This is cool! I like how it makes a natural link between how characters are composed of semantic parts ( radicals ), and how the non-semantic RRPL maps / abstra

This is cool! I like how it makes a natural link between how characters are composed of semantic parts ( radicals ), and how the non-semantic RRPL maps / abstracts that, maybe exposing the semantic structure with a language that doesn't have to be aware of semantics. Very cool. If the author is reading / or if I just missed it in the docs, were these 5000+ descriptions generated automatically somehow or by hand? Either way it's impressive