And the fact that you can run the indexing and visualization in separate passes is extremely useful, especially for E.g. large network file systems that are slow to index.