Impacts of open data: disintermediation

The following post is an excerpt from my thesis entitled Linked open data for public sector information.
Who draws and controls the maps controls how other people see the world [1]. Who interprets data from the public sector controls how other people see the things described in the data. By releasing raw open data the public sector also releases its total control over the interfaces in which the data is presented. In this way, the interpretive dominance of the public sector data is abolished and it no longer controls the way how citizens should see the world described in the data [2]. Civil servants perceive this as a loss of control over the released data, but in fact, it is only a loss of control over interfaces in which the data is presented.
Providing raw data is an example of disintermediation. It reduces the frictions and inherent cognitive biases that come with interpretations by intermediaries. It allows users to skip the intermediaries that stand between them and access to raw data. For example, both civil servants producing reports based on primary data and journalists transforming data into narratives conveyed in articles serve as intermediaries that affect how the public perceives public sector data.
Depending on the type of use mediation may be either a barrier or a help. It is a barrier for those that want to access raw data to interpret them themselves. However, common perception has it that too few people are interested in raw data [3, p. 71]. Yet one should not make such generalizations as there is evidence that suggests otherwise. For example, after the release of data from the Norwegian meteorological institute, the institute registered more data downloads (14.8 million) than page views (4.5 million). These numbers were given by Anton Eliassen, the institute’s director, during the first plenary on the revised public sector information directive at the ePSI Platform Conference 2012. In general, it is the case that raw data receives relatively few downloads, yet access to raw data is vital to build new applications on top of the data.
Disintermediation creates a demand for reintermediation. Mediation helps users that need to get user-friendly translations of data in order to reach understanding. Applications mediating data in ways that are accessible and compelling, such as visualizations, may attract a lot of attention proving the demand for public sector data. For instance, this has happened in the case of the UK crime statistics, the visualization of which crashed under the weight of 18 million requests per hour at the time it was released [4].

References

1. ERLE, Schuyler; GIBSON, Rich; WALSH, Jo. Mapping hacks: tips & tools for electronic cartography. Sebastopol: O’Reilly, 2005, 568 p. ISBN 978-0-596-00703-4.
2. BARNICKEL, Nils; HÖFIG, Edzard; KLESSMANN, Jens; SOTO, Juan. Organisational and societal obstacles to implementations of technical systems supporting PSI re-use. In Share-PSI Workshop: Removing the Roadblocks to a Pan-European Market for Public Sector Information Re-use [online]. 2011 [cit. 2012-03-08]. Available from WWW: http://share-psi.eu/submitted-papers/
3. HALONEN, Antti. Being open about data: analysis of the UK open data policies and applicability of open data [online]. Report. London: Finnish Institute, 2012 [cit. 2012-04-05]. Available from WWW: http://www.finnish-institute.org.uk/images/stories/pdf2012/being%20open%20about%20data.pdf
4. TRAVIS, Alan; MULHOLLAND, Hélène. Online crime maps crash under weight of 18 million hits an hour. Guardian [online]. February 1^st, 2011 [cit. 2012-04-17]. Available from WWW: http://www.guardian.co.uk/uk/2011/feb/01/online-crime-maps-power-hands-people