Science transcends form: publishing beyond papers

Comparing existing tools for openly sharing scientific documents that don’t fit the mold of scholarly papers

For the past 200 years, the primary tool for sharing science has been the journal article: two columns of text, static figures, and predictable structure from background to methods to conclusions.

But anyone who has worked in science knows that the practice of science doesn’t adhere to this structure. Rigid formatting and content requirements slow down the process of getting science out of the lab, and worse, stop many valuable resources from being shared at all. Ultimately, restricting the types and forms in which science is shared inhibits its progress and its societal impact.

Other publications are possible. Researchers in academia, industry, and beyond can and do produce documents that don’t fit the mold of traditional papers or preprints (versions of papers shared by their authors to a public server, usually before completion of peer review). These include grant applications, results from projects that are discontinued, null results, and individual results that don’t conform to a traditional paper format. Simply depositing code in GitHub or data in generalist repositories may leave behind context that could maximize reuse. Adding written descriptions of the researchers’ motivations, methods, and interpretations as a wrapper around (appropriately deposited) code and data can make these outputs more useful for humans and machines alike.

However, there are few publishing options that can host these diverse documents, and selecting among them requires making some tradeoffs. Where should researchers post their content that doesn’t fit the mold of a traditional article?

We set out to look at the options.

Seeking flexibility and durability in non-traditional publishing

Many experts have articulated best practices for repositories via the TRUST, POSI, and FAIR principles. Notably, these principles express a need for long-term thinking and the responsibility to archive and maintain resources indefinitely.

Astera’s open science team (Prachee Avasthi and Jessica Polka) have been working to evolve scientific publishing in the life sciences for nearly a decade, especially through advocacy for preprints (both via ASAPbio), changes to the journal peer review process (Jessica via Review Commons and Prachee via eLife), and pioneering non-journal publishing (Prachee via Arcadia Science).

Through these experiences, we’ve identified several important features of platforms and tools that can host non-standard scientific content. Each platform we looked at can do some of these things, but not necessarily all:

• Hosts non-paper content, going beyond the standard article format to include short reports, grant applications and reports, hypotheses, and more.

• Customizable display, including branding, interactive content/data, ability to change the layout e.g. hiding a byline.

• Issues DOIs, which are well-established identifiers for scholarly content, aiding discoverability and offering machine-readable metadata.

• Offers permanent archiving, because content that’s valuable should persist.

• Enables public commenting, which allows the community to add valuable context to the work, enabling readers to benefit from expert viewpoints.

• Is discoverable on Google and other popular search tools important to make the content visible, to experts and non-experts alike.

• Indexed, e.g. by Google Scholar, an important discovery tool for researchers.

• Anyone can post without self-hosting an instance, and they don’t need to be a part of an organization that does so or go through a peer review process (though submissions may be screened). For this reason, we’ve focused on existing centralized, hosted repositories that allow anyone to post. We acknowledge that there are software platforms that in principle could be used to host a suitable repository, but that depends on the choices of the future host organizations.

There are more criteria that may be important for different kinds of outputs that we didn’t specifically look at, like interactive code and data elements, and there are more and more tools continuing to come online. But for researchers who want to get started today to enable flexibility in format and durability of record, we compared some existing tools against these criteria. We acknowledge that this list is incomplete and that other options are possible, but these are the ones that come up most often in our conversations. The detailed table comparing features of various platforms can be found below.

Our recommendation: mix and match to get everything you need

For most researchers without other resources, we recommend using a combination of Zenodo and Substack to get a blend of complementary features.

Zenodo is a generalist repository that can host scholarly objects that are easy to cite and play nicely with scholarly communications infrastructure (including reference managers, ORCiD, and more). Zenodo is supported by CERN, which is likely to persist in the long term. Choosing an open license such as CC BY will help to promote reuse.

However, Zenodo doesn’t enable the flexibility, discoverability, or community of less specialized non-scientific publishing platforms more commonly used for blogs and newsletters. To enable commenting and have more formatting options, you can also cross-post to a discoverable website or blogging platform such as Medium or Substack (for example, the Astera-funded Pioneer Labs posts frequent scientific updates to its Substack newsletter). You can include a link to this more readable version in the description of the Zenodo record. At the top or the bottom of the blog post, you can help researchers cite your work properly by copying the “cite as” text provided by Zenodo (see below for example).

The below table compares the other tools that we considered in reaching this perspective. We compared both centralized, ready-to-use hosted repositories that allow anyone to post, such as the generalist repositories like Figshare or Zenodo, blogging and newsletter platforms like Medium and Substack, or preprint servers like bioRxiv or arXiv, as well as open source tools that could be used to set up a self-hosted repository, such as OJS, Janeway, and PubPub.

For organizations that have the capacity to host a repository, the open source platforms listed in the table are attractive options. Astera has supported Knowledge Futures as a grantee to enable the development of its fully-featured and customizable platform, with the goal of making it universally available for all researchers.

You can see an early example of PubPub’s potential implemented at Arcadia Science. Its publishing site is a customized installation of PubPub, which supports customized features like replacing the byline with a modified CRediT taxonomy, displaying alphabetically the contributors along with their contributions.

Today’s options don’t offer one simple solution for researchers, but by combining different platforms this way researchers can access everything they need to make available scientific texts that go beyond papers. We hope to see this change in the near future.

What are you doing to share your science? We’re curious to hear how others are solving this problem!

Cite as: Polka, J. (2025). Science transcends form: publishing beyond papers. Astera Institute. https://doi.org/10.5281/zenodo.14927522