Year: 2018
Type: conference-paper
Abstract: It is a cliché today to say that "data is the new oil", meaning that the commodity that will drive economic engines is access to and use of data.But if data are oil, what constitute the refineries that make raw data useful?Academic science has been grappling with the opportunities and challenges around routine, open sharing of primary research data for several decades.Many scientific fields have invested in the production of large, open data sets that are designed as a public resource to drive discovery using new data science approaches.But science produces thousands of smaller data sets every year, so called "long tailed data", much of which is not ever made available for reuse.Routine sharing of these data is still in early days in many scientific domains.Nevertheless, with increased scrutiny on scientific reproducibility and requirements from funding agencies for sharing of data, the issues surrounding data sharing are moving beyond whether data should be made available, to what are effective practices for making data useful beyond the initial purpose for which it was produced.Long tail data is highly heterogeneous, comprising many different data types, formats, sizes and data models.Because there are no standards for any given paradigm and new data types and techniques are constantly being developed, the challenges of how to make these data available and useful are considerable.Progress is being made in defining what constitute best practices for publishing of scientific data.Based on these efforts, Wilkinson et al. (Nature Scientific Data, 2016) encapsulated these best practices into a simple acronym, FAIR.Scientific data should be FAIR: Findable, Accessible, Interoperable and Reusable.These principles are realized through 15 attributes that when followed greatly enhance the utility of data.Although originating in science, the FAIR principles apply broadly to any organization or entity that is producing large amounts of data and faces challenges in managing it effectively for reuse.In this presentation, I will discuss the FAIR principles and provide examples of how they are being applied across scientific domains.
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4243818128