In recent years automated text recognition has developed impressively, and now it seems to become applicable to digitized archives. A computer can actually be taught to read and also 17th and 18th century Dutch in scanned VOC archival sources. A breakthrough is in reach in the near future. Texts of old manuscripts made digitally searchable can assist historians and other researchers in disclosing archives.

Not all archives are in prefect condition. Because of archival damage text recognition can be difficult.
Marco Roling (advisor of The Corts Foundation) has conducted research recently on the application of text recognition on archives with damage, and has looked at the effects of ink corrosion and discoloration. This article also takes a first step towards measuring archival damage and the possibilities of improving scans digitally.

The research is public and free, and can be downloaded here >>>