Writings on technology and society from Wellington, New Zealand

Monday, June 28, 2010

Good in parts

I spent today at the Microsoft Open Government Unconference in Wellington. It was an interesting experience. On the one hand, I got the impression of a company trying to weave something from the whole “open government, open data” thread which is being spun out of a genuine desire by government folk to share things, and on the other hand there was the disconnect of trying to script the agenda for an “unconference” – a contradiction if ever there was.

I should explain how I got involved. Oliver Bell of Microsoft suggested I should look at going – I had been aware of the session but hadn’t considered it before that. Invitations were flooding the government sector. The session was free (like an unconference or Bar Camp) but you needed to “apply” for a place which might or might not be granted, unlike an unconference. In the spirit of trying to help the government folks to get a good outcome I applied and was duly granted a place, on the understanding that I would call the whole day as I saw it, which I did on Twitter (check @ThisCJ and #opengovt2010) and am doing in this posting.

Subsequently I met Anni and Michele of Digitalbrand, who were contracted by Microsoft to run the session. They had recently run something similar for Microsoft in Canberra, and a couple of New Zealand government people had gone over for it. We talked a lot about the difference between a self-organising informal unconference and a formal conference with a set agenda and speakers.

This was certainly not the first informal session to be organised around open government data in Wellington. There have been several self-organised bar camps and there will presumably be more. The government folk and the open data types are used to just rolling up their sleeves and making these things happen. So, we were slightly amused and a little nervous to see how a large corporate would approach the notion.

The day before the session I met with the organisers again, and with Julian Carver of Seradigm who had just allowed himself to be talked into chairing it. It became clear then that there was a tension between the un- and the conference, and Julian spent a lot of time working out how to manage this.

I should say at this stage that my respect for Julian, already high, has gone up hugely. It was controversial that he even accepted the job of chairing (which he did pro bono). Nevertheless he put all his formidable talents into the planning and the day itself. His performance was stellar.

And so on to the day itself – today, although it seems longer ago. There was indeed the the tension between an organised agenda and an unconference-style self-directed group. This was partly managed by calling for lists of topics before and during the session, then self-selecting into breakout groups to brainstorm these topics and report back. Of the attendees I spoke to, some seemed uncomfortable with the informal part of the day, and others with the formal part, so perhaps a balance was a reasonable thing to attempt.

Many of the formal presentations were very good. Hamish McArdle of the NZ Police, Minister Steven Joyce and the idea-per-second Glen Barnes stick in my mind. There was a panel discussion which was probably the best run of any such I have seen. (I did say that Julian was good at this.) And there weren’t any presentations coming in over Skype, video feeds or other potential technical nightmares. But, most of all – especially – there wasn’t any selling. Product and specific technology did not get mentioned. That was good for several reasons, perhaps the most trivial being that Oliver and I had an agreement in advance that he would have to supply me with a beer immediately for delivery offsite if selling occurred. I didn’t collect on that one.

The break out groups produced some interesting material, although a lot of it was reasonably anodyne. It’s clear that there is great enthusiasm to open government data – the NZGOAL framework at SSC is an example – and external encouragement isn’t needed, although welcome.

I was concerned by the output of one group which evaluated open source software as a key plank of open data – it concluded, perhaps predictably, that OSS wasn’t necessary for open data or even fit for government’s purposes. After all, who are you going to sue if it goes wrong? And do people know how much investment commercial software houses put into their products? These rather tired points have been rebuffed many times and no-one saw the point in doing so again at the meeting. That was the prompt for my acerbic tweet Microsoft unconference decides open source no good. Film at 11.

To me, the self-serving bit about the alleged unsuitability of open source was the only serious fly in the ointment on the day. The rest was at least neutral and often worth while. But, the intended beneficiaries of this day were the government folk who are opening up data as part of a move towards more open government, so I asked a few of them how they thought the day went. “Mixed” was a typical response. Like the curate’s egg, said one (hence the title of this post).

My overall conclusion here is that there was an attempt to make an elephant dance, in that Microsoft did at least try to run a session with a fluid agenda. Anything could have happened. The day was sell-free and mostly free of the kind of rancour which has poisoned Microsoft’s relationships with open source and open data people over the last few years. Politely, no one mentioned OOXML, and even when Clare Curran expressed her frustration that the Select Committee’s clear view on disallowing software patents has yet to reach some quarters, she refrained from rubbing the host’s corporate nose in its own support for software patents.

For Microsoft I suspect this was a walk on the wild side. Good on them at least for trying something. From the government’s – well, many of the government folk involved were well-used to informal bar camps. The bar campers wanted more informality and the others felt happier being lectured. From the open source, open data and general “want to help” types such as yours truly – the day gave an opportunity to spread the message wider.

To repurpose an old Data General ad: People are saying that Microsoft’s entry into open government unconferences will legitimise the field. The bastards say: welcome!

posted by colin at 11:56 pm  


  1. […] get too much into detail for the main parts of the conference, others (including Brenda Wallace and Colin Jackson) have done that already. But I’m keen to share some of my observations/things I’ve […]

    Pingback by “Open” Government 2010? | Human Factors — 29 June 2010 @ 8:41 am

  2. Thanks for summarising the day Colin. Really useful.

    Perhaps I wasn’t listening closely enough, but I didn’t hear the open source’s group’s report back as OSS not being “fit for government’s purposes”.

    I did hear them say “OSS is /less/ important than open data and open standards for the data”. I have been thinking about it, and I think I agree.

    Open access to data, in open standards, which are openly licensed, are all essential. Open communication and open policy making are also extremely desirable (others might argue essential too). I think open source software in government is often the right choice, and almost always has the lowest total cost of ownership, but I’m not sure I see it as fundamentally essential (I could imagine a hypothetical world where proprietary software produced open standards data that was openly licensed and openly accessible, and had a low total cost of ownership). If you can articulate why OSS is essential, I am happy to be corrected (I simply haven’t been able to work it out myself – which may well be due to my cognitive deficiencies).

    Comment by Matt Lane — 30 June 2010 @ 5:56 pm

  3. Thank you Colin for your kind words. A huge thank you also for your help in the planning prior, and during the day. Your support and knowledge were invaluable to me.

    Comment by Julian Carver — 1 July 2010 @ 2:27 pm

  4. I would agree with Matt’s comment in principle—open data formats are much more vital than open source code.

    But theory and practice usually turn out to be two different things. Proprietary vendors have this irresistible incentive to try to lock you in to their products. With no counterbalancing incentive to check this, why would they not succumb? Look at Microsoft and the whole OOXML debacle—after getting it approved as an ISO standard, even they can’t be bothered using it.

    With Open Source, the incentives are different. There is no way to make vendor lock-in stick, therefore there is no point in even trying.

    In short, once you take human psychology into account, you see even more reasons why Open Source is important.

    Comment by Lawrence D'Oliveiro — 1 July 2010 @ 5:23 pm

  5. As usual, Lawrence, you’re spot on (in my opinion).


    Comment by Dave Lane — 6 July 2010 @ 8:15 pm

  6. There are many examples of open standards promulgated through free open source software. There are many examples of proprietary standards and proprietary extensions to open standards promulgated through proprietary software. I am not aware of a single example of an open standard promulgated through proprietary software.

    So the evidence is that if we want open data using open standards, we need to use free open source software. Continuing to reward A while hoping for B is rarely a successful strategy.

    Comment by John Rankin — 9 July 2010 @ 7:49 am

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress