Bowerbird wrote:

so jon, are you going to take up my challenge?

I didn't know you issued a challenge.

if not, tell me, and i'll do that o.c.r. myself. we're gonna see what accuracy-level we can get on your nice hi-res scans of "my antonia"...

I can only ask my friend so much for Abbyy scanning (he effectively pays a per page fee for using Abbyy), and your request does not qualify as anything important enough for me to spend "capital" on. So feel free to go ahead and do what you will with the "My Antonia" scans. That's why they're online (I will need to make some sort of usage statement for them, maybe a Creative Commons license -- but the intent is for the whole world to have ready access to them.) I'm curious to know how well various OCR packages will perform on "My Antonia" since the XHTML version is very accurate to the original -- so it can form sort of a test base. Of course, if you or anyone else finds an error in the XHTML version as a result of the OCR test, I'll appreciate being informed so I can make the correction. Others here who use their own OCR package, feel free to test it out on the My Antonia scans. Go to: http://www.openreader.org/myantonia/ Jon (btw, I plan to soon scan my original edition of Burton's "Kama Sutra", and that will be a much greater challenge to any OCR package, even if it were new, because of very small print, overall poor typesetting, and poor print quality.)