The high human cost of copying manuscripts during Antiquity made it possible for thousands of volumes - especially those that were not compatible with the religious doctrines of the time - to be burned or lost in the blink on an eye. There were not enough available hand-written copies of most books to protect them from the fires of disinterest or condemnation.
The scriptorium of the middle ages were the first institutions aimed at massively copying and diffusing written works in the West. It became the role of the clergy to maintain and expand the collection of available knowledge, in the institutions that forebode the academic editors of today. The medieval clergy was controlled by heavily centralized institutions. This constrained the knowledge that was available: we were living in a time where adhesion to dogma was a necessary condition for knowledge to survive.
Only when copying became decentralized - with Gutenberg famous' invention - could heresy, the heresy of Descartes, of Spinoza, or of the Marquis de Sade, become viable. During this classic period, academic books were published mostly by private editors. Books were then sold to a potentially large audience of educated scholars, since, as Thomas Kuhn pointed out in his seminal Structure of Scientific Revolutions, these academic book did not allow for a very high level of esoterism. After all, the authors rarely knew what other books their reader had read beforehand. And so, any educated reader of the 18th century should have been able to read Adam's Smith's Wealth of Nations just as any educated reader from the 21st century should be able to read Alice Goffman's On the Run.
During the XIXth century, natural science branched out into what has since become the official, modern ideal for all science to tend to : peer-reviewed journals. Subscribing to a journal meant you could be part of a special community, progressively building a common language, a common "paradigm". Academic journals appeared alongside modern academic disciplines and the academic division of labor that Auguste Comte was describing in his work. Academic communities became fragmented. If you publish in a journal, you must know what is usually published in this journal. But at the same time, it is acceptable to ignore other major contributions to the understanding of a topic, as long as this contribution is not in your "field".
Peer-reviewed journals have gone through several changes since their birth. Here is more or less how most of them work nowadays : researchers propose their papers to an editor, who may or may not send the paper for review depending on their editorial line. If the paper is sent for review, it is then critiqued by 2 or 3 anonymous researchers, chosen by the editor for their ability to understand the work - a choice that is sometimes arduous to make. After a series of back and forth between authors, reviewers and editors, the article can either be published months later - sometimes years - or rejected. If the paper is rejected, the researchers can either rework it, send it to another journal or give up completely.
While "double-blind" peer review has now become the sole process of production for the natural sciences, social sciences and humanities have gone a different way. Part of social sciences and humanities still hold the model of yore and rest on the publishing of books. Another growing part is imitating the natural sciences more or less blindly. Humanities and social sciences have often done this in the past. Immanuel Kant wanted to be the Newton of metaphysics and sociology could have been called "social physics". In that process, what has been missed is a deep reflection on the kind of publication model we would want in the humanities and social sciences. And while the production of knowledge has undoubtedly gained from such a diversity, it is now being built on a materiality that is already dead: the paper journal.
The material conditions of the diffusion of knowledge are changing, and we have to take advantage of this change to create the best possible environment for academic research.


Things to keep. Academic journals have greatly advanced the way research is done. Hundreds of fields and sub-fields have bloomed, and there is room for every imaginable heresy. There are peer-reviewed journals for evolutionary psychologists as well as for Jungian psychologists, for neo-classical economics as well as Marxian economics. Every school has its home.
At the same time the most prestigious outlets such as the American Journal of Sociology or The Philosophical Review are generally recognized by their disciplines as major actors: they produce the "must-read" lists, they are the place where we forge the basis of our disciplinary research standards so as to avoid the pitfall of multiplying ad hoc methodologies.
Books serve a major purpose in the humanities and social sciences. They exist alongside academic journals to cover different areas of specialism. In some disciplines, arguments shorter than a book are difficult to make, and in others it allows academia to be more connected to its public side.
Finally, the current system also allowed us to measure science. Science metrics such as citations metrics are a direct product of the current publishing model. These metrics, while they have been abused by the new public management of academia, have also permitted the rise of scientometry, and with it, of a better understanding of how research is used, more insight and more reflexivity into our own work.
Crises to overcome. Nonetheless, the current publishing model is facing several interrelated crises : from the multiplication of joke publications on the model of the Sokal Affair, to the tyranny of the publish or perish model, to the reproducibility crisis, the last twenty years have shown clear limits of the current system.
This is what does not make sense for us :
  • Researchers do not want to wait a year for their work to be read by a paying audience.
  • Being "accepted" in a peer-reviewed journal is a not a realistic make or break criterion for a paper to be "readable" 1
  • For empirical research, the current system favors a suboptimal research format where a "publication unit" consist of a team generating new data and interpreting it at the same time2
  • The digital era hit us more than ten years ago, and still, online articles look like paper articles printed on a computer screen. Most journals don't integrate an access to the paper's data, there is little to no interactivity, and actually many journals still only publish graphs in black and white...
A system to rethink from scratch. The limits of the current system are more and more visible, and a series of short term solutions were devised like band aids : some websites now carry datasets, others give access to preprints. These websites are extremely popular but they are the crutch to a broken system.... in the world of preprints, two versions of a given work coexist, the less attractive, sometimes unfinished, sometimes outdated free version on one side, the "nice", "good", "reviewed" version, on the other side, usually a side behind a paywall; this is certainly not the world we want for our research.
What if the system could be rethought completely? What's stopping us to reinvent the way research is produced, evaluated and disseminated according to our currents needs and capabilities?
What if we could offer the same opportunities to publish, to debate findings, to create journals and networks, to niche areas and to big players?
What if we could create a system where every researcher was incentivized to share their data as soon as possible in the research process and/or to contribute to the interpretation of the data shared by others?
What if we could invent a system that would record a plurality of contributions to the research processes? what is most cited? what is raising the most debates? where are we pointing out and correcting mistakes? How much mentorship do junior colleagues receive in each area?
What if we could have several "levels" of barriers, depending of the reader's level of understanding of the subject area and/or discipline? Less informed readers would be advised to trust papers that are added in the safest collections. More informed readers would want to read all kinds of publication, to learn from the debates attached to these publications and to form their own mind about what's to keep and what's not to keep.
If the academic environment was built from scratch, what would we need?


We need open access. The restricted access to some of the most important research is becoming less and less acceptable for academics. Most of us would like our research to be accessible to the greatest number as soon as possible. We want our students to have access to all of the available literature, and we don't want public money to be spent just to make the research we have written available.
We need open data. Back in the "paper era", data was not shared, but it was almost impossible to share. But now, there is no excuse for prohibiting the reader of an academic work access to a workable data set. The practice of open data is spreading slowly through some academic disciplines, while others ignore it completely. Sharing open data lacks unity in formatting and in practices. Ideally, researchers would share their data as soon as the research starts, so as to follow the highest standards of transparency.
Sharing your data is scary, so this move must be valued by our community. Everybody makes mistakes. The faster we know it, the better is it. The generalization of open data will allow for those mistakes to be pointed out, but it will also allow for a deep collective effort that can produce wonders.
We need open review. Reviews are essential for every actor of the research process. But we want them to be done openly, in front of everyone else.
  • For the most junior researchers, being reviewed is a necessity. It allows them to understand what is expected of them, what they are "allowed" to say, what they are missing.
  • For more experienced researchers, being reviewed is a more ambivalent experience. Reviews can be extremely helpful, moderately helpful, to... extremely frustrating.
Overt, civil criticism and commentary has always been a part of academia and there is no reason for the reviews of papers not to be modeled on the same model that the review of books. It was not feasible before but it is feasible today.
Readers will learn from the debates - sometimes more so than from the papers themselves. An overt record of reviews might well generate one of the largest progress currently imaginable for scientific research : formalizing, systematizing and standardizing the reviewing of research articles, the same way other academic productions are standardized.
The possibility to participate to the discussions or to revise one's research will shed the light on how research really progresses, and how a paper is really written : with the input of many, with an hypothesis changing in the middle of the process, with some key references added late in the process, with more inference than deduction. Research is a social process and it should not be ashamed of its processes: we hope that soon every junior researcher will learn that all research is debatable, that nobody has read everything, and that it is better to have your mistakes pointed out as soon as possible.


We are a group of dedicated scholars who went into this calling because we wanted to learn and to teach. This job is not the best paying, and it is not the most relaxing, and we do it because we want to advance our understanding of the world around us, and we want to share that knowledge. We have created a unique academic environment where authors, readers, reviewers and editors meet. We are providing authors tools to write their papers, readers tools to review them and editors tools to build collections of works that fit their requirements.
Academic research clearly has not yet taken full advantage of the digital era and of social networks, and we have a possibility at shaping the future we want for our community. If we think of how we would want an ideal place for this peers to thrive, we imagine a perpetual congress, where any new research project could be immediately presented and discussed with peers, data shared along the research work being done and commented by peers, young and old researchers presenting their work to their various communities and discussing the results, editorial committees gathering every other month to select what to put in their volume. The perpetual symposium.