[Standards-JIG] XMPP URIs was: Two questions regarding JEP-0124HTTP Binding

Ralph Meijer jabber.org at ralphm.ik.nu
Wed Nov 30 08:36:48 UTC 2005


On Tue, Nov 29, 2005 at 02:51:39PM -0700, Peter Saint-Andre wrote:
> [..]
>
> What about jabber80.com? It doesn't communicate on the standard XMPP 
> ports. So is the following identifier confusing?
> 
> xmpp:jabber80.com
> 
> Well, it *is* an XMPP server, so identifying it that way seems OK. If 
> you attempt to interact with that server following the procedures in 
> draft-saintandre-xmpp-iri then you will be fine if both the client and 
> server also support SRV records as specified in RFC 3920, since you'll 
> find out it uses port 80 rather than port 5222 for client-to-server 
> connections.

This is totally irrelevant. A URI of <xmpp:jabber80.com> means nothing
more than pointing out the resource jabber80.com that is accessable via
XMPP.

Suppose I have an account at jabber.80.com, the URI for my account would
be <xmpp:ralphm at jabber80.com>. When something dereferences that URI, for
example to send me a message, it should access its own XMPP server and
make it send the message to the JID ralphm at jabber80.com. Like any server
on the global Jabber network, the server jabber80.com is accessable via
server-to-server connections, and accessing it will always use the
standard mechanism to find out the port using SRV records and falling
back to port 5269. How the URI-dereferencing entity accesses its own
XMPP server is not specified.

Only if an XMPP URI would have an authority component (unlike the
example above) the connection of the sending entity comes into view. For
example, if someone had a URI like
<xmpp://ralphm@example.com:5222/ralphm@jabber80.com>, this would have to
mean: contact the XMPP server at the domain name example.com, port 5222
using the XMPP protocol and login as user ralphm. I believe this should
always use a standard XMPP client-to-server connection over TCP. Not
specifying the port, would use SRV records.

-- 
Groetjes,

ralphm



More information about the Standards mailing list