Re: extending status codes

Robert Sayre wrote:
> 
> I recently got into a debate with a colleague about whether it would
> be acceptable to use higher 5xx codes for custom status codes. I
> maintained no, it's a shared namespace, we shouldn't do that. But the
> response was "show me where it says you can't do that". To me it seems
> implied, but maybe it's not to someone who wants to Get Work Done as
> fast as possible by writing a switch statement on a number.
> 
> Could we get a sentence explaining that adding status codes requires
> coordination?

Yes, and that coordination is defined in 
<http://tools.ietf.org/html/rfc2817#section-7.1>. Not a good place, 
though, thus <http://www.w3.org/Protocols/HTTP/1.1/rfc2616bis/issues/#i59>.

Best regards, Julian