I would have to agree with mat. But to each their own. The return codes frequently make no sense from twitter, so i guess the fact that it doesn't make sense it is irrelevant, so long as it is consistent.
On Dec 3, 6:29 pm, mat <mat.st...@gmail.com> wrote: > Given that 400 is bad request, and the client SHOULD NOT repeat the > request without modifications (w3.org's emphasis), and 503 means > service unavailable, try again later, and can include a retry-after > header, would it not have made more sense to change the response code > of the REST API to the "more correct" one? > > On Dec 3, 10:41 pm, Wilhelm Bierbaum <wilh...@twitter.com> wrote: > > > > > In an effort to simplify our APIs, we are standardizing the response > > codes returned by our various systems. Historically, the Search API > > has returned 503 for rate limiting whereas the REST API has returned > > 400. So, we are changing the response codes sent back from the Search > > API. > > > Starting Wednesday, December 16th, 2009 the search API will respond > > with error code 400 in the event that the number of requests you have > > made exceeds the quota afforded by your rate limit. > > > Please update your response handler accordingly. > > > If you have any questions, please feel free to post them on twitter- > > development-talk.
