> besides the scientific (or at least practical, at the current level of OCR/machine learning) impossibility I'd hardly say impossible. It's being done, today.

> besides the scientific (or at least practical, at the current level of OCR/machine learning) impossibility I'd hardly say impossible. It's being done, today. See this article:https://www.uni-wuerzburg.de/en/sonstiges/meldungen/detail/artikel/tontafeln/ https://www.uni-wuerzburg.de/en/sonstiges/meldungen/detail/a... or these papers: http://pages.jh.edu/~dighamm/version2/research/2003_10_digital_preservation_of_ancient_cuneiform_tablets_using_3d_scanning.pdf http://pages.jh.edu/~dighamm/version2/research/2003_10_digit... and https://graphics.stanford.edu/papers/cuneiform/cuneiform-300dpi-images.pdf https://graphics.stanford.edu/papers/cuneiform/cuneiform-300... These describe scanning of ancient clay tablets, for archival and information retrieval purposes. Note that the Johns Hopkins research on archiving is from 2003. I suspect the blocker may be money, rather than practicality; these aren't simple flatbed scanners, the tables must be scanned in 3D to recover the full detail required. But, it's an incredibly fascinating area of research to read about. EDIT - So, I did some more reading, and it seems that the Johns Hopkins work I referenced above is part of The Digital Hammurabi Project which is doing exactly this, archiving cuneiform tablets, described here: http://pages.jh.edu/~dighamm/version2/research/2003_03_Museums%20and%20the%20Web%20Conference%202003.pdf http://pages.jh.edu/~dighamm/version2/research/2003_03_Museu... with more information on the projects website: http://www.jhu.edu/digitalhammurabi/ http://www.jhu.edu/digitalhammurabi/