Going back to my email that started this thread (not to say the subsequent discussion hasn’t been interesting), I wasn’t actually proposing a correction system but a quality feedback system. It seems to me this could be implemented much more quickly than a true editing system.
Readers of OL ebooks would be encouraged to rate OCR quality, perhaps a 5-step scale, with optional comments: 1. unreadable 2. errors seriously impede reading 3. some errors but able to cope 4. a few errors 5. no errors attributable to OCR It should be obvious that this needs to be a completely separate kind of “review” than is offer at Google Books or Amazon. For example, if these books [1,2] were potentially popular in plaintext, the negative reviews they would attract because of bad OCR would distort those ratings to the point of uselessness. I imagine that aggregated results from this feedback would be useful in deciding whether to tune the OCR algorithms or parameters, rescan the book, or pull the ebook from distribution. Likening OpenLibrary to OpenStreetMap... it would be the equivalent of OpenStreetBugs, not the JOSM editor. Likening OpenLibrary to Wikipedia... it would be the equivalent of the “Rate this page” and “Help improve this article” boxes, not the Edit tab. regards, - Laurence [1] http://books.google.co.uk/books?id=XlYuAAAAYAAJ&printsec=frontcover#v=onepage&q&f=false [2] http://openlibrary.org/books/OL23277054M/Anecdotes_from_Pliny's_letters _______________________________________________ Ol-discuss mailing list [email protected] http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss To unsubscribe from this mailing list, send email to [email protected]
