W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2012

Re: [httpbis] #364: Capturing more information in the method registry

From: Mark Nottingham <mnot@mnot.net>
Date: Wed, 4 Jul 2012 06:16:55 +1000
Cc: HTTP Working Group <ietf-http-wg@w3.org>
Message-Id: <6A49C006-E8A7-46C4-8509-DD7DDC99D074@mnot.net>
To: mike amundsen <mamund@yahoo.com>
It's common for the registration template to not reflect all of the requirements upon a registration.

For example, it would be at best inefficient to try to reflect the semantics of the body in the registry.

Cheers,


On 04/07/2012, at 6:14 AM, mike amundsen wrote:

> Thanks for the pointer, Mark:
> 
> In reading both 2.2 and 2.2.1 there is (in my mind at least) some discrepancy:
> 
> From 2.2:
> "Registrations MUST include the following fields:
> - Method Name (see Section 2)
> - Safe ("yes" or "no", see Section 2.1.1)
> - Pointer to specification text"
> 
> From 2.2.1:
> "New method definitions need to indicate whether they are safe
> (Section 2.1.1), what semantics (if any) the request body has, and
> whether they are idempotent (Section 2.1.2)."
> 
> So, when registering a new method in the future, indicating Safety is
> a MUST; indicating Idempotence is a .... SHOULD? MAY? And what of body
> length/semantics, and cachability? SHOULD? MAY?
> 
> From my POV, resolving the above Qs will better inform decisions on
> how to complete the table in "Initial Hypertext Transfer Protocol
> (HTTP) Method Registrations"
> 
> 
> mca
> http://amundsen.com/blog/
> http://twitter.com@mamund
> http://mamund.com/foaf.rdf#me
> 
> 
> 
> 
> On Tue, Jul 3, 2012 at 3:58 PM, Mark Nottingham <mnot@mnot.net> wrote:
>> 
>> 
>> On 04/07/2012, at 5:54 AM, mike amundsen wrote:
>> 
>>> Yes, I think safety and idempotence are most needed in this table of previously registered methods.
>>> 
>>> Per Julian's comment: if one of the reasons to leave idempotence off the list is that we don't know whether some of these methods are idempotent or not, then I'd opt for saying that in this table ("UNK", etc.) rather than leaving that property out of the table completely.
>>> 
>>> On a related note, I wasn't able to find any details on the method registry. Anyone able to give me some pointers?
>> 
>> It's set up here:
>>  https://svn.tools.ietf.org/svn/wg/httpbis/draft-ietf-httpbis/latest/p2-semantics.html#method.registry
>> 
>> Cheers,
>> 
>> 
>> --
>> Mark Nottingham   http://www.mnot.net/
>> 
>> 
>> 

--
Mark Nottingham   http://www.mnot.net/
Received on Tuesday, 3 July 2012 20:17:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 July 2012 20:17:32 GMT