This repository analyzes the licensing of bioRxiv preprints. The bioRxiv data was generated Omnes Res for PrePubMed — a search engine for biomedical preprints.
The findings from this analysis are summarized in a blog post titled The licensing of bioRxiv preprints, which analyzes preprints through but not past November 2016.
Click the badge above to launch this repository in binder, which allows you to
interact with the jupyter notebooks. Note that the mybinder.org build may be
outdated. You can rebuild it
here. Binder uses the
environment.yml in this repository to create a custom Docker image
with the environment for this analysis. If binder fails to load, check the system status and gitter before opening an issue here.
This repository uses conda to manage its environment as specified in environment.yml.
Install the environment with:
conda env create --file=environment.ymlThen use source activate biorxiv-licenses and source deactivate to activate or deactivate the environment.
The analysis is performed by running the following notebooks:
1.download.ipynbretrieves bioRxiv data from PrePubMed.2.create-figure-data.ipynbcreates JSON data files for vega-lite.3.impact.ipynbinvestigates the impact of the blog post on author license choices.
execute.sh automates running the analysis for command line
usage.
This repository is dual licensed as BSD 3-Clause and CC0 1.0, meaning any repository content can be used under either license. This licensing arrangement ensures source code is available under an OSI-approved License, while non-code content — such as figures, data, and documentation — is maximally reusable under a public domain dedication.