Re: Issue 72, was: Status of IANA Considerations (registrations and registries) -- issues 40, 59, 72, 79

Mark Nottingham wrote:
> Looks good from here, and yes, this should be a WG item.

OK, will prepare that as draft-ietf-httpbis-method-registrations. 
Publication will have to wait until draft-ietf-httpbis-p2-semantics-04 
is ready.

> Q: any more thoughts about whether to add things like safety, 
> idempotence to the registry?

I'm in favor of adding stuff that would be important to know in an 
implementation, such as: "is this method safe" (so can I retry, and can 
I automatically follow redirects?).

For "safety" this is pretty clear, for idempotency I'm not totally sure 

Furthermore: what about cacheability? In WebDAV space, I see many specs 
making statements about the cacheability of a method result. Is this a 
good idea? After all, 
<> states:

"Unless specifically constrained by a cache-control (Section 14.9) 
directive, a caching system MAY always store a successful response (see 
Section 13.8) as a cache entry, MAY return it without validation if it 
is fresh, and MAY return it after successful validation."

So if a (success) method result is known not to be cacheable, the server 
must send it with "Cache-Control: no-cache" anyway, right?

Proposed next steps:

- add "safeness" to the method registration template (Part 2)
- move proposed initial registration into a WG draft

BR, Julian

Received on Tuesday, 15 July 2008 11:57:07 UTC