XMPP Gets An IETF Working Group

An anonymous reader writes "The IETF has approved the formation of a Working Group to continue evolving the XMPP protocol." Interoperable instant messaging, who'd a thunk it. Our previous story has more information.

Re:Admirable, but ...

[rusty0101]

Considering that the real start to this was Microsoft's attempt to interoperate with AOL, using the open standards AOL published (and later stoped useing) to allow independent client developers to talk to AIM servers, I think there is a desire at some of the IM server platforms to interoperate.

Obviously there are problems that will still need to be adressed, however that is what the working group is attempting to address.

As far as I am concerned, knowing that one of my buddies is online, and being able to send an IM to them, regardless of the service or client either of us are using would be a major step forward.

-Rusty

what XMPP really is

[ageitgey]

The headline is a little misleading. This isn't a working group to create some new standard for interoperability. This is a working group to evaluate and possibly improve Jabber [jabber.org]'s protocol.

In other words, this new group will ensure that Jabber's existing protocol is secure and has good support for localization. But it has nothing to do with AIM/ICQ [businessweek.com], Yahoo Messanger, or anything like that. You can use XMPP today - it's called Jabber (and it's pretty cool).

I fell for the /. hype....

[brassman]

Actually installed a Jabber server yesterday. Should have dug a little deeper though; I got it from jabber.com instead of jabber.org, so it's going to expire in a month or so. (Duh.) Well, guess I've got some time to straighten that out.

What really makes me shake my head, though, is the client they provided. It's locked on the jabber.com server. What's up with that? They sell you a server, and then give you a client that you can only use with a server they didn't sell you?!

Re:Admirable, but ...

[Zorikin]

> Trillian and other clients already make interoperability a practical reality.

Only for end users (who this standardization effort isn't aimed at anyway, afaik). For programmers, feature P may exist in protocols X Y and Z, but they may have different requirements, so redundant code will have to be introduced to enable P across every protocol which supports it. Furthermore, users of protocol W might be screwed wrt P even though every other protocol supports it.

So AOL (for example) may be betting that it can keep its share of clients and keep selling AIM ad space.

Then again, some end users would like (for example) to have IM handles which match their email address, both of which have the same domain name as their web sites. Jabber solves this sort of namespace issue this by using a decentralized addressing scheme similar to that of email. Meanwhile, AIM users make a game of being the first to register distinctive screen names.

Just another IETF standard for IM...

[jamezilla]

FYI, this is only the latest offering in the arena of IM standards. There are 4 other IETF working groups related to IM standards:

Instant Messaging and Presence Protocol (IMPP) [ietf.org]
SIP for Instant Messaging and Presence Leveraging Extensions (SIMPLE) [ietf.org]
Session Initiation Protocol (SIP) [ietf.org]
Common Presence and Instant Messaging (CPIM) (Still a draft) [ietf.org]

In addition to these guys, Wireless Village [wireless-village.org] is an IM standard created by Ericsson, Motorola, and Nokia. It's getting very strong traction among wireless carriers who want to deploy IM on phones and other mobile devices. Of these different offerings, SIP isn't strictly an IM thing, but there are people trying to use it to set up IM sessions. Microsoft uses SIP in their Messenger offering (which is how they claim they are "standards-based").

CPIM is probably dead.

IMPP has some traction in the 3GPP wireless groups, but not really anywhere else (read "probably dead").

SIMPLE has tons of backers including IBM (Lotus) and is probably going to emerge as one of the dominant standards.

Jabber is just trying to stay afloat in all this standards chaos. This was a very good move for them since they actually have millions of deployed users. Jabber is the only IETF-related working group that can claim real-world deployment like this. None of the other standards have any subtantial deployed user base (if any users at all).

Probably what will happen is that as IM servers emerge, they will support a handful of these protocols, just like email servers currently support IMAP, POP, etc.

Notice that AOL, ICQ, MSN and Yahoo! are not pushing their protocols as standards anymore. They are plying the Mexican stand-off thing and probably will have to scramble to jump on one of these standards as things shake out.

IETF has SIMPLE working group too...

[mamahuhu]

An article on News.com [com.com] mentions that "the new working group could have some competition from IBM and Microsoft, which have promoted a separate standard known as SIMPLE". This also has a IETF working group - here's the charter [ietf.org]

Meanwhile [internetnews.com]a group of users in finance industry are pushing [com.com] for exactly this sort of integrated solution. Called FIMA they "say it is non-partisan, and is open to any company that wishes to promote Internet Engineering Task Force (IETF) IM standards and protocols within the financial services community. By endorsing IETF instant-messaging standards, FIMA wants to promote "interoperability and beneficial competition among instant-messaging vendors."

There is an air of enevitability about the integration of protocols - but it may not be based on Jabber.... but SIMPLE doesn't sound [news.com.au]all that hot...