Hey there,
This post isn’t intended to be a suggestion, nor (apologies) is it related improving the Anki stack. However, it is aimed at folks with enough Python and the codebase to answer.
Basically what I’ve got in mind is an Anki-like tool to help drill some Python snippet or trick (e.g. tricky corners of pandas GroupBy, or itertools one-liners). But instead of making a flash card, I was thinking of it closer to writing a unit test at “card creation” that determines the correct output. Then when a “card” is served to the user, they’re given a function outline and the docstring, and need to write the code that passes the test.
The user would be able to submit arbitrarily-many times until their output matches the expected output (or they can forfeit/pass), then they’d be greated with the original solution and a similar prompt of [Again, Hard, Good, Easy] and go from there.
The snippets I have in mind straddle the line between incredibly-useful and edge-case that I’m not writing frequently enough to memorize naturally. The reason I’m posting it here, is because I love the spaced-repetition method (and use it literally daily in my Jeopardy prep), and was wondering if there was a way I could leverage the Anki backend to handle that aspect of this tool.
Specifically, I’d like to know:
- If what I’m proposing sounds compatible with repurposing the existing code
- If so, what the most critical sections of the codebase to learn are
- Do any other projects come to mind, as far as leveraging the spaced-repetition backend that I could study?
- Are there any dumb, glaring development things I hadn’t considered in my proposal?
Thanks in advance for any responses this gets!