@lrhodes @lmorchard I don't know enough about their protocol, is having large indexing servers the trade-off for small "user servers" or is it the goal (is for data harvesting)?