As I prepare to implement rate limiting on some APIs, I've been considering how to communicate, to API clients, the "status" of their rate limit. The purpose of this communication it to allow API clients to adjust their behavior - for example, if the client is told "you are only 10 calls away from your rate limit, and your limit will reset in 30 minutes", the client might adjust its call frequency to once per 3 mins.
Is that the best approach? Are there other approaches you like? Are there any standards or best practices?
Thanks, Cooper