I've never played with any of the commercial OCR stuff, so I can't comment on how good it is, but I can say this with certainty. The free stuff isn't usable. I'd call it experimental at best, but my experience is, that neither ocrad nor gocr could recognize a printout from an inkjet printer without dozens of errors. Of course, there aren't a lot of programmers who are knowledgeable about how OCR works willing to write free software, plus, it's a very small market. So, I'm not holding my breath. I certainly don't understand much about how OCR works, so I don't think I'd be of much use. It's really a bummer, and the commercial stuff looks to be a real bear to get working, especially if you happen to use a Linux distro other than Redhat/Fedora Core. That's my biggest hold-up as well and why I still have Windows XP. That, and the numerous web pages which use Java Script in some way just to make sure you are using Microsoft's browser. Mozilla will hopefully eventually level the playing field here, but it's proceeding rather slowly. I think the biggest problem is that most of the big companies want nothing to do with Linux. This ends up being a mixed blessing. The good part is, most of what you can get for Linux is free, under the GPL or BSD license, but the draw-back is that unless you are on an email list like this, people will be totally clueless about support issues.
