It’s been a while but just a note; if a pdf of scanned documents once contained OCRd text, and if the displayed bitmap part was edited to blacken out the pixels in the scan - sometimes the original OCR text is just sitting there inside the pdf. Drop it in a text or hex editor, run strings etc.

The latest Cloudflare and Amazon outages have got me thinking about utilities. We can't build modern software without dependencies. The days of running entirely on the client are over, which I see as a good thing. The cost is that we rely on infrastructure.
1/5