Open data relates to the idea (perhaps ideal?) that some data should be freely available to everyone to use and republish as they wish, without restrictions. It is also a mechanism to allow data to be shared on a more restricted basis between institutions or businesses. Many businesses are using open data to their advantage, not just to support transparency but also to boost economic growth. One example is CarbonCulture, which is an open platform designed to help people and businesses use resources more efficiently.
A second example is with Arup, a U.K.-based global engineering consultancy. The new company uses open data as a key part of its work with smart cities and the technology that supports them. The benefits also extend to academia, according to Ian Powling, who develops digital programs to help universities. Powling has told the open data website Efficiency Exchange that sharing data openly offers considerable benefits for higher education and institutions must make the most of it. As an example, releasing data on procurement costs, energy costs and usage, or space utilization allows research institutions to benchmark one another and drive down costs.
There are two challenges for using open source data, as identified in a new study from the Open Research Data Task Force, which is based at University of Warwick in England. These challenges are where to find it and with compatibility issues once the data is found. Looking at accessibility first, even though open data is notionally accessible many researchers complain that they simply can locate files.
The problem with locating files is that the mechanisms available to identify and find research data considered to be “sub-optimal.” Often the search tools only work effectively for just a few subject areas. Most of the search tools available require considerable manual input, which can put some researchers off. The report suggests that further work can be done to help streamline workflows.
The second frustration is that once data is located there are often format variabilities and other compatibility issues. These issues are more likely to occur when researchers from different disciplines are attempting to review each other’s results. This means researchers often have to convert datasets, which requires effort and time.
It also raises risks and this can put the integrity of the data in doubt, as well as drawing out security vulnerabilities. The report highlights these as concerns with the “reusability of data.” Things are not helped where the quality of the data is not of sufficient quality. They security issue also raises a dilemma with open access: between the degree of openness with the access and the need for sufficient security to avoid data breaches.
The research into the use of open source data has been led by Professor Pam Thomas and she has focused on the U.K.’s national open research data infrastructure (although the findings can just as easily be applied to other countries).
The report from Professor Thomas also draws on some other limitations affecting the use of open source data. Included here are software issues. One limitation is with the users of data themselves, especially in relation to software skills — calling into question the abilities of many in the research community. Another limitation relates to how easily data can be stored and preserved.
