I spent Thursday evening and all day Friday attending Wikimania 2014 at the Barbican in London. For the uninitiated Wikimania is the annual conference for the Wikimedia community (that is Wikipedia and all its sister projects.) Like the Olympics or the World Cup the event takes place in a different country/city each time and there is a comprehensive bidding and voting process to select the successful proposal. It is considered quite an honour for the successful ‘chapter’ and Wikimedia UK have certainly made the most of the opportunity. Next year the conference is in Mexico City so I doubt my ‘travel & subsidence’ budget will stretch to that 🙂
I was invited to say a few words at a small fringe event on the Thursday evening alongside the CEOs of both Wikimedia UK and Creative Commons which was an unexpected honour. I spoke about the work ONS is doing to open up our data and also our burgeoning relationship with Wikimedia UK. It seemed to go down well with the crowd and spurred some interesting conversations that I need to follow up.
Friday afternoon I had been accepted to give a talk in the main programme about the work ONS has been doing to contribute to Wikipedia and to talk a little about some personal ambitions for future collaborations – not just from the ONS but the wider official statistics community. I was especially interested to talk to those in the know about Wikidata, a project that was something of a hot topic at this conference.
Despite a focus that I considered would probably be of niche interest around 100 people attended the talk and it spurred a lively Q and A session that I greatly enjoyed and left me with more ideas and inspiration than I quite know what to do with. I had particularly interesting conversations with attendees from Google, the Metropolitan Police, Wikimedia South Africa and the e-Government Institute in Switzerland. It was a nerve-wracking experience as they were a knowledgeable audience who asked many insightful, hard questions but I am very glad I did it.
The rest of my day was focused on attending sessions on the ‘Featured Speaker’ track. The event organisers had put together an all-star list of speakers from the ‘open data’ community and more widely.
Nigel Shadbolt, co-founder of the Open Data Institute, started the day with his talk about ‘social machines’. It was a different focus than I expected from a man I most closely associate with the ‘semantic web’ and the harder end of ‘web science’. I’m a big fan of Zooniverse, having even briefly worked with the team on a project, and was interested to listen to him weave a narrative around the power of computing + people to solve problems that computing alone could not resolve where Zooniverse was held up as an exemplar.
The next two speakers were Lydia Pintscher, the Product Manager of the Wikidata project, and then Markus Krotzsch, who is one of the technical founders of the project, which as I mentioned earlier was the hot topic on the day. It was referenced in almost every talk or conversation I was involved in.
Wikidata is a fascinating project that is seeking to create an open, query-able, machine readable, database of verifiable ‘facts’ that can be integrated with any Wikimedia project and beyond. This create some incredible opportunities and the Wikidata team have already created some impressive demos (the Siri like answer engine being a particular crowd pleaser.) For more of an idea of what is possible take a look at the Reasonator or for a very visual view of the potential power of the database check out Histropedia which has some very fun possibilities. Between Lydia and Markus I came away with a much better understanding of what they are trying to build I think and why it is potentially so important.
Viktor Mayer-Schonberger, from the Oxford Internet Institute, gave a talk about the power and possibilities of ‘big data’. It was pretty much a whistle stop tour of the big issues for ‘big data’ and what we can, could, should and shouldn’t do with it. It is a topic I find hard to really get a handle on – the data science needed to really comprehend it at any really useful level is beyond me and without that you are left with buzzwords and the same couple of success stories time and again (impressive though they are).
After a break from the sessions for my own moment in the spotlight I returned to the main room to listen to Mike Bracken talk about the work of the Government Digital Service. Mike is the Executive Director for Digital at the Cabinet Office and while it was a talk I knew well it is always useful for me to hear – the work of GDS is the standard we all need to live up to in digital Government projects these days. It was telling I think that Mike was the only person in the main plenary track to take questions direct from the audience – and the questions were far from soft balls. GDS talk a lot about being open and it is impressive to see the man at the top practice what they preach.
Mike was followed by Rufus Pollock, founder of ‘Open Knowledge’, who gave a thoughtful talk using shipping and shipping containers as a metaphor for open data – particular the discoverability and reuse of it. The talk was a triumph of staying true to your metaphor of choice and I was particularly pleased it worked towards a pitch for the use of ‘data packages’ which we’ve already started to experiment a little with as a lightweight way of providing our data.
The final talk of the day (for me) was from Peter Murray-Rust, a ‘reader’ of molecular chemistry at Cambridge University and advocate for open science, open access and, as he said in his talk, pretty much open everything! Amongst other things Peter can lay a strong claim on coining the term ‘open data’. He certainly created the entry for the term on Wikipedia. He led the campaign against the use of PDF in scientific publications because of the lack of machine readable data included and has been working on a new tool, ContentMine, that uses data mining techniques to make sense of the huge number of impenetrable publications out there and provide researchers with actionable data. The tool can even make sense of those PDFs now – including the ability to retroactively take PDF’d tables and turn them back to CSV.
It was Peter who used this term of ‘digital age of enlightenment’ with the flair only a well practiced public speaker can make. He was talking about how the power of things like Wikipedia to make knowledge freely available to huge amounts of people should be embraced by all – including scientists as this was an opportunity to change the world.
Inspiring stuff and appropriate for what was an inspiring day all around. Sadly I was unable to stay for the rest of the conference but based on Friday alone I can say it was a huge success and the team behind it at Wikimedia UK deserve congratulations on a job well done. It is the community that makes these things work though and again I was amazed by the generosity and passion of the open community. Keep up the good work.
Now about those monkey selfies…