Status Updates From How to Design Programs: An ...

«The crucial problem is that beginners make mistakes before they know much of the language, yet programming languages always diagnose these errors as if the programmer already knew the whole language. As a result, diagnosis reports often stump beginners.»

Small lang=better error msgs, hence Elm & Hugs. Should Haskell be extremely granular+modular to improve its error msgs?

«Getting everything right at once is nearly impossible». Iterative refinement: strip inessential, focus on the core, add detail 1by1 & resolve the expanded problem using existing solution as much as possible. They compare it with science, where models are revised until they fit predictions. Interesting comparison.

«we use iterative refinement to state increasingly complex variants of the same problem»—good.

I'm a bit confused: they say functions are in pre-algebra, which really only have equations, inequalities & arith expressions, functions come in Algebra 1. Also, Khan Academy added new content (https://khanacademy.zendesk.com/hc/en...-).

Systematic program design=design recipes+iterative refinement. The writing is so bad, like HEA PowerPoint!! >:(

Lotsa good message under insufferable writing, eg «program design—but not programming—deserves the same role in a liberal-arts education as mathematics and language skills» (that also implies dumb “teach kids to code” is bad; I agree).

https://en.wikipedia.org/wiki/Michael... is a real person :).

Lol, Felleisen is a ‘centrist’ concern troll: https://felleisen.org/matthias/Though....

Current version I'm reading (https://www.htdp.org/2019-02-24/) says: «Released on Saturday, February 23rd, 2019 4:54:46pm». Page numbers will be approximate.

Key phrases: good programming, program design (as opposed to programming), systematic design.