[nestorD]

Great ! It is not far from one of my dream project : numerising a maximum of old text and let historian do research on them using state of the art tools (that work across synonims and languages) with parameters to restrict by time of publication and localisation obviously.

[toomuchtodo]

Have you considered working with the Internet Archive on this across their corpus? They are open to such work being done. And if some of the material you need isn’t in the archive, let’s get it in there.

[nestorD]

I have not but I am going to file the idea, it would indeed be a good starting point.