Adding to the Wind in Wellington

by Rick Jelliffe

I've just returned from a super-interesting week in Wellington, the capital of New Zealand. The government there has a very forward-thinking e-government program at the State Services Commission (SSC) that has to finesse its way around a severely fragmented organizations structure (they made several friendly digs at Australia's three levels of government, but then someone mentioned they had 45 different government departments, or did I hear wrong? :-) and the emergence of powerful blocks of topical industry standards (Health, Education) which were not designed with inter-department data sharing in mind.

The e-GIFprogram has quite a few topics, including an advanced activity in authentication. I think Colin Wallis' RFC 4350 A Uniform Resource Name (URN) Formal Namespace for the New Zealand Government gives a pretty good indication that they are asking the practical questions about providing the mortar between international standards and locally-useful standards.

They are putting in various services as part of a second-generation of pilots: of most interest to me is that their first set of pilots indicated to them that there was a real practical problem with validation: with XSD validation begin too difficult to set up, too difficult to interpret, not powerful enough, and not flexible enough to cope with the variants and derivations of real life schemas. So they are putting in a Schematron online validator, among several other measures. SSC's Liz Kolster seems to be approaching it with a pretty hard nose.

My main task was to address a series of meetings on the topic of XML Governance: everyone's mind on the same page was the goal. The first meeting was internal for the immediate staff. The second was for government department representatives ("stakeholders"). The third was for commercial integrators, information architects and so on ("vendors") which was moderated by a really good professional facilitator, who used something called World Cafe that worked very well and efficiently, I thought.

What surprised me was that vendors really wanted more direction from government: more use of XML, more standardization, more best practices, more forums, more dialog, and so on. Terribly positive and encouraging. It seems to me that one reason why XML projects tend to succeed is that people adopt XML because they are asking the question "How can I make it easier for the other people in this ecosystem?", and that attitude (even more than the technology) is the big win. Where projects are conducted where the ease of implementation of the other blokes in the project is ignored, you have a disaster waiting to happen.

As I was in town at SSC's behest, on Thursday night I was invited by the New Zealand Open Source Society to speak on Wikigate and the Open XML process. They were so welcoming, sweet and smart, not to mention alcoholic, that I had a really enjoyable time. Plus I met up with an old friend from SGML days (Hi Richard) which was a real bonus. Andrew MacMillan has blogged about the talk Chinese Whimpers and kindly added my pic to my Wikipedia entry, (The reference to "bribery" is a joke based on comments in the talk btw.)

The talk was a run-through for a keynote I'll be giving next week at Open Publish 2007conference in Sydney, which will be called "The True Saga of Wikigate" that should be fun.