Open data filing cabinets or open data clubs?

Do successful open data portals work as filing cabinets or open data clubs?

Updated: A few people have asked about examples and illustrations so I’ve updated the post.

As part of the Media Mill project, building on thoughts about ways to think about open data ecosystems, we’ve been looking at what factors drive ‘successful’ portals.

It’s not new (or surprising) to note the key role sustained community building has in successful open data engagement. In our research, the successful open data portals all manifest a level of community involvement that is encouraged either through an initial stage of community engagement which has led to a level of autonomy for the portal or through sustained community engagement and an open door policy to the sources of data on the store.

Organisational hub — open data filing cabinets

Many open government data portals represent an organisation’s structural aims. Whilst they enable data users and intermediaries, the engagement value is often reduced because the main driver for the choice of data is driven by internal motivations rather than user demand. Whilst there is still value to be gained from interaction with the data, this is often weighted to service reform/infrastructure benefits to the organisation. This can create a public facing side of a portal that is more like a clearing house of data akin to an unlocked filing cabinet in an empty room. Interaction between the broader inhabitants of the data infrastructure still remains but often the opportunities for Citizens to interact with the data are often limited to traditional ‘front door’ services provided by the government organisation.


  • York Open Data , like many open data portals takes its transparency and accountability data and extends the functionality of standard transparency pages to take a more structured approach to publishing and supporting internal moves for more effective and efficient use of data in reporting
  • The London Borough of Redbridge’s Datashare portal was developed specifically to address data needs within the council and the portal reflects the idea of creating efficiencies in the process of publishing data. The Redbridge example does highlight that opportunities to release broader value are still there in these models of portal. The Datashare platform was ‘spun-out’ as a commercial offering and a large percentage of the data portals run by local councils use the Datashare platform.

Community hub — open data clubs

When a government organisation moves a data portal to an arms length (management and governance happens outside of government organisation) they frame themselves as a one ‘spoke’ of hub that often contains a broader range of stakeholders. Government organisations are more likely to also become consumers of their own data — often through projects managed and championed at a community level. In some cases the community can represent a club and single issues can dominate. But the club is in principle and practice open to all members which means in the this model citizens are also, conceptually at least, more proximal to the sources of the data and there may be a diversity of sources which is more attract attention.


  • Leedsdatamill: Although the costs of setting up the DataMill have been met by the Council, there was a good deal of support for the idea from community groups like Leeds Data Thing. Whilst the portal itself remains a council project, it is managed outside of the council and the platform operates as a city data hub used by a number of organisations and events in the broader open data community e.g Leeds Data City
  • BathHacked: Like Leeds Data Mill, Bath’s open data portal had initial funding from Bath Council but is now run as a CIC (Community Interest Company). The relationship between the CIC and the council offers an interesting insight into a reciprocal model that has evolved to underline the social value of open data.

Measuring success

What constitutes a successful open government data portal is, obviously, up for debate. The ideas above might represent a simplistic, binary view. We recognize that often an internal, data infrastructure focus, can be as big a win in encouraging a level of understanding and engagement with open data that might pay off further down the line e.g Trafford Innovation and Intelligence Lab.

There are also examples of emerging community engagement activity as portals develop. Bristol’s Open Data portal, like others developing alongside Smart Cities or similar initiatives has focused on developer communities in its outreach. Projects like B-Open which encouraged hack and make activities to promote open data use. Calderdale Dataworks is a good example of a developing organisational portal. The initial set up involved a visible level of community engagement at its launch (particularly during the Wuthering Bytes festival). As one of the founder members of Leeds ODI node, Calderdale also broaden the opportunity for engagement but this may present new set of challenges aligning local interest to the broader aims of the collective members of the ODI node.

However, in order to give the best possible chance to develop value and encourage innovation from the data on open data portals — what we might think of as the golden halo around data — an engaged supporting community is a key element of creating sustainability. Rather than focus on narrow interests in thematic datasets or pragmatic data releases, emerging data portals are likely to be more sustainable if they encourage communities to form or develop around data portals with a view to sharing control.

What do you think?


* These examples highlight a blurring of the definition of open data portals. We are making a distinction here between transparency portals and open data portals. All councils have (or should have) sections of their website where transparency data is published. This is often in open data formats and released under open data licensing which would make them, by a broad reading of the term open data portals. However there are a number of advanced features of open data portals that mean even though they perform the same core task of publishing transparency data, create the opportunity of more value creation e.g the addition of api access to the data. For more on the issues that open data portals address see van der Waal, Sander et al. “Lifting open data portals to the data web.” Linked Open Data — Creating Knowledge Out of Interlinked Data (2014): 175–195.