Right, and thanks for your thoughts. Doing a bit of code archaeology suggests that everything was kept minimal because the websocket implementation wasn't real

Right, and thanks for your thoughts. Doing a bit of code archaeology suggests that everything was kept minimal because the websocket implementation wasn't really meant to be used beyond the narrow scope of its initial implementation. Any updates (if we needed more data, for example), would just mean expanding the return payload with the necessary data. The whole question really comes into play specifically when looking at designing an API for public consumption, because you really then can't know what the end user wants to receive.