How to handle when a user puts in a query that returns 20000 results (it isn't hard to do).
Limit the number of results in a principled way: Add degree constraints to interior nodes, remove subclass_of edges (especially ones that go A-[subclass_of]->B<-[subclass_of]-C).
paginate results: return 100 or whateve
Return aggregate results? Maybe just the nodes and let users add more selectivity to the query based on that?
https://github.com/RobokopU24/Feedback/issues/108
How to handle when a user puts in a query that returns 20000 results (it isn't hard to do).