[Standards] Proposed XMPP Extension: Unique Room Names for Multi-User Chat

Kurt Zeilenga Kurt.Zeilenga at Isode.COM
Tue Sep 27 16:16:40 UTC 2011


On Sep 27, 2011, at 9:10 AM, Peter Saint-Andre wrote:

> On 9/27/11 10:07 AM, Kurt Zeilenga wrote:
>> 
>> On Sep 27, 2011, at 8:35 AM, XMPP Extensions Editor wrote:
>> 
>>> The XMPP Extensions Editor has received a proposal for a new XEP.
>>> 
>>> Title: Unique Room Names for Multi-User Chat
>>> 
>>> Abstract: This specification defines an XMPP protocol extension for requesting a unique room ID from a multi-user chat service.
>>> 
>>> URL: http://xmpp.org/extensions/inbox/muc-unique.html
>>> 
>>> The XMPP Council will decide in the next two weeks whether to accept this proposal as an official XEP.
>>> 
>> 
>> Would be nice if this spec would offer some rationale for the introduction of this extension…
> 
> It's simply being moved from XEP-0045 to a standalone spec (perhaps,
> i.e., if people still think it's useful).

yes, s/introduction/existence/

Just some note as to why this spec is considered by some as useful, some thing which gives some applicability to real world problems.

> 
>> and it would be good if the spec stated that return of a name does not imply the name is reserved in way.
> 
> In order to be useful, the room ID would need to be reserved for some
> period of time, no?

Depending on what the problem one is trying to solve by server generation of the name instead of client side generation, possibly so.

If reservation is desired and to be required of servers, that that should be clearly stated.

I note that this would prevent a user from asking the server to get them a unique name and passing that to another user for the actual room creation.

Without understanding the problem folks are trying to solve with this, hard to say whether reservations is desirable or not.

> 
> By the way, I think this feature needs to be deprecated. I submitted
> this ProtoXEP only to foster discussion on this list (previous
> discussion happened on the muc at xmpp.org list).
> 
> Peter
> 
> -- 
> Peter Saint-Andre
> https://stpeter.im/
> 
> 




More information about the Standards mailing list