Agile production of high quality open data

This paper introduces an Agile Methodology to support the production of Open Data (OD) which can be internally adopted and adapted by Public Agencies (PAs) in order to make data publicly available. The methodology is iterative, incremental, evolutionary, test-driven, and collaborative. The novel idea is that PAs produce open data and, at same time, envision and anticipate possible uses and re-uses by citizens. The opportunity is that PAs will publish use cases along with Open Data as a way to engage citizens. The fit-for-use test is a specific step of the methodology in which PAs conceive possible uses of the dataset by creating a set of relevant visualisations (e.g., charts). This step mitigates some well-known barriers in the field of the OD, such as, that the data could be not accurate, not interesting, or too costly to be re-used [27]. The paper describes a platform named SPOD, based on the above agile methodology, meant to be used by PAs over their Intranet to support the collaborative process to make data publicly available. Moreover, the methodology and the SPOD have specific steps to assist PAs in checking the dataset quality through syntactic, sanity and domain-specific steps implemented through a set of heuristics. Both the methodology and the SPOD have been successfully adopted by the Council of the Campania Region in Italy to produce their open data, this experience has detailed described and reported in the paper.