To start wrapping up my overview of how a hypothetical functional JIT would work, today I'll describe I/O, concurrency, and Spectre-resiliance.
In dealing with I/O the JIT I've described has incredible flexibility in how it applies optimizations! (Input) data is stored as pointers to read functions. And output can decide whichever order it wants to most efficiently compute and output the data, even allowing for easy concurrency!
We just published a post with some #FDroid stats from 2019 and a call for support.
Was just looking at #Hy language — it makes no sense to me whatsoever. Doesn't even support `let`, what's the point then?
@climagic I even research problems I think I have a solution for, just to make sure I do not overlook something, consider all edge cases, avoid duplicate work and, most generally, learn from others' mistakes.
I believe that doing so eventually enables me to actually come up with something sound.
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!