Currently unique IDs for a stream are generated using repetitive code in multiple locations, possibly allowing for inconsistent behavior.