In order to be able to remix the work of Neel Doff, the content of the old books needs to be converted in plain text-files. Therefore, I used a bookscanner, I would have loved to use this one, but it needs some repairing; and an OCR-software that turns the image into legible text.
There is a direct relationship between the history of printing techniques and the quality of the final text-file. Could it be an interesting topic for a research project one day? For the moment, comparing the results is already an adventure.
This beautiful edition on heavy textured paper gives the following result:
Jours1943
And this more recent edition from 1974 gives a much more performative character recognition: Jours1974_ocr_tesseract