Jabber Journal #10 (2003-04-15)

In the last issue of the Jabber Journal, I promised to publish this little missive fortnightly. So here we are, two weeks later, jabbering away about everyone's favorite real-time messaging technology. Promise kept!

Also in the last issue I asked the question: what is going on with the explosion of Jabber interest and adoption in Poland? My new friend kocio reports that it's no coincidence! It seems that Jabberites in Poland launched a coordinated attack: an overview article in a major newspaper, public servers announced in the dominant free software / open source news services, a gateway to the popular Polish IM service Gadu-Gadu, documentation and client translations, several local open-source projects, and the JabberPL website (where translations of the Jabber Journal are published). A true grassroots effort! Now if only we could duplicate that success in every major nation on the planet, world domination would be just around the corner...

In other European news, I recently discovered a fascinating project called Lluna that enables people to share presence and chat with other people who are visiting the same web pages (including fun features like avatars, small video clips, and "bubble chat"). The first phase of this "virtual presence" project is almost finished and the client is ready for public use, so check it out. By the way, in another sign of the diversity of the Jabber community, Lluna is a joint creation of SURFnet (a Dutch academic network) and bluehands (a German software development company), who have generously open-sourced the code under the LGPL.

In other project news, I just found out about two more cool projects. The first is WebMessenger, an early-phase project to create an open-source, browser-based Jabber client (we've needed that for a long time!). The second is Jabbee, a graphical configuration tool for the jabberd server. And speaking of jabberd, development is proceeding apace (yes, I actually read the CVS checkins via the cvslog mailing list on JabberStudio), and there is even a nascent effort to port jabberd to win32 using Apache Portable Runtime Modules. Pretty sweet.

Let's see, last time I also promised to provide a report on the XMPP Working Group, including the in-person meeting in San Francisco and where we go from here. Basically the XMPP specs, which document the core Jabber protocol with appropriate modifications to meet the IETF's high standards for security and internationalization, are in great shape. We've had a lot of discussion recently on the xmppwg@jabber.org mailing list (thanks mainly to Lisa Dusseault, our wonderfully thorough working group chair), but the issues raised are fairly minor and I've been busy incorporating list consensus into our five Internet-Drafts. The next step is to hold a Working Group Last Call, which will give mailing list participants a final chance to comment on the documents. After that we will submit our documents for review by the Applications Area Director assigned to the XMPP WG. Once the Area Director is comfortable with the documents, we will submit them for review by the IESG (whose role in the IETF is similar to that of the Jabber Council in the JSF). The final step is an IETF Last Call, which lets the entire Internet community provide feedback on the XMPP specs. Each of these stages may take two or three weeks (perhaps more depending on how much feedback we receive), so if all goes well the XMPP Internet-Drafts will be approved by the IETF in two or three months. Stay tuned for further updates!

Speaking of protocol, I would remiss if I did not mention the many new JEPs that have been discussed and published in the last month or so. Although some of these merely document protocols in use (e.g., in-band registration and non-SASL authentication), others break significant new ground (e.g., Jabber Object Access Protocol (JOAP), Message Delivery Semantics, and Geographic Location Information -- a.k.a. GPS over Jabber). There has also been much discussion of Peter Millard's publish-subscribe proposal (which is nearing completion) and of Justin Karneges's JEP on in-band bytestreams (a fall-back protocol for file transfer). Another proposal that will have a significant impact on interoperability and compliance testing is JEP-0073, which seeks to define a protocol suite for basic instant messaging functionality in Jabber; look for finalization of this protocol suite in the next months, to be followed by a protocol suite defining advanced IM support (e.g., groupchat and XHTML messages).

Last but not least, the Jabber community showed its dedication to standards compliance exactly two weeks ago today by publishing JEP-0076 to complement the IETF's RFC 3514; the very next day, jabberd developer Rob Norris announced that he had implemented the protocol in jabberd. It was a true team effort, and enough to bring a tear to the eye of protocol geeks everywhere. ;-)

Until next time, Jabber on!

--stpeter