[glax]

Bookmarked a post about Document OCR that recreates the whole tables with text. Lost to wanacry, cannot find it since. Backup is important, learned the hard way.

[severine]

My guess:

Probabilistic Scraping of Plain Text Tables (edinburghhacklab.com) 92 points by tlarkworthy on Sept 5, 2013 | 12 comments

https://news.ycombinator.com/item?id=6334178#6337769

http://edinburghhacklab.com/2013/09/probabalistic-scraping-o...