Consumer Focus: From broadcast to conversation – making open data work for you workshop

Spent this morning at the consumer Focus "From broadcast to conversation – making open data work for you" workshop aimed primarily at the voluntary sector making use of Open Data. First speaker Mark O'Neil from the cabinet office and part of the recently formed Government Digital Service (DGS) - he talked about the 'single domain' which is the governments replacement of direct.gov. The website is built around a API allowing data to be shared, however it is currently a beta and provides services. He also talked about 'agile delivery' methods, which equates to the wisdom of doing things quickly in a focused manner, rather than spending years building a massive system that is either inappropriate for use or no longer relevant. (Which is a bit similar to James Wallbank's promotion of 'good enough ICT') Mark stressed that he spent a lot of time talking to users, which seemed primarily to mean business users who provided services to what I would call the 'general public'. Within government he identified four areas of operation, at the level of citizen, business, inside government and specialist area of operation. He was then asked about the issues of read /write access and admitted we are a long way of that yet. He was up for email conversations though, contact mark.oneil@digital.cabinet.office.gov.uk.

Next up was Ed Parks who also worked for the cabinet office (is there a theme developing ?), in the transparency team. He talked a bit about the process which promoted Open Data to its current position as an explicit part of Open Government project (this came about as a result of a request from the PM), and although a project of transparency and citizenship, was mostly justified against business metrics. As he stated "Open Data as an engine for growth". Someone in the audience queried the method for deciding which datasets are released, as his company had come into direct competition with another government agency. The point was that the decision to publish particular datasets is driven by financial and political agendas, and that Open Data was used, as the last speaker said, 'as a Trojan horse'. Ed mentioned the the public data group was recently set up to offer the possibility to query the governments policy, and we should all take part (address here) He also mentioned a recent Guardian article (here is a response to it) that addressed some of these concerns, other initiatives included the setting up of the Open Data Initiative in Shoreditch, to help small business take advantage of Open Data, the Open Data/Privacy white paper and the Open Business forum and the government consultation group that featured the stars of Open Data promotion in this country (Rufus Pollock (OKFN) , Tim Berners-Lee W3C and Nigel Shadbolt. Cynically Harwood's  "people who are awaiting power'  springs to mind.

Next up was Dan ? from Nominetrust ?. He consulted with charities in their possible use of Open Data. These included wheelmap.org, RSPB and Barnsley hospital. He gave the example of the toilet map whereby a service (http://toiletfinder.org/about)  (there is also one built by  a charity around incontinence issues) also acted as a campaign tool, as in 'Why are their no toilets here". Dan cited the Indigo Trust as an organisation that had done considerable work in making use of and talking about the difficulties of implementing Open Data practices in charities.

Javier Ruiz Diaz from the Open Rights Group talked about the realities of Open Data publishing. He talked about three basic types of government data, PSI (Public Service Industry) this is core data - geotagged public data, this is where the money is (OS, Census etc). The second Policy/Accountability and the third public services. He also made reference to the use of Open Data as a Trojan Horse under which government ministers (Francis Maude) could include plans to share welfare data with private company Experian to reduce welfare fraud. He cited a recent UK based survey that the majority of the public want public data to Open and available. And again referring to the usability of Open Goverment data he cited the example of the publishing of a central government data set which contained all the expenditure for a particular year which was largely unusable due to the lack of metadata to contextualise it. So the metric used for volumes was not detailed 300 what ? the frequency per year/month/week ? etc, rendering much of the data unusable. He also pointed to some non commercial projects which have made use of Open Data, the Europeana has published Open Data about may of the cultural archives held across Europe, and on speaking about broken data cylces (i.e. the government not allowing read/write privileges of open government data) he invoked a distributed methodology whereby the dataset with the most users would be the one that was used, and would be default by the one that was most uptodate. Some interesting companies were mentioned most prominently Swirrl who provide consultancy to large public entities, in this case Hampshire County Council.