Re: Additional status codes in HTTP/1.1

["Nicolas Mailhot" <[e-mail]>]

Le Jeu 25 juillet 2013 10:24, Julian Reschke a écrit :
> On 2013-07-25 09:51, Nicolas Mailhot wrote:
>>
>> Le Jeu 25 juillet 2013 08:18, Julian Reschke a écrit :
>>> On 2013-07-25 07:24, Willy Tarreau wrote:
>>>> Hi,
>>>>
>>>> A user notified me that status codes 428, 429, 431, 511 introduced in
>>>> RFC6585 by Mark & Roy are not mentionned at all in the current 1.1
>>>> draft.
>>>>
>>>> Shouldn't we copy them there, or at least add a reference to RFC6585
>>>> so
>>>> that implementers know that these codes exist ?
>>>>
>>>> Thanks,
>>>> Willy
>>>
>>> No, that would send the wrong message.
>>>
>>> The list in the spec is not exhaustive; there's an IANA registry for a
>>> reason.
>>
>> But is there any good reason not to consolidate the codes that were
>> known
>> at the time?
>
> As I said: it sends the wrong message. What's relevant is the IANA
> registry.

Then why are the other codes documented at all? They should be in the IANA
registry! Some are so obsolete they're almost never used in the wild
nowadays.

Even though the ietf process allows creation of new rfcs to extend past
ones, the common practice has been to simplify implementor's lives and
merge extensions when the extended rfc is revised. Is this process
suddenly frowned upon?

-- 
Nicolas Mailhot