The day Hal came

Hal Abelson stopped by the day we inaugurated our SICP reading group.

I wish we could have recorded the session; off the top of my head, though, we touched on a few talking points:

• “Amb” was a McCarthyesque back-tracking apparatus that they stuck “in the guts of the language.”
• They had considered first-class irrationals like pi; the question is, what do you compute on such things?
• Linguistic programming is key; the idea where you think primarily about what sort of language you want to think in, then implement it.
• Should they have done more with the object-oriented stuff in chapter 3?
• SICP is about peeling away successive layers of magic, until you have a full mental model for how (+ 2 2) computes 4.
• There is a lot you can do with functional decomposition.
• Math was a way to “sell” SICP to MIT-frosh; nowadays, calculus seems to scare people away.