[Foundation] JSF == Jabber Standards Foundation?

Matt Tucker matt at jivesoftware.com
Thu May 22 19:04:55 CDT 2003


Peter,

> 1. We've been building mindshare around the term "jabber" since 1999. I
>    see no compelling reason to throw that away.

I think there are many compelling reasons (even from the perspective of 
JINC):

  * There is still very broad confusion over Jabber as an open protocol 
vs. Jabber as a commercial company. I can cite two specific cases:

   a) A recent discussion on Slashdot where a large group of technical 
people have no idea that Jabber is protocol rather than a service or 
company. There is one person trying very hard to record straight, but 
it's a lone voice in the masses:

<http://slashdot.org/article.pl?sid=03/04/11/0113239&mode=thread&tid=102>

  b) The Jabber thread in the Trillian forums. There are a few people 
that know what is going on, but most have a totally wrong idea about 
Jabber and actually vote to have Trillian *not* support it because "they 
used Jabber way back when and hated it".

<http://www.ceruleanstudios.com/forums/showthread.php?s=db68b72b0806d70b81d09f706f9b867a&threadid=6777>

These two discussions represent opinions from people that should be "in 
the know" -- technical people on Slashdot and IM fanatics in the 
Trillian forums. The fact that there is so much confusion indicates that 
the current "Jabber as an open protocol" branding strategies have not 
been entirely effective. My own personal experience with customers 
validates this -- they actually understand that XMPP is an open protocol 
much better than they do that Jabber is.

Further, all recent articles I've seen use the term XMPP (even if they 
mention the term Jabber too). XMPP is the official name of the protocol 
so that's what people will use.

I'll put it frankly, as a commercial company implementing the 
Jabber/XMPP protocol, the JSF would be much more valuable to me with a 
name that doesn't promote a competitor. I'm willing to put my money 
where my mouth is -- I'd gladly double our current donation to the JSF 
after a name change, since the organization would become much more 
valuable to our commercial efforts (since we really want to see an open 
standard around IM succeed).

> 2. Jabber is not co-equal with IM. Jabber is about streaming XML. We're 
>    using XML streams for IM, but we already know it can do a lot more
>    than IM.

I agree. That still doesn't mean we should be using the term "Jabber". 
Maybe it means we can't use "XMPP" for everything the protocol does, but 
I'd much rather pick some un-encumbered term rather than Jabber.

> 3. XMPP is essentially owned by the IETF's XMPP WG. As the person who is
>    on the front lines of the Jabber community's standards efforts in the
>    IETF, I can tell you that there is no way we want to send every one
>    of our protocols through the IETF process. While that process is and
>    has been benefecial for the core protocol, there is absolutely no
>    justification for sending every one of our protocols through the
>    IETF. If you think that is the best course of action, I cordially
>    invite you to initiate the charter change for the XMPP WG and take
>    over writing Internet-Drafts. Warning: it is not a day at the beach.

I totally agree with this and wouldn't want to see us try to push 
everything through the IETF. I love the JEP process and want to see it 
continue. I think there are many ways we could extend XMPP while not 
using the term Jabber -- it could be an "XMPP extension proposal" or 
even some term that hasn't been invented yet. It's a semantic difference 
rather than a procedural one, but the semantic difference is a very 
important one.

> 4. It is much more productive for the Jabber community to develop its
>    own protocols on top of the XMPP core (you could argue that even the
>    XMPP IM Internet-Draft would have been better to keep out of the IETF
>    process, since it's an application protocol built on top of our data
>    transport layer, but that wasn't possible for various reasons). This
>    gives us great freedom to innovate (take that, Microsoft!) without
>    being beholden to the slower pace of IETF work.

I 100% agree with this too, but it still doesn't mean we should be using 
the term Jabber.

>>I don't think this is true. The broader term is "Instant Messaging" not 
>>"Jabber". Nobody will ever say, "Hey Sally, I'll jabber you when I get 
>>home".
> 
> You are wrong. I hear it all the time from the 1,000 or so people in my
> roster.

Heh. But, what about the rest of the tens of millions of people using IM? :)

>>The "broad audience" we need to appeal to is those that will use and 
>>implement the protocol. I think compelling arguments have been made that 
>>those implementing the protocol (especially commercial people) only want 
>>to use the term XMPP. My kid brother who uses AIM to chat with his 
>>friends is not someone that the JSF is trying to reach. The executive at 
>>AOL who might someday be forced to choose an open protocol for their IM 
>>network is.
> 
> So they'll call it XMPP.

Right, they will. But, will they be incented to participate in the 
"Jabber Software Foundation" or "Jabber Standards Foundation"? Perhaps, 
but only with great trepidation. At least, that is how we've felt as a 
commercial company -- we love the protocol and the vision but are very 
nervous about the promotion of the term Jabber.

>>Yes, the world already has Jabber and associates the term with a 
>>commercial company. This is the point of confusion that I and others 
>>have brought up. XMPP needs to be the term that we brand as "an open IM 
>>protocol".
> 
> You assert, you do not offer arguments. 

Please see above for some specific examples. Also, others besides myself 
have pointed out the commercial problems with the term Jabber.

>> I'd rather call 
>>it "community extensions on top of XMPP".
> 
> Catchy.

Sorry, it's all I could come up with on short notice. :) I'm sure the 
community can come up with some much better alternatives.

> That is false. Maybe some commercial entities don't want to use the term
> Jabber, but it's false to say that none do.

Would you accept "most"? I think that any reasonable person has to admit 
that there is a prima facie dis-incentive for any commercial company to 
use the term Jabber. As a specific example, we, Jive Software, 
specifically avoid using the term Jabber since we've been burned by 
doing so already.

> I agree that in an ideal world, Jabber.com (yes, that was the original
> name back in early 2000) would have chosen a different name. Personally
> I think it would be great if we could say:
> 
> "Jabber is like the Linux of instant messaging, and LoDo is like the
> RedHat of Jabber" (where "LoDo" is my fictitious name for Jabber Inc.)

Yeah, I totally agree. I like the term Jabber quite a bit and wish we 
could use it. But, as you say, we can't change reality, and the reality 
is that there is a company named Jabber Inc. So, despite the fact that 
they have been such great contributors to the community and protocol, I 
think it's time our organization moved on to a name that will be more 
viable to everyone.

>>For example, we have a commercial XMPP server that appeals to certain 
>>vertical markets (places where the Jabber Inc. solution wouldn't work at 
>>all). However, just because of the Jabber protocol/product confusion, 
>>I've had a customer ask, "Wait, this is based on Jabber, right? Why 
>>wouldn't I be using the server from Jabber Inc. then?"
> 
> Sounds like you need to educate your customers.

Trust me, we do. :) But, why should I have to? The fact that we need to 
is a strong enough argument by itself, I think.

> No one is saying that Jabber will supersede IM (as in, no one uses the 
> term IM anymore and they just use Jabber). Obviously it's too late for 
> that. However, based on the number of non-IM applications of the Jabber
> protocol, I think it's safe to say that Jabber is more than merely IM.

Yep, agreed. I apologize if I confused the discussion with this.

Regards,
Matt




More information about the Members mailing list