Open access is the principle that the output of academic research, in particular research papers, should be made available to everyone at no cost on the Web. This principle is intuitively reasonable but not yet widespread. This page describes briefly the issue of open access, presents my perspective on it, and outlines the steps I have been taking to push academia towards open access.

In summary: I try to avoid submitting to closed-access conferences and journals, I refuse to review for such venues, and I put the definitive version of my papers online: you can jump to the specific commitments if you don't care about the motivation. If you wish to take specific commitments yourself, you can sign the pledge No free view? No review!.

What is open access?

At first glance, the publication of scientific papers looks a lot like newspapers: scientists, like journalists, write papers; journals publish the papers; and readers can pay a subscription fee to have access to the papers.

There is a key difference, however: newspapers pay journalists, and their subscription fees partly go towards supporting the journalism. But academic journals do not pay researchers or peer reviewers. All the subscription fees go towards financing the journal's operations and its shareholders, not towards financing research.

Academics want their work to be widely available and read (and cited!) by as many people as possible. In the light of this, it seems utterly incomprehensible that their work can only be accessed by paying a subscription fee that doesn't bring any benefit to them.

The reason for this is historical: before the Internet, distributing scientific articles was a difficult and costly business, which relied entirely on publishing houses. But the Internet has appeared, making the distribution of knowledge easy and essentially free, and academia hasn't caught on yet.

Virtually anyone from outside academia who learns about the academic publishing system believes it is absurd. But within academia, researchers do not care: they work in institutions which are subscribed to the journals that they need (at the cost of millions of dollars to taxpayers), they do not hope that their work can be of interest to anyone else than professional researchers subscribed to these journals (a self-fulfilling prophecy), and they are simply used to the current system being how it is.

For this reason, academics do not see open access as an important concern. When pushed, they will defend the current model by arguing about the "cost" of open access, i.e., of hosting articles online1. They do not see the absurdity with commercial journals charging upwards of 2000 EUR to publish without a paywall. In the real world, hosting PDF files online is essentially free: you can host gigabytes of research articles for free on the reputable Internet Archive, you can host articles for free on arXiv which is operating at a yearly cost of less than $2 per article. But for now, academia doesn't seem to have noticed.

Why do I care about open access?

I have already written many times about the issue of open access: it is the first one in my list of problems with academic research, I have explained in detail everything that open access would make possible. Yet, none of this seems to strike a chord with my colleagues, who don't seem to understand why I care so much about this small inconvenience, instead of ignoring it like everyone else. So here's an attempt at explaining why this personally matters to me.

I'm old enough to remember what the world was like before having the Internet at home (TL;DR: it was boring), but still, the Internet appeared pretty early in my life. I was fascinated by it as a kid, and I had my first webpage in middle school (fortunately no longer online). I think what I liked was the possibility of making what I did available to a huge (virtual) audience. Of course, getting anyone to care was an entirely different story, but sometimes I'd have the joy of seeing that people had stumbled upon my page -- the Web was a much smaller place back then. So, for all the bad programs and other ramblings I was writing at that age, my end goal whenever I was proud of my work was to put it online for the world to see. While creating stuff on my own would have seemed boring, putting them online felt meaningful.

As I teenager, I discovered free software and permissive licenses such as Creative Commons. This went even further: not only could everyone access what you created, but they could also redistribute it, build upon it, etc., provided they acknowledged their source. This recognition was more than I asked for – I certainly didn't need anyone to pay me. So now I was putting everything online under an open license, and I had the feeling of being part of this grand movement of openness and freedom.

When I started doing academic research, I was attracted by essentially the same motivations. Research is about contributing to something which is larger than yourself, which belongs to everyone, and is ruled by no one. You want what you do to be seen by others, as widely as possible, because your salary does not come from royalties. But as it turns out, there is big difference between academia and the open source community: academia was born before the Internet, whereas open source was born with it. So open source happens openly on the Internet (with a large hobbyist involvement), whereas academia is entrenched in existing prestigious institutions and paywalled journals and run by highly specialized professionals.

The academic publishing model made no sense at all when I discovered it (during my master's internship in 2012). I was extremely dismayed, when I published my first article, to have to sign a copyright agreement promising that the article could only be distributed by the publisher, and would only be accessible to subscribers. I was proud of my work and wanted to share it with the world, but I wasn't even allowed to put it online myself under a free license, because some company needed to get exclusive rights so they could "help make my research available" by hiding it behind a paywall. This was ridiculous. I stayed in academia nevertheless, choosing to ignore this cognitive dissonance during my PhD in the hope that I would eventually understand.

Several years later, being done with my PhD and having a full-time tenured job as an academic, I have a better understanding of the historical reasons of how we got there, and I realize that moving away from this model is going to be hard. But this hasn't changed my mind: it is an absurd model. So I figured I would take action.

My personal commitments for open access

Academia's current publication system is outdated and broken, but moving out of it won't happen overnight. Academia has no leader to steer it in the right direction2. Everyone in academia is used to the current system, companies such as Elsevier make billions of dollars off it, all academics are used to the names of prestigious journals in their fields that are owned by commercial publishers, all hiring and promotion decisions are made by looking for these journals in résumés. What's more, moving towards open access is an extremely low priority on everyone's radar3.

So the most realistic route towards open access involves advocacy, activism, getting researchers to care about this issue (as opposed to their own research), and mostly waiting for more digital natives to be old enough to reach the top of the academic pyramid. I'm doing this kind of things to some extent: I'm a board member of the CAPSH association for open access that started the initiative No free view? No review!, I'm that annoying guy who raises that boring issue at every single conference and occasion, I'm trying (with little success so far) to organize colleagues in my area to act together, and I'm pushing with all of my minuscule weight against academia's unfathomable inertia.

That said, as a tenured researcher, and as a human being who cares about doing things according to my principles, I also want to act individually, and do the right thing for the choices that I can make. So here are my personal commitments for open access:


(Since times immemorial:) I will put the definitive version of all my research output (including all corrections, reviewer feedback, etc.) available for free online, on this website and/or on a centralized, well-archived repository such as arXiv.

This disclaimer appears at the top of my publication list: the most up-to-date version of my work is always the one available for free online. It is never the paywalled version that publishers are selling because of the outdated system.


(Since 2016:) I will not publish single-author articles in closed-access conferences and journals. For collaborations (= virtually all my research), I will argue with my co-authors against publishing in such venues, and advocate in favor of open access alternatives.

Sadly, I cannot promise never to publish in closed-access venues, because most of my articles are collaborations, and my co-authors also have a say in where the articles get published: this is a collective choice not an individual one. My co-authors sometimes have more at stake than I do: e.g., because they lack job security and need to publish quickly, or simply because they have contributed more to the paper than me. In all cases, they usually have different concerns from mine, and we need to make compromises to work together. But I will always argue against closed-access venues, and make it clear that we are submitting to that venue at their insistence.

I will also try to be less involved when the publisher process in closed-access venues, and care more about the actual publication of the work online in open access (see "Self-archiving" above). In particular, to the extent possible, I will avoid signing any copyright transfer paperwork, relying on my co-authors to do it when possible, but this is just personal preference. I will also advertise open-access versions of the work instead of the closed-access publisher version.


(Since 2016:) I will not review articles, or serve as committee member, for closed-access journals and conferences.

I have had this stance for several years, and it is the one that has attracted the most criticism, because journals and conferences crucially rely on the free work of reviewers to be able to function. For that reason, people have tried rather hard to convince me to do otherwise. But reviewing is an individual choice, and I want to make the right one. Besides, I have plenty of reviewing to do for other open access venues, so it's not like I'm not contributing at all here.


(Since 2016:) I will think twice before paying the registration fee to a closed-access conference, or otherwise engaging with such conferences.

This is another question: paying the registration fee of closed-access conferences is another thing that supports their model. For now, I am not ready to commit that I will not attend such conferences, but I am considering it.


Are other researchers with similar commitments?

Yes! With a friend, we have even started an initiative allowing you to take such a committment: No free view? No review!.

Here are some specific examples of people that I know about which have taken such a pledge (some of which are signatories of the "No free view, no review" initiative):

There are other movements targeted at a specific publisher, such as The Cost of Knowledge or the defunct No deal, no review, or targeted at specific journals, e.g., against Nature Machine Intelligence. There are also initiatives such as Research without walls or Free our knowledge.

I will update this page if I know of other examples.

When is something open access?

There are many open access models, so the picture is somewhat complicated. Basically:

I reserve the right to make my own judgement calls, and bend these criteria when I think it's a good idea.

Do you refuse to publish in closed-access venues?

I try to avoid publishing my work in closed-access venues. However, most of my work is collaborations, and some of my collaborators do not share my principles (for reasons such as those outlined above). When my co-authors insist, I accept to publish my work in closed-access venues, but limit my involvement with the publisher, and advertise the open-access copy of the work on my webpage instead of the closed-access publisher version.

Isn't it unethical to refuse to review for venues where you publish?

I get asked a lot about the apparent inconsistency of accepting to publish in closed-access venues (point above) but of refusing to review for such venues. Some colleagues argue that this is unethical because it violates the norm of reciprocity: they see a moral imperative to help do reviews for the places where you publish. Not everyone interprets this imperative in the same way.

I understand that this point is subtle. Here is how I justify my stand:

I will also add that I dislike implicit imperatives where you get blamed for refusing a reviewing request based on vague notions that you ought to accept it. If a conference or journal only wants to welcome papers for which every author would be willing to serve as a reviewer, then this should be spelled out in the call for papers.

Are you not worried about the impact on your career?

Taking a stand for open access, and refusing to review for prestigious closed-access venues in my field, has certainly made my life a bit more complicated, and attracted some well-meaning concern from colleagues about the impact on my career (not always easy to distinguish from community pressure).

Fortunately, there are enough open access venues in my research field to keep me busy and to publish my research, so my stand in favor of open access isn't making me an outcast.

In any case, the honest answer is: I care about my work, and I care about being able to publish what I do openly and without paywalls so anyone can access it. I value this more than career advancement in the traditional, narrow sense. So if I ever feel that my principles are standing in the way, and academia isn't giving me enough opportunities because of them, then I'll consider doing something else with my time.

Why not limit your boycott to commercial publishers?

I agree that the worst offenders are closed-access commercial publishers such as Springer or Elsevier, who will milk the current subscription model as long as they can, and even try to make money off open access publishing too, so as to maximize their shareholders' profits (Elsevier had a 37% operating profit margin in 2018). By contrast, closed-access scholarly societies, such as ACM or IEEE, are theoretically not-for-profit and are closer to the interests of the community, so there is some hope that they can eventually adjust their model in the right direction.

Nevertheless, I do not believe that it is reasonable for these societies to put articles behind paywalls, especially given their alleged non-profit nature and their claimed mission to advance research. So I'm not limiting my open access stance to the worst offenders, and I'm also applying it to them until they adjust to a healthier model.

Will individual boycotts change anything?

They won't. Trying to move society in the right direction doesn't only happen via individual choices, and the same is true of academia. So a more effective course towards open access is collective action, which I also pursue, e.g., with the initiative No free view? No review! to try to get more people onboard.

But my main goal with these personal commitments isn't to make open access win or to get the rest of academia to agree. My goals here are to follow my principles, do what I believe to be the right thing, make my work as openly available as I want, and avoid wasting time dealing with an outdated model.

Are you judgemental towards people who don't share your views?

Fortunately, I'm not, otherwise I would have complicated relationships with essentially all my colleagues. :)

I understand that my strong preference for the open access model may not be shared by everyone, and that others may care less about these issues that I do. I accept that other researchers, especially at earlier stages of their careers, or in other fields with no reputable open access venues, may not be in a position to take action.

I will, however, confess that I feel some disappointment with how the academic community is collectively failing to update its practices and has been slow in moving away from a model which has now been outdated for decades. This is especially disheartening given that tenured academics have considerable freedom and can take essentially any stand they want with extremely little risk compared to other professionals: see my list of problems in academia for more about this. In any case, this is a collective failure, and I don't blame anyone individually for it.

Can I draw inspiration from this to design my own pledge?

Please do, I'd be honored! I also encourage you to put it online to help change mentalities, and I can link to it, just drop me a line.

This page is licensed under the Creative Commons Attribution 4.0 International license.

I don't agree with you on something

Sure, I'm open to discussion. Feel free to reach out by email at a3nm<REMOVETHIS>

  1. To give orders of magnitude: most computer science research is on DBLP, which recently passed 5 million references. Estimating around 1 MB per article, this suggests that the entire corpus of research in CS fits on one single hard drive... 

  2. The closest thing to a leadership is countries, grant agencies and universities that sometimes mandate requirements on the research that they fund. But this is moving very slowly, and not always in the right direction thanks to copious lobbying by publishers. 

  3. There are also some questions about the precise solution that we should move to, and how to support the moderate costs of publication: we need orders of magnitude less money than what is currently paid in subscriptions to paywalled literature, but it is tricky to change how the money is allocated. The picture is further complicated by commercial publishers offering open access at a charge (so-called "gold open access"), spammy predatory publishers being "open access" because no one would ever subscribe to them, etc. So "open access" can now mean lots of different things depending on who you talk to.