I’m looking for a self hosted solution to this problem:

I want to create a full text search index from a collection of PDF manuals (text, not images, I don’t care about OCR here). There is a UI to search for text matches in documents, and clicking a search hit opens the PDF scrolled to where the search hit is (bonus points if the search hit is hilighted)

If you host an instance and allow people to sign up and post, what is preventing bad actors from filling your storage with cp and other illegal shit?

Are you just signed up for permanent content moderation assisted by some tooling?

That liability sounds absolutely terrifying to me…

#fediverse