It’s no secret that there are significant structural problems in the scientific publishing ecosystem, both in publication quality and confidence in the process of publication. There have been many analyses provided by the scientific community and the publishers themselves regarding this crisis.1 2 3 4

This post will briefly analyse the current issues and potential solutions before presenting a technical solution designed to improve scientific publishing. We are actively developing an end-to-end, open-source scientific publication and peer-review system, called Reflx. If you’re interested in contributing, then please get in touch!

Current Issues

There is a reasonable expectation that scientific publication should be accurate, accessible and affordable. However, a number of pressures have led to the publication ecosystem undermining these aims.

Some of these pressures have arisen incidentally from the evolution of online publishing, while others are deliberate barriers resulting from the financial models of publishers. The result is that scientists need to bend their paper-writing to the whims of publishers that often impose significant burdens on authors. There is considerable pressure to publish and it is not always clear that papers will land at the “right” journal, be received fairly or proceed though the peer-review process in a timely manner. Furthermore, re-submission to more than one journal is common, time-consuming and involves significant editing overhead.

The other side of the coin, is that scientists invited to peer-review can find the process burdensome and may be providing their expertise to companies that exploit their market control for profit. Scientist understand the importance of peer-review but it is not surprising that they may be reluctant to provide their time for free in a system that actively exploits their labour.

The result is that scientific productivity is lost and discourse is curtailed. Even worse, the pressure to publish, reduced effectiveness of peer review and commercial pressures create an environment exposed to fraud, scientific errors, and junk publications. For the public and institutions, the outcomes are dire: reduced trust, lost productivity and wasted money.

Issue, Affecting: scientists public institutions government
reviewer fatigue reduced quality and engagement opacity of process poor trust lost opportunities
fraud loss of trust loss of trust reputational risk financial mismanagement
cost of publication barrier to innovation inaccessibility financial impact financial impact
delays and reworking lost productivity delayed access lost productivity lost productivity
junk journals/publications impaired knowledge discovery confusion regarding quality reputational risk wasted funds
reproducibility and bias reduced confidence loss of trust strategic uncertainty lost opportunities

Online publishing provides the opportunity for great efficiency, reproducibility and low costs, or at least it should. The marginal cost of publishing papers is low with a well-designed publication system, meaning that the high costs of publication are unjustified.

How can these issues be addressed?

Innovations and Failures

There have been a number of responses to these issues. However, I will contend that recent innovations have failed to solve the core issues, largely due to lack of integration.

An encouraging area has been the rise of free and open-access pre-prints (most notably the arXiv and bioRxiv distribution services). These services provides open submission processes; papers need to meet some minimum guidelines but are only lightly moderated and are usually published within 24 hours. The publication format uses the (La)TeX typesetting language, which is broadly used in the mathematics and physical sciences communities.

Pre-print systems have made it easier for papers to reach audiences for discussion and the normalisation of this method of publishing has been an important step in improving scientific discourse. Importantly, traditional journals have relaxed their policies regarding prior publication and distribution. This is important so that authors are not penalised for making their work open to the public.

However, there are some drawbacks to pre-print systems. They can be difficult to use for authors not familiar with (La)Tex, do not provide an inbuilt review process and there is no guarantee of the quality of the publications.

There have been some attempts to address these limitations. For instance, the alphaXiv system layers a commenting system on top of arXiv. However, these type of layered systems have not received wide adoption and are poorly integrated.

There are also commercial and open-source systems to make publishing in (La)Tex less technically demanding (for instance, ShareLatex, it’s commercial fork Overleaf, and the desktop application LyX). However, it is not reasonable to expect all scientific authors to learn (La)Tex, especially when it is somewhat arcane.

A path forward…

In summary, we are presented with three main issues that have prevented better access in scientific publishing:

  1. misaligned incentives (both in peer-review and publication)
  2. technical complexity (of online authoring systems)
  3. lack of integration (particularly of commenting)

Our proposal is to address these issues by utilising the best current technology to provide a simple to use end-to-end publication system with open peer-review built in.

Thankfully this is not a project which needs to be built from-scratch, as excellent open source software is already available which provides many of the prerequisite features.

From a design perspective, the system (which is called Reflx), will include the following key features:

  1. an authoring environment, focusing on real-time collaboration, reproducible research principles and accessible tools
  2. a rich commenting and proposed-changes environment to assist collaboration
  3. an open peer-review system with incentives for accurate and insightful review built-in
  4. a publication system which surfaces and collates the best papers into “digests”

The intent is for system to support authors in the full process of scientific publication, from concept to collaborative writing to review and, finally, publication; all within the one environment. For reviewers and auditors, the process is transparent with clear authorship and document history. For the public, the transparency of the process and the visibility of the peer-review process will encourage trust in the scientific process.

Bringing the whole publication flow into a unified system is an ambitious aim, but we believe that it is necessary to do so to achieve integration and transparency in the process.

Reflx - a specification and work in progress

for further details see the current Reflx whitepaper (in draft) is available online

Following a review of existing technologies, we have devised an implementation plan for the platform. The main areas of work will include:

  1. Formalise specification (WIP)
    • define components
    • define interactions
    • build data object and view specifications
    • user journey and interaction specifications
  2. Extend the Hedgedoc Markdown Editor as the basis for the authoring front-end
    • improve alignment with Quarto Markdown syntax and features
    • build commenting and track-changes system into Hedgedoc
    • integrate git “checkpoints” into Hedgedoc Revision System
    • automate publication quality rendering of content (example available here)
  3. Build peer-review system (re-utilise code from commenting system)
    • classifier and documents/paragraphs
    • peer-review invite system
    • peer-reviewer reputation system (similar in concept to StackExchange reputation system)
  4. Publication pipelining
    • ensure smooth transition between document stages
    • adjust incentive structure
    • define digest editor promotion and dissemination of content
  5. Promotion and beta program
    • build developer community
    • build interest in scientific community
    • beta program invitations and feedback

Some notes on architectural choices (FAQ)

Why are you integrating these functions together, wouldn’t they be better in separate systems?

We considered building separate publication and peer-review systems. However, there are significant benefits to integration. Having a competent authoring system which also allows integrating changes following peer-review has the advantage of supporting the authors in making changes without format-switching. Integration also ensures that the full research paper creation process remains transparent (a key reproducible research principle). There is also a lot of overlap of features which make integration a natural architectural choice. Nevertheless, the code developed for the project will generally be maintained in a modular framework, allowing sub-systems to be built if desired.


What license will the project use?

The project will be published under the GNU Affero General Public License v3.0 (AGPLv3). this license has the benefit of ensuring community benefit of code contributions even if the project is forked to other online systems. It’s also the underlying license of Hedgedoc.


Why have you chosen Markdown as the format?

There are many good reasons for why Markdown is an excellent format for scientific publication. For example, see this post. In summary, Markdown hits a sweet spot between usability and precision. Authors seeing Markdown for the first time can read it clearly and edit it safely (unlike LateX). It also allows precise document control while separating format from content. The Quarto extensions to R Markdown are specifically designed for scientific publishing and allow excellent document control while hiding technical complexity.


How will you deal with templates?

Different digests in the platform may choose to render documents with specific styles. However, we believe that styling of documents is an unnecessary distraction to authors. Therefore, we have built a carefully designed template for HTML and PDF output (based on Quarto) which includes all the necessary features for scientific publication. It provides excellent support for document structure, citations and cross-references, images, mathematical notation, tables, and inbuilt search functionality. This format will be refined over time, but is designed to be easy to use and beautiful. For an example, see this test publication (WIP). We understand that this choice will be a bit controversial, but believe that simplifying the authoring process is essential to ease-of-use.


Will peer reviewers be identified or anonymous?

Registration on the platform (for authors and peer-reviewers) will be pseudo-anonymous. This is important to protect peer reviewers and encourage academic freedom. Each user will have a username (handle). If a user chooses to identify themselves that will be permitted.


What is the timeline of the Reflx project?

We plan to have the beta program available in Q2 of 2027.


How can I help?

Please contact me if you’d like to help. We will soon be publishing a pubic repository (using Gitlab) where you can also make direct contributions - details to follow. In the meantime, please feel free to share this post and discuss!


References and Footnotes

  1. Adam, D. (2025). The peer-review crisis: How to fix an overloaded system. Nature, 644(8075), 24–27. https://doi.org/10.1038/d41586-025-02457-2 ↩︎

  2. Vineis, P. (2024). Scientific publishing: Crisis, challenges, and new opportunities. Frontiers in Public Health, 12. https://doi.org/10.3389/fpubh.2024.1417019 ↩︎

  3. Altbach, P. G., & Wit, H. de. (2024). The Dysfunctional Academic Publishing Ecosystem: The Need for Reform. International Higher Education. https://ihe.bc.edu/pub/ro4sc1tg/release/1 ↩︎

  4. High Costs, Long Waits, and Ethical Dilemmas: A Review of Challenges in Academic Publishing. (2026). Journal of Scientometric Research, 14(2), 413–423. https://doi.org/10.5530/jscires.20250015 ↩︎