[jt2190]

An interesting, related thing I’ve noticed is in many learn to program tutorials and exercises there’s a failure to be explicit about when we are creating a model, and that modeling is a skill, and that skilled model-builders first and foremost create a model to solve a problem. I’ve seen too many cases where students are left on their own to flounder about deciding if cars should be composed of four wheels and an engine, and what about the doors? etc.

Another flavor that traps experts in endless, pointless debate are taxonomies. Edit: See “The narcissism of small differences”, https://en.m.wikipedia.org/wiki/Narcissism_of_small_differen...

[gopher_space]

I've been trying to find the name of this film for ages, but it's a scientist explaining taxonomy by lining up a bunch of pencils sharpened to different heights in order of height and then arbitrarily dividing them into sections.

Very much "the divisions are important but also decided on by some random guy, so keep that in mind."

[BLKNSLVR]

The best introduction to taxonomies: https://en.m.wikipedia.org/wiki/Celestial_Emporium_of_Benevo...

Jorge Borges also has written something directly related to 'the map is not the terrain'

[theoa]

https://en.wikipedia.org/wiki/On_Exactitude_in_Science

[BLKNSLVR]

And wrapping irony around itself in Borges' special way, parts of the map actually become the terrain:

"tattered Fragments of the Map are still to be found, Sheltering an occasional Beast or beggar..."

[lionkor]

OOP can be a minefield in this sense, yeah. It needs to be taught correctly, and even then, it brings the problem domain into the solution, which makes it difficult to reuse.