Re: [Uri-review] [xmpp] XMPP jabber: URI scheme not registered?

[Peter Saint-Andre <[e-mail]>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

[ removing apps-discuss ]

Actually that list of minted namespaces is incomplete. Here is what I
consider to be a complete list (note that some of these were rarely
used, some are no longer in use, some have been obsoleted, etc.):

      - jabber:client
      - jabber:component:accept
      - jabber:component:connect
      - jabber:component:exec
      - jabber:iq:agent
      - jabber:iq:agents
      - jabber:iq:auth
      - jabber:iq:autoupdate
      - jabber:iq:browse
      - jabber:iq:conference
      - jabber:iq:gateway
      - jabber:iq:groupchat
      - jabber:iq:last
      - jabber:iq:oob
      - jabber:iq:privacy
      - jabber:iq:private
      - jabber:iq:register
      - jabber:iq:roster
      - jabber:iq:rpc
      - jabber:iq:search
      - jabber:iq:time
      - jabber:iq:version
      - jabber:server
      - jabber:x:autoupdate
      - jabber:x:conference
      - jabber:x:data
      - jabber:x:delay
      - jabber:x:encrypted
      - jabber:x:envelope
      - jabber:x:event
      - jabber:x:expire
      - jabber:x:oob
      - jabber:x:roster
      - jabber:x:signed

On 11/28/12 11:57 AM, Peter Saint-Andre wrote:
> Something like this seems reasonable:
> 
> URI scheme name. jabber Status. permanent URI scheme syntax. 
> jabberuri = "jabber" ":" 1*(ALPHA) [ ":" 1*(ALPHA) ] URI scheme
> semantics. Strings of the form 'jabber:*' and 'jabber:*:*' were
> used as XML namespaces during development of the technology that
> became the Extensible Messaging and Presence Protocol (XMPP).  The
> scheme was never used for any other purpose.  The only namespace
> names minted with this scheme were: - jabber:client -
> jabber:component:accept - jabber:component:connect -
> jabber:iq:auth - jabber:iq:gateway - jabber:iq:last -
> jabber:iq:oob - jabber:iq:privacy - jabber:iq:private -
> jabber:iq:register - jabber:iq:roster - jabber:iq:rpc -
> jabber:iq:search - jabber:iq:version - jabber:server -
> jabber:x:conference - jabber:x:data - jabber:x:encrypted -
> jabber:x:oob - jabber:x:signed No other strings were minted, and no
> other strings shall be minted. Encoding considerations. Encoded as
> UTF-8 within XMPP protocol streams. Applications/protocols that use
> this URI scheme name. Extensible Messaging and Presence Protocol
> (XMPP). Interoperability considerations. The 'jabber' scheme must
> not be used to identify or enable interaction with XMPP addresses;
> the 'xmpp' scheme defined in RFC 5122 is to be used in such cases. 
> Security considerations. See Section 13 of RFC 6120. Contact. Peter
> Saint-Andre <[e-mail]> Author/Change controller. XMPP WG
> <[e-mail]> References. RFC 6120
> 
> On 11/28/12 11:06 AM, Graham Klyne wrote:
>> Peter, all,
>> 
>> I don't think it even needs to be a draft.  XMPP spec already
>> exists, so we should just be able to submit the registration
>> template to IANA.
>> 
>> I was thinking I'd try and draft something and run it by you (but
>> don't hold your breath).
>> 
>> #g --
>> 
>> On 27/11/2012 17:05, Peter Saint-Andre wrote: On 11/27/12 10:04
>> AM, Joe Hildebrand (jhildebr) wrote:
>>>>> On 11/27/12 9:34 AM, "Graham Klyne" <[e-mail]>
>>>>> wrote:
>>>>> 
>>>>>>> Yes, jabber:client and jabber:server are required by
>>>>>>> RFC 6120 (and RFC 6121 requires support for
>>>>>>> jabber:iq:roster).
>>>>>> 
>>>>>> OK, that's what I originally thought.  In which case, I
>>>>>> think the text from RFC 4395 that you cited does not
>>>>>> apply, since use of these jabber: URIs is still required
>>>>>> (and others as you note below).
>>>>>> 
>>>>>> I think the appropriate course would be to register the
>>>>>> URI scheme, maybe list the URIs in use for this scheme,
>>>>>> and add a note that no more jabber: URIs should be
>>>>>> minted.
>>>>> 
>>>>> (as individual)
>>>>> 
>>>>> As long as the registry has a policy of "Closed" or
>>>>> similar, I don't really care what status the doc has.
>>>>> Let's not bog down.
>>>>> 
>>>>> (as XMPP co-chair)
>>>>> 
>>>>> This isn't on our charter at the moment, so whoever wants
>>>>> to write an individual draft first should just pick a
>>>>> status, and that will probably stick.
>> 
>> I think it can be an informational I-D outside any WG, and will
>> find time to bang that out before the end of the year.
>> 
>> Peter
-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.18 (Darwin)
Comment: Using GnuPG with undefined - http://www.enigmail.net/

iEYEARECAAYFAlC2aNUACgkQNL8k5A2w/vzqfwCbBnpyz+knu+LZ+z30myXmsH1P
y+cAn0DhCbwHSkJfJaVx3NhxlCjBpgip
=LSGj
-----END PGP SIGNATURE-----