On Tue, Apr 6, 2010 at 14:29, David Mimno <[email protected]> wrote:

> From my perspective, what would be especially useful would be to
> specifically separate out the places where a human has made a specific
> change -- correcting a field, merging two records, etc -- including the
> original state, the final state, and a reasonable amount of context (I
> realize that's vague, sorry!). This type of human-supervised data can be
> used to train automated techniques that try to emulate such processes.
>

I think this kind of dump could prove to be very useful and may benefit Open
Library in the future.

James
_______________________________________________
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]

Reply via email to