Research Software vs. Research Data I: Towards a Research Data definition in the Open Science context

Background: Research Software is a concept that has been only recently clarified. In this paper we address the need for a similar enlightenment concerning the Research Data concept. Methods: Our contribution begins by reviewing the Research Software definition, which includes the analysis of software as a legal concept, followed by the study of its production in the research environment and within the Open Science framework. Then we explore the challenges of a data definition and some of the Research Data definitions proposed in the literature. Results: We propose a Research Data concept featuring three characteristics: the data should be produced (collected, processed, analyzed, shared & disseminated) to answer a scientific question, by a scientific team, and has yield a result published or disseminated in some article or scientific contribution of any kind. Conclusions: The analysis of this definition and the context in which it is proposed provides some answers to the Borgman’s conundrum challenges, that is, which Research Data might be shared, by whom, with whom, under what conditions, why, and to what effects. They are completed with answers to the questions: how? and where?