The move by government organizations to public cloud infrastructure is often motivated by a desire to provide better services to citizens, but an academic piece warns about the need to maintain public access to open data in the cloud.
Written by Dr. Mariel Borowitz, a professor at the Sam Nunn School of International Affairs at the Georgia Institute of Technology, the piece notes that commercial cloud platforms often charge a fee to download data, or use cloud-based tools to analyze data. While moving to the cloud can enable more storage and computing power for agencies, it also can make previously available data harder to access.
“Such challenges to the implementation of open data policies must be addressed to ensure that the current benefits of these policies are not lost and to realize the opportunities for researchers and society presented by big data and cloud computing,” Dr. Borowitz wrote.
While open data works on the assumption that there is no limit on the quantity of people who can use it, and that providing access does not add additional cost per user, the move to commercial cloud can complicate that equation. Downloading data comes with a cost, as does using analytical tools on cloud platforms.
Borowitz cited an example at the National Oceanic and Atmospheric Administration (NASA), where a cloud service provider agreed to host a massive 100 petabytes of data at no cost to the agencies, but charge end users to download the data. As a counter-example, Borowitz pointed to NASA, which is exploring cloud options for its environmental data, but plans to pay the costs of distribution and analysis.
With the passage of the OPEN Government Data Act and the recently released Action Plan for Open Government, the push for open data among Congress and the White House is strong. Agencies will need to figure out how to balance access to data with the costs of a commercial cloud platform. Borowitz details several suggestions to keep open data free for users, such as budgeting for new expenses, collecting metrics, and implementing reasonable limits.
“The transition to cloud environments for the distribution and analysis of government data is under way, but in most cases, long-term decisions have not been made. Now is the best opportunity to design systems that help us to understand and maximize the value of open government data,” she concludes.