Almost two years ago, on 1 January 2021, Nature Astronomy became a transformative journal. It means that we are committed to publishing a growing number of open access (OA) articles until we’re a fully OA journal, in compliance with Plan S — an initiative to make academic work fully and immediately accessible upon publication. Until then, we will continue to offer authors a choice between OA and the traditional subscription model, so that there are no financial barriers to publishing primary research in Nature Astronomy, Nature or any of our sister titles.

So far, less than 10% of our authors have opted for the OA route. There is some confusion regarding which funders require OA, who pays and how much. Many institutions now combine Nature journal subscription access with OA publication costs in ‘transformative agreements’. These agreements provide clarity to authors to help them comply with their funders’ requirements, simplify administration and cover costs. An evolving list of participating institutions can be found here:

Whether OA or not, we encourage all authors to post their preprints at any stage of submission (including before) on arXiv. Regarding access to the final version of the paper, OA papers are immediately accessible for free and have liberal licences for re-usability. For non-OA papers, authors can self-archive the accepted version (after peer-review but before typesetting) of the paper six months after publication, in what is known as ‘green OA’. On publication of all our papers, we also provide a free, shareable PDF version to the authors through our SharedIt initiative.

In August 2022, the White House Office of Science and Technology Policy published a memorandum to federal agencies “to make publications and their supporting data resulting from federally funded research publicly accessible without an embargo on their free and public release” by 31 December 2025 at the latest. While the reasoning behind the policy update is just and sound, the memorandum makes no distinction between green and gold OA models, and no specific references to how much funding would be available for OA publications. Such details will be decided by the individual agencies in due course, subject to funding.

Making supporting data available on publication is an obvious win for open science, for the paper alone may not contain enough details for complete scientific scrutiny and reproduction. However, a general framework can be short on practical implementation instructions. Many researchers simply do not know how to make their data shareable with the public, and in perpetuity. For this reason, Nature Astronomy is joining the free pilot between Springer Nature and Figshare to help authors make their data freely available.

Figshare is a data repository for any file format (examples pictured), with files being citable, shareable and discoverable. Authors can now upload their spreadsheets, tables and images at the point of submission by using the ‘Research Data Deposition’ tab. The process is free and automated for datasets up to 50 GB (larger files are possible, but have a separate workflow and may incur a fee). The data will remain the property of the authors and will be available to the editors and reviewers during peer review. Published papers will then include links to any Figshare files, in addition to any Supplementary Information associated with the article.

And what about data availability before publication of the results? There is a lively debate over whether publicly funded (NASA) telescopes should have a proprietary period at all. The first five months of JWST observations are not under embargo, partly because observers in the queue need to see the full capabilities of the various instruments ahead of their allocated time. But after that, researchers will have one year to analyse the results before they are publicly released. For other telescopes, the proprietary time varies from six to eighteen months. Before making a decision on JWST and other telescopes, the Space Telescope Science Institute plans to survey a wider community of 12,000 astronomers. Open science is very much an open issue.