Aim: One of the primary outputs of scientific research is the paper. Many papers are structured in a similar way, covering abstracts, methods, results analysis, discussion, conclusion and supplementary materials. The bulk of this work – setting up the structure of a paper and embedding data – is repeated time and time again by researchers, despite being a relatively routine and theoretically automatable problem. We would like to take the pain out of this process, making it easy to automatically generate the parts of a paper that can be easily scaffolded and incentivising reproducible papers by ensuring the scaffolds include well-structured data and metadata.
Work at the Sprint: Our data paper skeleton tools for biology/bioinformatics data are currently in a prototype state. We hope to flesh out the prototypes by user testing with hands-on use cases at the Sprint, and exploring integration with the eLife Reproducible Document Stack. A stretch goal would be to automatically include figures (e.g. scatterplots, bar charts) or enrichment statistics that draw from the frictionless data packages associated with the auto-generated paper skeletons.
We are looking for… people with domain expertise in JATS, the Reproducible Document Stack, and other research publishing tooling who would be able to offer valuable insight into our tool design. Input from designers and UX consultants would also be welcome. It would also be helpful to have scientists/researchers and editors, or other members of the manuscript publishing field, as test users to give us feedback, to help with documentation and bug reports.