Open science can be variously defined. In some communities it is related principally to open access to scientific publications, for others it includes open access to research data and for others still it includes opening up the processes of academic research to engage all interested civil society stakeholders. The absence of a common understanding of what is, and isn't, included in open science creates confusion in discussions across these different communities. It is potentially holding back efforts to develop effective policies for promoting open science at the international level.
This paper builds on the limited conceptual work that has been published to date and proposes a broad framework for open science. The framework is not meant to be prescriptive but should help different communities and policy makers to decide on their own priorities within the open science space and to better visualise how these priorities link to different stage of the scientific process and to different actors. Such a framework can be useful also in considering how best to incentivise and measure various aspects of open science. Digitalisation is fundamentally changing science and the paper lays out some of the opportunities, risks and major policy challenges associated with these changes.