About as sexy as an eye exam, but damn, this technology is difficult to get right. So yesterday Google announced the open sourcing of Tesseract OCR, character/text-recognition software it developed back in the 80’s that it claims is better than most of the open source alternatives (I’d believe that) but not quite as good as some of the commercially available technologies (I’d buy that too).
But hmm, isn’t there a lot that could be done with this? Personally, can’t wait until we see this make it’s way into OpenOffice among other places.
Factory Joe 