Speaker
Description
To be truly useful, Open Data must be more than a file deposited in a public repository. The recently published FAIR Principles for scholarly data - Findability, Accessibility, Interoperabiliy, and Reusability - provide guideposts toward an emergent scholarly publication landscape where data, and its provenance, are transparently available for evaluation and reuse by both humans and machines. However, the FAIR Principles (purposely) do not advocate a technical implementation, preferring to simply describe the behaviors and qualities expected from contemporary data publications that allow them to fully participate in the Internet of Open Data and Services. As such, several key stakeholders involved in the establishment of the FAIR Principles felt it would be useful to create an exemplar implementation, to answer the question "what does FAIR look like, in practice?" In solving this problem, the stakeholder group committed itself to the use of existing Web technologies and standards, and in particular, refused to consider creating any new API. The outcome was a novel combination of emergent Web technologies and standards that can be applied, incrementally and in myriad combinations, to gradually increase the FAIRness of both legacy and prospective data publications. In this presentation, I will provide a walk-through of the design and application of this novel approach. We will explore how even simple enhancements to a data publication, applicable to any repository, dramatically increases its FAIRness. We will then examine the upper end of FAIRness, where data can be discovered and dynamically "projected" into a variety of formats, according to the user's needs, without any human intervention.
