Socrates calls Sane to scan, then Gocr to recognize the text. It then pipes the results through whatever pager is set by your $PAGER variable. Gocr isn't all that good, and I found that while Socrates worked with my old HP scanner, it did not with the Umax I now have because it scans in color mode by default, which Gocr can't handle. Fine is a very nice OCR engine, but extremely slow. I may have a look at their web site. It's too bad gocr isn't better, but I don't know of many people with the knowledge of how OCR works who would be willing to produce a free OCR engine.
