both outstanding teaching tools), but a major conceptual failure of IDEs. By shielding the programmer from the complexities (and we can endlessly argue whether those are needed or not) of the typical Java framework or build tool, they make that complexity tolerable and thus create fertile ground for adding new complexities on top of it the already high pile of complexity, a layer that will, eventually, be mitigated by the next iteration of the IDE (or, if this specific layer of complexity fails to get much traction, by an IDE plugin).

And so, as we add layer upon layer of complexity, many of today's software engineers grow accustomed to be so removed from whatever they are actually doing (or, more precisely, what their IDEs are doing for them) that they risk being unable not only to see, but to accurately grasp the full depth of the stack they are standing upon. They become surgeons who can't say where the patient's lungs are located or what they do.