> PS - Just a hunch, but I think that the type of content one is likely to
> see in Peter Daniels' ( :) ) works is not going to be very amenable to
> any type of existing OCR.
In the early days of home OCR, there allegedly were programs that had to
be "trained" to recognize characters. Then they were "improved" so they
could read roman alphabets right out of the box, and there went all the
flexibility.
--
Peter T. Daniels grammatim@...