On Saturday 27 March 2010 I attended the Rewired State: Rewired Culture event. This was the latest in the series of events organised under the Rewired State banner which, we heard, aims to address the difficulties in adopting rapid developments approaches in central government.

The event, held at the Guardian office in  King’s Place, London, was organised by Rewired State (a non-for-profit organisation) together with the DCMS. The context to the event was summarised on the Web site:

Britain is a creative culture. We have a vibrant developer community, a growing and active entrepreneurial base and a vast, rich array of culture assets. How can we bring these together to create new opportunities for data owners and developers? How do we encourage links between data repositories such as museums, broadcasters and the wider community like data.gov.uk or the “London Datastore”? How do we ensure that the exciting work already underway in a number of organizations is shared more generally, so even smaller bodies and SMEs can learn from best practice and find workable routes to market? What are the cultural content business models for the 21st century? How do creators, curators, developers and entrepreneurs work together?

The were two strands to the day. The first was the hackday which brought together data owners, data users, developers and people with ideas to see what they can create in a day. The second strand was a half-day unconference style event which  started at midday and ran  in parallel with the hackday and sought to engage  data owners, entrepreneurs, data users and communities in discussions on business models, funding mechanisms and challenges.

The Unconference

As I am not a developer I attended the unconference.  From the show of hands at the start (together with discussions I’d seen on the Twitter back-channel – #rsrc was the hashtag for the day) I was aware that a number of the participants were unfamiliar with the concept of an unconference. In order to try and help avoid the embarrassment of finding no suggestions for topics for the first unconference session, I proposed a session on ‘Avoiding institutional barriers‘. I was pleased that there were about nine  people who had an interest in this topic. My interest related to UKOLN work for the cultural heritage sector in supporting best practices for cultural heritage organisations wishing to exploit the potential of Web 2.0.   We have recently completed a series of workshops on Web 2.0, which have enabled us to share best practices and also to hear about barriers faced by institutions wishing to implement Web 2.0 services themselves.  I was looking for an opportunity to hear if others were facing similar barriers and how they were addressing such challenges.  However in the discussions it became clear that although there were interests in this, there was more interest in the barriers faced in making use of open data.  The concerns that were mentioned included the reasons for providing open data; identification of who the beneficiaries of such open data would be and the business models for engaging in new areas of work at a time when large-scale cuts in public sector funding are in everybody’s mind.

There was useful discussion about the relevance of open data as a provider and as a consumer – issues which had been raised at a recent meeting on Linked Data which I attended and helpfully summarised by Jeremy Ottevanger in his post on Linked Data meeting at the Collections Trust. We heard that as more open data begins to be provided by, for example, central government departments, this raises issue about the quality of the data and the context of its use and reuse.  We heard that for museum’s collections database, there may be large number of catalogue records simply stating ‘coin’ or ‘shard’ – making such data which was previously only available within the museum  made lead to both doubts as to whether useful services can be developed using such data and concerns that poor quality data could be used against an institution – and with cuts looming would institutions be willing to see their dirty washing?

Although such concerns are legitimate the people in the unconference group shared the views express recently by Sir Tim Berners-Lee who argued that public sector organisations should be opening up access to their data, even if it had deficiencies. It is only by opening up the data that improvements to the data, and the underlying workflow practices seemed to be the shared consensus. But whether this view will be shared by senior managers and policy-makers may be questionable.

This first unconference lasted for an hour. After a coffee break I attended an unconference session on ‘Why are museum APIs so poor?‘ – although this session was subsequently merged with another session on ‘The role of museums‘ so the discussions were not as focussed as I personally would have liked. We heard from a number of developers in museums and archives who were frustrated at the difficulties in exploited data provided by others due to the limitations of the APIs. Unfortunately as the group was so large it was difficult to pin down whether the problems were due to the APIs themselves or the underlying data or data modelling.  I did mention UKOLN’s Good APIs project which produced a report on best practices in both providing and consuming APIs which had been developed in conjunction with developers in the JISC community.   However this work did not address the issues of the underlying data, which is particularly relevant to the developers form museums who were participating in the unconference session.  The subsequent discussions there was a view that in order to make available the richness of the cultural resources  which are available there will be a need for greater investment in providing and managing the metadata in order for interoperable systems to be deployed. However the likelihood that a successful case for such increases in investment was felt to be questionable.

The Hackday

After a break for beer and pizza the developers arrived in the main room and gave a summary of the development work that had taken place between 9 am and 6 pm. I must admit I was impressed with the ideas that had been explored and, on a number of occasions, the prototype which was demonstrated.
There were, I think, fourteen presentations made, but not all of which managed to develop a prototype which could be demonstrated on the day. A list of five of the projects for which software was developed is available on the Rewired State Web site.

I will summarise a couple of the presentations which I felt were particularly noteworthy.

Josh Pickett and Isabell L, two 15-year old developers, produced Landing Zone which they described as “the *ultimate* tool for crop circle enthusiasts around the country and indeed the globe. The LandingZone app fetches co-ordinates from cropcirclearchive.com and plots each occurrence on an interactive map (courtesy of google maps). The crop circles can be filtered to view only the most relevant circles that appeal most to you, each map can be filtered by crop type and by date. The data used to build the app is also freely available and each page of the app has a direct link to the source googledocs spreadsheet.”

Julian Burgess (@aubergene) developed HMRC Artworks. Julian described how many works of art are donated to the state in order to avoid inheritance taxes, with the condition that access to the art works should be provided to the general public. The HMRC Artworks prototype aims to enhance access to information about the availability of the works.

Presentation by Richard Light and Chris Foulston

Richard Light and Chris Foulston (shown in photograph) converted 7,000+ events from Visit Britain into a map layer. A value aspect of this work was the summary of the difficulties they experienced in processing the data from various sources. As they described “We reviewed as many of the provided datasets as possible, looking for common data points such as people, events or places on which we could do a linking exercise. We drew a blank on this, and instead did a mapping exercise based on NGR co-ordinates.” They also described how they “had issues with quotation marks in the CSV data (which can be confused with field delimiters) which caused errors in mapping the data to XML“.

For me, however, the most impressive demonstration and funniest accompanying presentation was given by Rupert Redington of Neontribe who described the Awesometer – a response to Jeff Gilfelt’s ASBOrometer. However rather than the ASBOrometer’s approach to spreading fear and distrust (it has been described in the Daily Mail as an application fore those who “want to know how dodgy their area is“) the Awesometer “gives hope, security and warm feelings” . More seriously, the Awesometer is “an app for mobile devices using the Android operating system, it uses a wide variety of open data sources to find the number of positive actions, institutions and events near the current location“.

In his talk about the development of this Android app Rubert described the APIs the app used: http://feeds.plings.net/ to find out what positive activities are nearby; http://data.gov.uk/ to find nearby schools and other educational establishments and http://culture-grid.org.uk/ to find nearby cultural institutions. In addition a temporary Talis datastore holding event information taken from http://visitbritain.org.uk/ was also used.  Rubert also described that, as he didn’t own an Android phone, he used an Android emulator to develop the app, which he used in his demonstration.

Discussion

The presentations given by the mashup developers made me aware that processing and visualising data stored as a Google Spreadsheet seems to becoming an easy way of rapid development of certain types of Web services – taken some data provided as a CSV file or MS Excel spreadsheet from data.gov.uk, upload as Google spreadsheet and then use various processing and display tools to enhance access to such data.

But other experiences confirmed the concerns raised in the unconference discussions – when data is reused by others, the inconsistencies or errors in the data can lead to the data being unusable or misleading services being deployed.

I feel that the Rewired State: Rewired Culture Event proved successful in demonstrating that useful prototypes could be developed quite quickly. And even when it was not possible to provide live demonstrations the event was also useful in explored new ways in which open data could be exploited.

But there will be barriers which will have to be addressed. I would hope that the vision described by Gordon Brown of an open government providing access to non-personal data will be realised, and that the technical leadership of Sir Tim Berners-Lee will ensure that a realistic technical understanding is applied to the political rhetoric. It will be interesting to see how this pans out after the General Election