use Amazon's Mechanical Turk to transcribe my notebooks
-
it might work well to use Evernote as a backend for the scanned images and transcribed text - it already uses OCR to make the text in the images searchable, and it looks like they have an API for modifying those text annotations, so it could be integrated into a custom UI
http://forum.evernote.com/phpbb/viewtopic.php?f=43&t=11056&p=44184
http://www.evernote.com/about/developer/api/ref/NoteStore.html#Fn_NoteStore_updateResource -
This one is still pretty close to the front of my project queue, and I think it would be useful for a lot of different people, like http://castingwords.com/ for text.
-
Joel commented
I am interested to see what comes of this. Have many notebooks to scan and convert to something searchable and shareable.
-
http://fotonotes.net/ could probably be used for the tagging, perhaps over individual words? then you could calculate % overlap of the boxes and compare the strings that the turkers wrote for each one
-
example page scan with annotations: http://www.flickr.com/photos/lehrblogger/3960518206/
-
(ping me for more information about this, I'm seriously considering building it)