Interesting stuff, but to see anything other than the abstract
would cost me $39.95. If you could send me a (p)(r)eprint I could
definitely share my thoughts.
In semantic work, you've got to evaluate the usefulness of
something against particular use cases, so I'll bring up two examples
of provenance in DBpedia-derived work that came up recently:
(1) I got a complaint from a user who found a highly offensive slur
in an abstract that came from DBpedia. In this case I think a student
at a high school wrote something bad about the principal. Certainly we
could trace this from the Wikipedia history. I like to be proactive
when problems turn up, so it would be nice to be able to find other
texts that were touched by the same person in case he's a serial vandal.
(2) It's a bit out of DBpedia's scope, but I was contacted by a
person who wanted to use an image in a book and needed a higher
resolution image than was available from Wikimedia Commons. In this
case the only provenance information that was there (in semistructured
form) told us the image was public domain because it was scanned out of
the old book. The provenance information that I wanted was the identity
of the old book so these people could have found the book in a library
and scanned it themselves. In this case the information just wasn't
there and the only route forward would be a perscomm to the person who
uploaded it.
------------------------------------------------------------------------------
What Every C/C++ and Fortran developer Should Know!
Read this article and learn how Intel has extended the reach of its
next-generation tools to help Windows* and Linux* C/C++ and Fortran
developers boost performance applications - including clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Dbpedia-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
