blog

Here’s the concluding part of the interview with Peter Murray-Rust at the University of Cambridge.

We talked a bit about publishing data, and indeed science as a whole: where open data started for Peter was when he suggested all science should be published as supplemental data. Taking such a suggestion forward, the problem then becomes one of where to publish that data. Data are necessary to back up assertions in published papers, and if as much data as possible should be made available, then it doesn’t have to be associated with the journal in which the article is published. With DOIs and appropriate links, it could be published anywhere…

We also touched on cases of fraud, and where we might be going with all of this.

We can assume [fraud] happens in other disciplines, it hasn’t been detected.

Using protein crystallography as an example, Peter suggests that re-calculation of results (structures), or actually checking assertions, will become part of the review/validation/publication process.

Also see Parts 1 and 2 in this series.