The Importance of Data Set Provenance for Science

Abstract

Data do not exist in a vacuum. To be useful, data must be accompanied by context on how they are captured, processed, analyzed, and validated and other information that enables interpretation and use.

Labor Out of Place: On the Varieties and Valences of (In)visible Labor in Data-Intensive Science

“…Today, those partial truths are spread among instruments, papers, data sets, and software. In dataintensive science, authoring is both writing papers and grants and authoring data sets and research software (Mayernik et al 2015;Green 2009;Hills et al 2015). Authorship is also a dividing line separating those whose name appears on the byline of scientific papers, and hence registered as a citation, from those merely acknowledged.…”

Certifying Provenance of Scientific Datasets with Self-sovereign Identity and Verifiable Credentials

“…However, due to a lack of proven methods to assess the quality of data and inadequate archival processes, secondary users can effectively utilize only a small subset of the collected data. For the data to be useful to its full extent it should be verifiable for its origin and trustworthiness [10], especially when it is applied in the field of scientific research.…”

The Geosciences DeVL Experiment: new information generated from old magnetotelluric data of The University of Adelaide on the NCI High Performance Computing Platform

“…However, to make MT datasets more interoperable and reusable, particularly on HPC, requires greater international agreement on data formats and consensus on what metadata attributes are to be collected both at the time of acquisition and during processing. To be really useful, data must be accompanied with information about how they are captured, processed, analysed and validated and other information that enables interpretation and use (Hills et al, 2015). The Australian proposal for metadata attributes (Kirkby et al, 2019) needs to be socialised with the international community.…”