Hello M. Freiert, Please read my comments. I hope they will be of use. Regards, Fernando
(as a side note, eventually another goal is to get a significant ammount of our rare books online in an e-book/pdf format for both public and research access, but that's a whole nother nightmare involving long nights with a right angle scanner, and getting ocr software to play with 1700 era german fonts)
You may consider not bothering to use OCR to start with. That is a NIGHTMARE and if you are going to end up trying to get a computer to read Fraktur, Schwabacher and Textualis typefaces I wish you good luck, lots of time and pacience. In our library we don't bother with OCR for digitised documents. We can have reasonably good-quality/small-size images scanned and made into pdfs. Anyway, if you keep the scanned images you will be able to run them trough OCR in a few years time when OCR technology is better.
questions: -how hard will it be to search/compare/semi autonomously add LOC cataloge numbers to known volumes? i know we have a number of privat printings that are probably not catalogued thus far, and will need to be hand numbered, but it'd be nice to not have to look up ALL of them.
I do not run Koha, we have a different library management system but we found that LOC records can be imported via z39.50 and it works very well. LOC (unlike the British Library) offers the full record for download. You can also try other libraries (such as the Danish ones) that also offer their records for free. I guess you may want to download records from German books, so perhaps you could look around German catalogues. Good luck. Fernando Andreu Systems Support Analyst Southampton Solent University +44 (0) 2380 319 689