Infrastructure, collaboration and open data “scenes” - reflections on my first Open Knowledge Festival
There will likely be many blogs about the recent Open Knowledge Festival in Berlin but, as it was my first time and because I came away continuing to think about a number of issues, I decided to write about what I took from the event.
As with all conferences, a large part of the value of the Festival came from having all of the people from your community in the same place at the same time. At points, it was a bit like my Twitter feed had come to life and I was fortunate to have a number of rewarding conversations with new as well as familiar people.
The discussions were both about the content of the keynotes and sessions as well as the missed opportunities. For me the keynotes were the highlight of the Festival and, in particular, those on the second day have continued to get me thinking about the state of the open data agenda.
Open Data Stagecoaches and Turnpikes
The keynote which a number of the people I talked to mentioned was the one by Eric Hysen, who leads Google’s elections and civic engagement programmes. The full text of the keynote can be found here and the accompanying slides here.
Eric presented a three stage recommendation to the broader open data community based on their analysis of current challenges. These recommendations were (and I paraphrase so apologies if I miss some of the sense):
1. Open isn’t enough - just releasing open data isn’t enough, you have to make sure it’s machine readable, you’ve made clear what the license is and you’re committed to keeping it updated;
2. Interoperable data - here Google was suggesting that as much as possible data should be released as APIs; and
To support this recommendation Eric used the analogy of the beginnings of the English road network, where new technology, in the form of the stagecoach, was created on the back of collaboration to build better trunk roads. This collaboration was between sectors with shared interest, such central and local government and commerce, coming together in turnpike trusts which built and maintained the roads.
Leaving aside the fact that these recommendations were made by an organisation which in some parts isn’t viewed as always being collaborative, this contribution was mostly met with respect and consideration by those I talked to afterwards.
It also appeared to be a well timed intervention in the development of the conversations at OKFest - a firm but gentle nudge to push the conversation to develop a bit more.
Recognising the power of analogy, I suggested in a tweet that Eric’s presentation was a bit like a rich, distant uncle coming to Christmas dinner, telling you what you’re doing wrong with your life and, of course, being completely right. The keynote also demonstrated a slickness and mastery of rhetoric which, through its presence, highlighted a lack in the rest of the offerings in the Festival.
UK’s Open Data Infrastructure vs Ecosystems
Taken on face value the main analogy of the presentation made me reflect on our own “infrastructural” organisations in the UK - such as Open Knowledge, the ODI, MySociety, Nominet and indeed Nesta - and what role they have in developing shared infrastructure as well as supporting the development of an ecosystem.
Several of these organisations are already collaboratively developing parts of the open data infrastructure. For instance, the ODI with their Open Data Certificates and MySociety with their Poplus components, making available components of collaborative software for civic sites - not to mention Open Knowledge bringing everyone together at the OKFest in the first place.
No doubt there is still room for greater collaboration and co-ordination, and given the Nesta representation at OKFest, I suspect this will spur some discussion internally about our role.
Open Data "Scenes"
However, I’m not sure I see an easy distinction between activity to support the development of a shared open data infrastructure and that which supports the broader development of an ecosystem. Do the growing open data “scenes” in the UK’s cities such as Manchester, Leeds, Glasgow and Sheffield amount to development of local infrastructure or different ecosystems?
For me, the challenge is how to collaborate on an infrastructure whilst also maintaining our own open data “scenes” tackling the local issues and speaking to the motivations of those involved. This speaks to some of the broader issues picked up in the Google research i.e. around capacity and motivation for engagement in civic technology.
For all the rational argument for standardisation and collaboration on infrastructure, the open data community is not a homogeneous and neatly organised phenomenon but more a series of interlinked “scenes”, much like other movements such as the music industry.
I suspect this might make collaboration on infrastructure more difficult, unless there is a clearer narrative that appeals to all about the benefit and, at the same time, how local efforts and individual open data products fit in the mix.
From Turnpikes to Motorways
The analogy of the stage coach also begged the question about what happens after all this collaboration. Eric’s analogy stopped with the development of the stagecoach, but the English history of road development didn’t stop there.
Should we deduce that Google also supports the further development of the collaborative open data infrastructure as being taken over by the state and funded through taxation, similar to the way that we provide roads at the moment?
Potentially one role for Government here is, rather than set out and over specify the infrastructure from the outset, to allow it to develop, potentially by being matchmaker for those with similar interests in seeing it develop, and then once it’s more mature adopting it into state provision much like the way in which roads became nationalised. Or indeed how platforms such as Twitter adopt innovations in their ecosystem, by incorporating features from apps in to the main body of the platform over time.
Overall, I was impressed by the Google contribution, which was expanded on in other sessions and will be followed up with published research. It’s nice to have some research in this area and hopefully these contributions will continue to spur further discussions.
Thanks also to all at Open Knowledge for getting us all together and mixing us all up in Berlin. It seems that the work now, if we're convinced about the need for collaboration on infrastructure and ecosystems, is about how we go about doing this with the activity we already have.
Photo credit: Sticker from Open Data Manchester