Fair digital object

expanding the boundaries of data stewardship

Abstract

The exercise of scientific research, especially in the long tail which is populated by small laboratories working on a multitude of projects, generates and consumes a vast amount of digital objects that are heterogeneous in several dimensions: in terms of volume, formats, models, technologies, structural complexity, configuring genuine scientific big data The automation becomes decisive for those digital objects to be appropriately managed according to their peculiarities. The need to identify the types of different digital objects, their disciplinary context, as well as the qualification of the relationship between them, also becomes of great importance. The problem that arises in this context is that current data management practices have proven to be inefficient, relegating a large part of digital assets to a condition of invisibility and without conditions for access and reuse by human and computer stakeholders. Looking for a solution to this dilemma is proposed an abstract unit of knowledge, referenced by persistent identifiers, and described by metadata capable of interacting with automated systems, named FAIR Digital Object (FDO). In this sense, the methodology adopted by the present study has qualitative approach  and an exploratory nature, presenting a conceptual, epistemological, historical, and technical study, whose objective is to analyze the concept of FAIR Digital Object (FDO), examining its possible contributions to the technological and informational convergences necessary to overcome the current inefficiency of data management. As a result, a new conception of digital object management is presented, which considers FAIR principles already in its architecture and which, through machine actionability, can lead us to a new generation of the Internet: A FAIR Internet of Data and Services. It is concluded that the concept of FDO significantly changes the vision of what a dataset should be, making it necessary to decode and disseminate knowledge that enables these data to become integrated sources of services, within the scope of an automated information ecosystem that can meet the needs of the most diverse sectors.