OSS: Open Scientific Standard

  1. 1.  Dept. Geosciences, University of Oslo

As the Open Science community grows and the zoo of Open Science platforms expand, it is becoming clear that the future of Open Science will not be in the dominance of one perfect system with killer tools, but rather numerous systems, much as it is today, focused on discipline or purpose specific applications. But there is a way to ensure a deep-level of compatibility between these burgeoning systems, and ensure that the Open Science future is one of interconnectivity: the Open Scientific Standard (OSS) format.

The OSS is a proposal of a standard format for Open Scientific Work. OSS can bridge the gap between past publications and also facilitate the creation of new types of articles, laying the foundation for the creation of the scientific tools of the future. By converting all open access publications to a standard format, we can build APIs to interact with the content, aggregate, and cross-reference as never before. (Think BibTex entries on steroids.) Irrespective of journal or article type, it would allow people to build systems which allow users to interact with content in new ways, and to focus on the particular aspects of scientific content that they wish, for example: highlighting publication trends; facilitating post-publication review; rapidly creating meta-studies to outline whole fields and interconnectivity of research (including highlighting the influence of individuals through a field); or, simply creating RSS-type feeds to stream user-defined content through unified platforms into final readable formats set by the users. These are just a few of the most obvious and immediate applications.

The OSS can readily be implemented in an agreed-upon JSON structure (objects made of name value pairs), with support for Unicode and LaTeX, ready for access by community developed APIs. No centralised organisation would be required to serve the data, as anyone could create REST APIs as required to serve the JSON data. However, it is probable (and desirable) that large organisations create centralised databases for convenient access (such as Google Scholar).

It is inevitable that there are a great number of researchers who, for a wide variety of reasons, will never change from a traditional publishing model. I.e. some researchers will never change from only submitting to closed peer-review in established pay-to-publish journals, with no post-review comments or article revisions, they will used closed code, in licensed languages, and provide data on approved request only. However, if the scientific community were to agree on the proposed OSS format, then it makes little difference if there are traditional researchers who are unaware or unwilling to actively participate in Open Science, as the creation of OSS compatible version of their work can be done in an automated way by traditional publishers and institutes as they add OSS support to their platforms: it is just a case of re-formatting Open access articles, so retrospectively adding work should be simple. Perhaps the only requirement to doing so is that is is Open and has a digital object identifier (DOI, - a unique code that identifies that piece of work). Thus, while the OSS can contain many attributes which accommodate/facilitate authors who work in new types of articles, it will also maintain backwards compatibility with past work. Perhaps compatibility with Pay-walled articles could be possible at a future time, although for now I acknowledge that this proposal effectively ignores them.

More than just being a fancy way to serve up traditional publications, this format can also open the door to a truly effective methods to implement new types of articles, including post-publication reviews, and (perhaps my favourite) Minimum Publishable Units (MPUs). MPUs could be readily grouped together under Publication Unit Themes (PUTs). Investigators tend to work for extended periods on a given research line, from the scale of individual researchers, to international collaborations. How much effort is wasted writing up results as self contained units within these research lines? How much more effective could these research lines be if they were grouped together, in essence wrapping encasing individual articles in an overarching theme containing meta-information (from rationale aims/scope, to comprehensive literature review, and methods/data). Imagine an outline of a research area and the problems to be addressed are published as a PUT, then successive minimum publication units linked to this and published when they are ready (small scale results and discussions pertaining to the PUT). Such MPUs would also perhaps be more likely to receive post-review discussion and comments due to their focused nature, and instantly group-centric attachment focusing interest within disciplines on currently active research themes.

The OSS format could be readily version controlled, so that authors can upload work at successive stages of development (pre-pub/self-pub, accepted, post-pub versions), with the tracked changes and reviews (if available) automatically highlighted. Interactive systems could be developed from this information to enable readers to roll through versions with ease.

Potential data fields in the OSS

  • DOI (digital object identifier, the method of identifying and cross-linking OSS work)
  • Type: Original Research, Comment, Synthesis, Review (review links to DOI of original), MPU (requires DOI of PUT)
  • Version: [integer value to track status, 0 can indicate self/pre-pub, with >1 indicating reviewed and successive modified versions]
  • Author(s)
  • Affiliation(s)
  • Contact info
  • Researcher ID value: (e.g. ORCID)
  • Title
  • Abstract
  • Disciplines
  • Keywords
  • Main content (references within articles are simply integers linked to Ref section)
  • Equations:(LaTeX format?)
  • Figures (ordered list of links or embedded items)
  • Legends [ordered list of text for figures]
  • Acknowledgements
  • Disclaimer/Conflict of Interests
  • References (lists of integer:DOI pairs)
  • Supplementary material
  • Supplementary Equations
  • Supplementary Figures (link)
  • Supplementary Legends
  • Supporting data link (external service or archive?)
  • Supporting code link (external link such as G-hub?)
  • Pre-publication reviews

License

This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.