∂ part of node [[2021-10-22]]


I wonder if I should use [[ntlk]] a bit in the Agora -- currently it's all regexes, all the way down, but some lever of text understanding would go a long way w.r.t. surfacing actually related content.

I'm also wondering if I should do the Wikipedia integration client-side-first or with some support from [[agora server]]. Nice thing of keeping it wholly in the client is that it would keep the server-side rendering path lean (no network calls, only filesystem).