Introduction

A beginning is the time for taking the most delicate care that the balances are correct

Princess Irulan (Frank Herbert: Dune)

Hi. My name is Adam Nemeth, known as Aadaam. I'm a software developer from Hungary, EU. For years I've been using ICQ, as it was the first instant messaging service, and for me, it was a standard, too. However, as years passed by, more and more instant messaging solutions came, the (at least in my viewpoint that time) ICQ-clone MSN Messenger, and the peer-to-peer VoIP software Skype.

I first used Jabber protocol sometimes around 2004 or so: I heard of it before, and some of my friends used it ocassionally, but it was a struggle with icq-transports and clients of that time. But I realized that ICQ is just a service, and as more and more people starts using instant messaging every day, we need to consider it as a part of a communication solution - most of the universities, student clubs were already providing e-mail and webhosting services, as well as some forum and so on. Realtime is getting closer as our life gets faster every day.

When Google introduced its XMPP-service called Google Talk, I was way in jabber: I knew how to use the protocol with a pure telnet, knew how to write a client and even had some ideas how to use it on web with XEP (or, at that time JEP) 124. Now I see it as a communication standard.

But much work has to be done: for example, it's not clear how we could use alias and forwarding features of e-mail. It's not clear how we could connect to accounts to each other. Practically only ejabberd provides virtual hosting, and erlang is an unusual language for an average programmer working with imperative code like Java or C++ or whatever.

There's also a problem with the protocol design itself: given that you have a heterogenous network with all kind of clients, the interoperability between those is the intersection of the supported XEPs of all clients; as more and more clients come in, it gets worse. We need to bundle the logic with the XEP itself, and there are some very good experiments out there for such.

We also need to make jabber easier to spread: embeddable clients in webpages, open hosting providers, etc... its not as easy as it seems but it can be done. Also we need a lot of features from other networks - in Hungary, mostly MSN is used today, because of custom smiley handling, for example. Teens love it, and can't imagine a client without that feature- it'd be 'boring'.

This blog will focus on mainly three topics: first, and not least, possible use cases on jabber services. Then protocol implementation examples, with simple, educational scripts in GPL or even public domain. And also there'll be some discussions about possible new protocol extensions, like the above mentioned custom smiley protocol I'm designing.