Assessing the socio-economic impacts and value of ‘open’ geospatial information
Workshop held at George Washington University, 28-29 October 2014
I headed off to the ‘capital of the free world’ in late October. The sun was shining on this surprisingly green and pleasant city, recognizable from films and TV. Driving past the Pentagon and the World Bank on the way from the airport was a reminder of the military and economic power in this town. I also saw the Watergate building, a reminder of a more disreputable moment in US history.
I had been invited to comment (for 15 minutes, longer than one sometimes gets for a full presentation at other events) on a position paper called ‘impact of increased access to data and new modes of consumption’. Three other position papers were discussed: about citizen science & participation in the production of data, measuring economic impact of data, and societal impacts of open data. The position papers are not yet publicly available, but the organisers plan to prepare them for publication in some form.
About 65 people attended. The vast majority were American, a very interesting mix of university researchers, researchers based in government agencies (such as NASA, EPA, US Geological Survey), private sector (Digital Globe & other consultants, large and small) and NGOs (such as Open Street Map, and Map my Rights). Given the focus on geospatial information, many were indeed concerned with such data but the discussion ranged more broadly, and many points came up that are relevant for all discussions about open access to research data, and about the socio-economic value of public-private collaboration (relevant for many discussions within universities).
There was a lot of attention to public sector information, as geospatial data are largely generated by government agencies (sometimes forgotten in US political discussions – apparently a Republican member of Congress recently called to stop funding the collection of such data, now that we have the Weather Channel). Many present were searching for answers to the question of how to justify the expenditure on producing such data and data infrastructures in times of austerity and crisis. Some of the tensions in the meeting were between economists (many present) and others. Economists still believe that everything can be quantified in monetary terms, so that trade-offs between competing needs can be easily compared. For many of the economists present, the public good nature of data (non-rivalry and non-excludability) are part of what makes it very difficult to conduct standard economic impact analyses. Others questioned this logic, raising issues of social quality, and whether or not all data really are public goods (in the economists’ sense of the term).
A number of other themes to emerge are briefly summarized below:
- Data do not travel easily. There is much work involved in making data open (and discoverable, and (re)usable, etc). The costs & benefits of doing should be evaluated on a case by case basis. In other words, open as default might not be most sensible way forward.
- Open data doesn’t mean much without open code, open access publication, open standards, etc.
- Open data needs to be funded. Who is responsible for making public data user friendly (not something for which public sector agencies are usually funded?)
- At the moment, the extent of open data (research and public) is fragmented and uneven, across countries and across disciplines. This affects the potential impact of open data as a whole.
- What are the appropriate roles for public and private sector researchers, for public sector, for citizens, for private sector to add value to public data? Who is responsible for data quality, in the short and long term? Is citizen engagement always sustainable, and under what conditions?
- Privacy & confidentiality remain important considerations, especially as the granularity of geospatial data is getting very fine.
- Questions about how to manage data abundance were raised, drawing attention to the problems of integration across different types of data (over time, and across disciplines where different definitions of data apply). Some people also raised concerns about negative network externalities, in other words, more open data increases costs of finding data and reduces privacy.
The results of the RECODE project (see Announcements for details of final RECODE conference) about the fragmentation of data ecosystems, and the importance of paying attention not only to researchers but also to the nature of data in different fields were certainly confirmed by this meeting.
The highlight of the meeting (in a perverse, schadenfreude way) was being close to the failed launch of Antares, an unmanned rocket to deliver supplies to the International Space Shuttle. We should have been able to see it from the windows of our meeting room during the reception at the end of the first day. There were some very worried looking NASA people at the workshop who suddenly disappeared to try and find out what had happened. Better images would have been available on TV, but there was something special about being just a few miles from the launch site.