The European Council is encouraging public institutions to provide information as open data in the context of administrative modernisation, better services and transparency. As the General Secretariat of the Council is the public administration supporting the Council of the EU and the European Council, we have taken this call to heart and decided to embark on a pilot project to scope out the practical challenges involved in providing data that we already make available in as linked open data. Adding a semantic layer to our information could also allow us to provide our clients with a much richer experience by simplifying information retrieval and by providing context on multiple dimensions (theme, meeting, process).
In addition, we are currently undertaking a major overhaul - conceptually and technically - of our information, data and knowledge management approach. As information provision to our clients (EU member states) is our core business, we are exploring semantic technologies as a core component in this overhaul. We are currently in an exploratory stage, but see an important role for semantic technologies in the way we provide information to our own staff, by allowing them to retrieve information and formalised knowledge more efficiently.
We aim to transform the General Secretariat of the EU Council into an organisation, which places a strong emphasis on how information is created, managed, and aggregated into knowledge. Therefore, we have defined that "knowledge management is the ultimate objective of all information management activities" as one of the guiding principles of our Enterprise Architecture framework. This is a challenge as it is a daunting task due to its complexity. We distribute far over 30000 documents to 28 member states, often in 24 languages. We manage the content production of 3000 staff, we ensure that the information is made available to those who need it and preserved for posteriority. And it is also a challenges, because we are only in the process of developing a policy-based approach to the question of semantics and open data. At the same time it is an opportunity, as both the information management backend and its public face are being overhauled and the two elements can be established in a way that fits together.
The first pilot project - and the focus of this presentation - has been a first test to identify the opportunities and challenges that this approach brings to the public and to the organisation itself. The EU Council is, together with the European Parliament, the co-legislator and all act adopted at EU level are voted on by all 28 EU member states. It is a crucial player in making law in the European Union and represents the interests of EU member states. Therefore we decided to make transform information into our first open data set that is of high interest to researchers and civil society and that represents a core part of the EU Councils role as a legislator: the results of EU Council votes on laws and other acts. This information has for a long time been available to the public. But it had been stored in a closed data base for internal use. Citizens, researchers and journalist were only be able to access that information through pictures embedded in pdfs . Now a linked open data set of the EU Council's voting records from 1.12.2009 (post-Lisbon) onward is available on the EU Council's website. Users can download the data, browse it and query it using an API.
The project implementation has been a great success and has led to a significant number of lessons learned. These include the need to not only shoot for the sky (5 star data), but to take the capacity of our 'normal' users more into account; the need to build up knowledge and capabilities on semantics and open data in house and the need to include reflections on semantics and open data in the how data/information generation process, from creation to destruction/archiving.
Wilfried HOC, Directorate for CIS, General Secretariat of the Council of the EU,