Interview: Wendy Carrara and Dinand Tinholt of the EU Data Portal project talk about the lessons from two new studies on open data
This week’s launch of the European Data Portal has provided further momentum for the cause of open data in the EU, and been accompanied by two reports that provide fresh evidence that the UK is one of the star performers in the field.
Capgemini Consulting, leader of the consortium that built the portal, compiled the studies – Creating Value through Open Data and Open Data Maturity in Europe 2015 – on behalf of the European Commission in an effort to clarify where the work on open data now needs to be focused.
The lessons suggest that, while there is steady progress throughout the EU, there is still a need for a more structured approach, even in the UK.
While the reports are not primarily for benchmarking, they include indicators of member states’ achievements so far with open data. The UK is second only to Spain on the maturity indicator, taking into account its readiness with policy, licensing norms, national coordination and use of data, and the maturity of its prime portal, data.gov.uk.
It also attracts the highest monthly number of visitors to its national portal at 175,400, more than double the figure for the second highest, France.
Speaking to UKAuthority, Wendy Carrara (pictured), principal consultant at Capgemini and project manager for the European Data Portal, says: “The UK is one of the top performers. It has been pushing open data for some time and ranks highly in the assessment, with a dedicated portal and a clear vision in terms of re-use.”
But she says it did not come out as number one because there has been no clear assessment of the macroeconomic impact of open data, such as in the number of jobs created.
“In terms of demonstrating that political and social impact, more could be done to promote the re-use of open data. But this is to be taken in inverted commas because the UK is one of the top performers and really paving the way for other countries.”
Dinand Tinholt,vice president and EU lead of Capgemini Consulting, adds: “There is a lot of value in the re-use of open data, but what we see through conversations with the public sector is that there is often not a structural, ongoing collaboration between public and private sectors about what they want.”
He suggests there is little sign of feedback from re-users being picked up by government, especially in prioritising data from within the vast volumes that can be released. This could be limiting the economic and societal benefits in the short to medium term.
Carrara also emphasises, making clear that it applies to all EU members, that the data has to be machine readable and accompanied by the appropriate licensing to ensure it can provide its full value for re-use. Work is being done on this in the UK, along with using open formats in the datasets more consistently.
“We’re looking at the nitty gritty of how to go from a top performer to a sustainable top performer,” she says.
On an EU-wide basis, the Creating Value report says that a third of countries are now ‘trend setters’ while only 19% are ‘beginners’, with maturity assessed largely on a combination of readiness for open data and usability. The direct market size of open data is expected to increase from €55.3 billion in 2015 to between €75.7 billion in 2020, with the indirect market being almost four times as large, and the number of jobs in the sector is forecast to rise from 75,000 to 100,000.
The report also makes a number of recommendations for government, including the creation of a marginal or free cost model for open data. There have been debates around charging fees for data, but Carrara says it has effectively been paid for already by taxpayers, and that the more open and free of charge it is the more likely it will be re-used.
“When it’s out there for free it generates more re-use than if only part of it is for free,” she says. “In line with the recent European Directive we urge countries to drop the charging model as far as possible.”
The issue of feedback comes up again with the recommendation that open data portals should include mechanisms for users to leave comments, although there is a question mark over how it should be done.
“It’s a culturally sensitive question as in some countries they might naturally provide feedback in a more structured fashion, some might require some guidance and some might not appreciate being guided,” she says. “Whether it’s more structured or totally open feedback is a decision that might come from more traditional reasons.
“What we believe is that the feedback should not necessarily be publicly displayed, as administrations might not want to be bashed on the portal. But if it is around issues like broken links or misleading information or inconsistencies, it could improve the quality of the data.
“Remember that administrations collect the data in the interest of the users.”
The other recommendations are:
- There should be more detail on the costs and benefits of releasing data.
- Government portals should maintain site analytics to obtain information on who uses the websites, which data sets are being downloaded and how many downloads take place.
- Governments should conduct surveys into the re-use of open data in the private sector.
- The workforce should be empowered to make the most of open data.
Carrara and Tinholt’s other lesson is that there should be some flexibility around the submission and capture of open data. There is a case for a central authority in each country to be in charge of its collation, and this would help in providing the consistency in its presentation, and ensuring it has the right licensing and is machine readable.
But there could also be cases when a city or agency is ready to release the data well before a central authority is ready, and they should not be constrained if there is a danger of it being left on the shelf.
In addition, Tinholt makes the point that in some countries agencies that could provide data are often independent public organisations, not under central organisations, and it would not be possible to dictate to them.
It leads to what could be the key element in the future of open data – creating a culture of “open by default”.
“Not opening things up would be an exception,” he says, “rather than the present situation of asking why we should open them up.”