Has anyone seen a third party product or attempted develop their own that would attempt to detect duplicate documents based on documents content? Like comparing textshots I suppose?