I assume the technique used to detect the 'leaks' would be something like: - operate authoritative dns for detector.tld - create a client-specific ID - have

I assume the technique used to detect the 'leaks' would be something like: - operate authoritative dns for detector.tld - create a client-specific ID - have the client request something from client-specific-id.detector.tld - log the origin of requests to resolve client-specific-id.detector.tld - repeat with different ids until happy - return the log results to the client I wonder if there are significant malign use cases for this (outside of standard megacorp privacy invasion, I mean)?