Revisiting: Is There a Business Case for Open Data? – The Scholarly Kitchen

Looking back at this 2017 post brings a mixed bag of thoughts. First, the fortunes being made with collecting, curating, and selling access to consumer data still haven’t spilled across into research data, and that’s likely because a) relatively few research datasets are available, and b) for the most part, the ones that are available have inadequate metadata and incompatible structures, so that combining datasets for meta-analyses is scarcely worthwhile. Until we address the problem of missing research data – which (full disclosure) we’re trying to do with DataSeer – we can’t really make much headway with getting it all into a consistent format. However, while combining datasets for re-use is a core feature for consumer data, it’s only one of the reasons for sharing research data. Open data also allows readers to check the results for the paper itself, and perhaps this is where our attention for the ‘business model for open data’ should turn. In particular, peer review is considerably simpler when the authors submit computationally reproducible manuscripts. Editors and reviewers can then be sure that the datasets support the analyses and hence the results, allowing them to focus solely on the appropriateness of the experimental design and the significance of the conclusions. It’s therefore conceivable that journals could reduce the APC for computationally reproducible articles (or hike it for non-reproducible ones), thereby incentivizing the extra effort required to required to produce them. No matter what route we choose, it’s clear that our current incentive structures around open science (mostly strongly worded policies and the lure of extra citations) are not getting the job done, and we need to consider alternatives. Money can enter the equation at a few places: by only funding open science, as exemplified by Aligning Science Across Parkinson’s, or by offsetting the extra effort required by researchers with additional financial resources, by making things cheaper or non-open science more expensive. Let’s see where we go.

Leave a Reply