Added by Maarja Toots 2019/08/31
The European Data Portal (EDP) finished another round of data collection this summer, asking EU member states to report the status of the supply and use of open data in their country. The fifth edition of the pan-European Open Data Maturity Landscaping survey measures countries’ progress in four dimensions: open data policy, open data portal, open data impact and data quality. How is Estonia doing compared to other European countries?
We didn’t do particularly well last year – Estonia’s ’maturity score’ of 0.43 on a scale from 0 to 1 placed us fifth from the bottom among the 31 participating countries, just behind Latvia. Ranking in the top were Ireland and Spain with scores of 0.87 and more. The survey involved 27 EU Member States (Hungary did not participate) as well as Iceland, Liechtenstein, Norway and Switzerland.
Figure: Estonia’s results in the 2018 EDP open data survey (source: EDP)
As it appears, Estonia’s regulatory framework was given a high score according to European standards, also because the Estonian law states that public sector data are public, redistributable and reusable by default. However, the results were not as good as regards the quality and impact of the data published on the national open data portal. For instance, in the previous version of the portal, datasets were mostly updated manually without a clear process for updating and improving the quality of metadata. Another shortcoming was the poor findability and cross-border interoperability of the data, which mostly was due to the portal not following the DCAT model for data descriptions, which is recommended at the European level.
However, the actual impact of open data was considered even more problematic. It must immediately be said, though, that EDP’s approach to evaluating impact is rather (perhaps even too) simple – the larger the number of open data-driven applications, services and processes, the bigger the presumed impact. The number of reuse cases is the basis of measuring the impact of open data on policy-making, social issues and environment (or public awareness of the environmental situation to be more precise). The assessment of economic impact is a bit different – in this case the survey looks for evidence of studies assessing the economic impact of open data in the respective country. As there were no such studies to report for Estonia, the score in this dimension remained zero.
To some extent Estonia’s poor result last year was due to unhappy circumstances. Although the year 2018 marked several important developments in the Estonian open data landscape, the data for EDP’s survey was already collected in May, at a time when many of these developments were still under way. As a result, Estonia’s ranking did not reflect events such as the relaunch of the national portal opendata.riik.ee on a new platform, the addition of tens of new datasets to the portal, or new requirements on metadata which give users more detailed information on data quality – and which also follow the common European DCAT-AP standard, enabling aggregators such as the Eurpean Data Portal to automatically harvest data from the portal.
The good news is that all these developments could be reported in this year’s survey and, based on EDP’s initial analysis, we have a reason to expect considerable improvements in Estonia’s score. Where exactly Estonia ranks compared to other countries will become clear around November when the survey results are expected to be published. However, according to first signs, we are gradually beginning to move closer to current open data benchmarks, as defined by European experts and trendsetters.
This said, it would be premature to sit back and rest on our laurels. Looking at the speed at which Europe’s understanding of open data is evolving, doing open data ‘right’ is kind of a moving target. Whereas some years ago countries were still working on creating a legal obligation for the public sector to publish open data, it is no longer sufficient to adopt policies and publish new datasets on the portal. Instead, countries need to learn to understand and assess the actual use and impact of open data.
On the one hand, impact is an important dimension in EDP’s landscaping survey, whose methodology assigns impact an equal weight to open data policy, portal usability and data quality. For instance, Ireland’s recent success in open data (from 18th to 1st position in a couple of years) is related to the country’s dedicated efforts to improve the quality and usability of the data on the portal. To this end, Ireland is paying close attention to detailed documentation of the datasets on the portal to help users really understand and reuse the data – check for examples HERE. For all last year’s top three countries – Ireland, Spain and France – EDP’s study valued their strategic approach to publishing data and assessing its impact. These governments have defined open data reuse as a strategic objective, defined high-priority datasets that should be accessible to the public, and are working to continously monitor the supply and use of open data. France sets a good example here with its systematic approach to assessing and increasing the impact of open data. Etalab, the French governmental agency responsible for open data, has actively worked to bring data users and providers together to identify datasets that are important for users and make sure they get published. At the same time, the French think tank FING is running a multi-year project with support from Etalab that aims to increase the impact of open data by 2025. The project has three streams of activity: analyzing the progress so far and establishing the baseline, encouraging innovative and experimental ways of stimulating the provision and use of open data, and developing a roadmap stating the next steps for increasing the strategic impact of open data. Those who speak French can learn more about the project HERE.
On the other hand, EU Member States recently adopted a new open data directive, which introduces the concept of ’high-value datasets’ to require governments to make datasets with significant economic and social impact available free of charge, with a license permitting reuse, in machine-readable formats, through APIs and, if possible, as bulk download. So far, states have identified six categories of data as having particularly high value for the public: spatial data, environment and earth observation, meteorological data, mobility, companies and company ownership, and statistics. The identification of specific datasets under these categories is still in progress and more clarity is expected towards the end of the year. In any case, these developments seem to be signs of the emergence of a new open data paradigm centered around reuse and impact, which Estonia – an EU Member State and experienced e-government – has the opportunity to actively shape. In order for the shift not to take us by surprise (and drag us down in rankings), data providers would do well to already start asking who and for which purpose is using their data. If the answer is not clear, it is high time to start a dialogue with existing or potential users, for example by organizing a hackathon dedicated to the reuse of specific datasets, attaching a simple feedback survey to the published datasets or asking users to reveal themselves on the GitHub issue tracker.
The Open Data Portal's content is created as part of the EU structural funds' programme 'Raising Public Awareness about the Information Society' financed through the EU Regional Development Fund. The project is implemented by Open Knowledge Estonia.