Because converting and scanning books, along with the matching process, is time consuming, this information should persist across runs of BookSift. Books which have been compared on previous completed runs should not have to be compared again. Books should also only need to have their fingerprints generated once; then this information should persist as well.
Because converting and scanning books, along with the matching process, is time consuming, this information should persist across runs of BookSift. Books which have been compared on previous completed runs should not have to be compared again. Books should also only need to have their fingerprints generated once; then this information should persist as well.