The aim of this brief was to carry out awareness-raising in Sweden to engage stakeholders in discussions about the situation with open data and to understand their needs to enable actions towards further data openness.
Method
The awareness raising was based on the case study findings and included interviews with different types of stakeholders working with and/or engaged in open data topics (i.e. academics, public officials, journalists and active users/practitioners of open data, and people promoting and/or using open data) This resulted in a brief and an Open Letter to public officials and policy makers with recommendations about the way forward in the field of open data in Sweden.
Background: open data in Sweden
Data produced by public actors constitute a comprehensive and valuable resource that can benefit society as a whole. According to a recent report about high-value open data made by the Swedish Mapping, Cadastral and Land Registration Authority (Lantmäteriet), the value of these data is estimated to be SEK 10-21 billion per year for Sweden.
Sweden is a leader in Internet usage and digitalisation, and has a long tradition of openness and public access legislation. At the same time, several international and national studies show that Sweden is underperforming when it comes to open data. The release of open data has a low priority among public sector authorities. The publication of open data is often delayed due to feelings of uneasiness regarding the quality of data or doubts about the legal implications of disclosing data (e..g regarding GDPR). Some of the challenges are also related to Sweden’s decentralised administration, where the responsibility for making data available lies with individual authorities, regions and municipalities. These often lack the awareness about open data and the knowledge gap is especially apparent at the higher levels of leadership. The resources to publish data are also lacking. When the data is published it is often considered as a one-off exercise, instead of a continuous process and system that should be in place. The connections between open data and other important areas, such as AI, IoT, digitisation and transparency, are often missed.
In the view of the upcoming national elections that will take place in 2022 and the upcoming development of the next OGP plan Sweden (also due in 2022) it is a good moment to raise awareness about open data for public good in Sweden.
Open data priorities in Sweden
According to the interviews conducted for this brief, these should be the open data priorities in Sweden:
Invest in basic public data
These could be defined as high-quality, standardised public data that can be used across systems, countries and sectors to create value and expand knowledge. Basic public data is basic information about our society contained in registers with information in other areas such as property, buildings, geography and climate. These data, especially if combined (e.g. company and social security data), are much needed by many actors ranging from tax authorities to journalists. At the same time, there are challenges in terms of handling risks to personal data security when handling these large and linked datasets.
Sweden could learn from Denmark’s “Basic Data Programme”, in place since 2013. This programme collects basic data across the country’s IT systems, so that the public and private sector can use them efficiently, and has adopted a common solution for distributing basic data (The Data Distribution Platform). The Platform provides authorities and companies with easy and secure access to basic data in one joint system rather than many different systems and interfaces. This gives a joint saving on operational costs and support, while the data reusers can find information all in one place.
Follow “high-value, high-impact” and “publish with purpose” principles
According to the doctoral thesis of one of the interviewees, Jonathan Crusoe, open data practitioners “should follow the principles “high-value, high-impact” and “publish with purpose” rather than “open by default” or “raw data now” principles”. Crusoe asserts that “open by default” and “raw data, now” might build momentum and are useful for building open data skills in the short run. However, the released data might not be used or of low quality. In the long run, these could be used as arguments against open data, especially considering that the costs might outweigh the benefits. Being context-specific, high-value and high-impact data have to first be defined at the national and local levels in order to create value for open data actors and the related ecosystem.
The work on defining high value datasets at the national level has just recently begun in Sweden, prompted by the EU’s Open Data Directive that requires Member States to publish a list of high-value datasets free of charge. The EU considered these datasets to have a high commercial potential and the ability to speed up the development of value-added European information products, as well as to serve as key data sources for the development of Artificial Intelligence (AI).
Types of data that should be prioritised
The interviewees also mentioned specific datasets that should be prioritised:
- Data on public procurement (advertisements, award decisions, contracts, lists of procured suppliers, etc.)
- Government economy systems with transactions ranged in different accounts.
- Authorities’ diaries (registers) of public documents that can be requested by using freedom of information (FoI) requests.
- Climate data could serve as an entry point for releasing more public sector data and to make municipalities climate neutral and more resource-efficient. The Internet Foundation (Internetstiftelsen) is correctly supporting local leaders in this. As a first step these leaders are asked to sign “climate contracts” with climate goals that they aim at achieving. Then it becomes easier to propose the implementation of these contracts with analytical tools driven by (open) data.
Create an open data ombudsman
There are already a number of intermediators or “cultivators” that develop open data and enable progress of open data in Sweden, such as civic tech associations and groups. The interviewed Crusoe stresses that it is important to ensure the stability of the open data field given that it is “the ability of the ecosystem to avoid permanent changes, shocks, and disruptions that can harm the ecosystem and its elements and actors”. To promote sustainability and stability he proposes to introduce an “open data ombudsman” or a “data governor” in Sweden.
Stimulate collaboration across sectors and between actors
There is a need for forums where data holders and reusers can meet, exchange knowledge and where the future of open data can be discussed. Some initiatives are already planned. In June 2021, the Government gave an assignment to the Agency for Digital Government (DIGG) to increase the amount of open data and the ability to use it. One of the key aims of this assignment is to boost collaboration between actors who share and use data, and the development of a digital arena that enables actors to data share and reuse data. This would support the exchange of experience and institutional learning about data sharing, as well as stimulating experimentation and development activity with data. Special focus will be placed on the use and value of data in authorities’ day-to-day operations. Although the results of this assignment will probably be useful, more structural support and funding will surely be needed for Sweden to rise to the top-ranks in open data.
Data workshop
A knowledge and coordination hub is needed as a sort of a “one stop shop” for data in Sweden where national and local authorities can turn for support and advise, and that can work as a sort of institutional memory of all the past and current data initiatives. To address this need, a number of authorities that are already far ahead in terms of open data together with Internstiftelsen, are planning to implement “Dataverkstan” (Data workshop). They will use the model of Västragötalandsregionen (one of the partners), which has been successful in getting on board their municipalities and opening up data.
Develop a taxonomy to classify and showcase open data uses and solutions
There is a need to develop a taxonomy to classify open data uses and solutions in Sweden, in order to understand where open data can create value and what values should have priority. There are already some attempts in this direction that can be built upon, e.g. Gebka and Castiaux’s (2021) typology of municipality and user roles in open data release and reuse.
It is also important to showcase stories of national and local authorities that have found ways to create value from open data. Areas that are far ahead in related innovations – e.g. the city of Jönköping that has a neighbourhood using IoT technology – can also serve as a test beds for open dat solutions.
Think in terms of “data for the people” and involve civil society
The government should invest in data for the people. Open data should make the difference for ordinary citizens, not just replicate existing data or app solutions. Citizens’ needs should be in focus. This is also where the civil society can support the release and reuse of open data, e.g. by identifying groups of users of certain data so as to show the authorities that the data can create value of the general public if it becomes open.
The interviewees were not very optimistic about the rise of the open data issues high on the political agenda in the next parliamentary election in 2022. Open data is not considered to be a politically charged question. The general perception is that democracy is working well in Sweden, and that the level of corruption is low. Data for transparency is therefore not a real driving force.
Moreover, Swedish politicians are generally not very knowledgeable about open data, and parties have traditionally not been promoting open data in their electoral programmes. However, there might be an opening for open data questions as a necessary instrument for getting information and building analytical tools for other political issues such as climate change or environment.
Interviewees
Jonathan Crusoe, lecturer, University of Borås.
Daniel Dersen, responsible for open data, Internetstiftelsen.
Per Hagström, freedom of information expert.
Sumbat Daniel Sarkis, Head of Section, Ministry of Infrastructure.
References
Crusoe J. (2021). Open Government Data as a Reform and Ecosystem – A conceptual framework for evolution and health. Doctoral thesis. Linköping 2021.
Gebka, E., & Castiaux, A. (2021). A Typology of Municipalities’ Roles and Expected User’s Roles in Open Government Data Release and Reuse. In H. J. Scholl, J. R. Gil-Garcia, M. Janssen, E. Kalampokis, E. Kalampokis, I. Lindgren, & M. P. Rodríguez Bolívar (Eds.), Electronic Government: 20th IFIP WG 8.5 International Conference, EGOV 2021, Proceedings (pp. 137-152). (Lecture Notes in Computer Science; Vol. 12850 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-84789-0_10