Thursday, December 21, 2017

Realising the BOAI vision: Peter Suber's Advice

This year marks the 15th anniversary of the Budapest Open Access Initiative (BOAI), the meeting that led to the launch of the open access movement, and which defined open access thus:

“By ‘open access’ to this literature, we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited.”

A great deal of water has passed under the bridge since 2002, but as 2017 draws to an end what should the stakeholders of scholarly communication be doing now to fully realise the vision outlined at the Budapest meeting?

That is a question I have been putting to a number of people, inviting them to say what they believe the priorities should be going forward for the following stakeholders: researchers, research institutions, research funders, politicians and governments, librarians, and publishers.

Today I am publishing the response I received from Peter Suber. Peter is Director of the Harvard Office for Scholarly Communication, Director of the Harvard Open Access Project, Senior Researcher at the Berkman Klein Center for Internet & Society, and widely regarded as the de facto leader of the open access movement.

This is what Peter had to say:

Here are some recommendations for key stakeholders. I haven’t included all the useful and effective actions I find worth recommending, just those at the top of my priority list today. In almost every case I could add more detail and supporting arguments.

In some cases, I want to say much more, for example on the kinds of OA policies universities and funders should adopt, the ways that promotion and tenure committees could help the cause, the strategic restrictions to put on OA funds, and the harmful myths about OA to recognize and correct. I haven’t had time to dive into more detail here, but have almost always provided more detail elsewhere.


Above all, make your own work OA. For this purpose, it doesn’t matter whether you make it OA through journals (“gold OA”) or through repositories (“green OA”). Just do it. Make all your new articles OA, starting now. As you find time, make all your past articles OA.

Do the same with your data and code. Do the same with your books and book chapters, though it’s harder to get permission to make books and book chapters OA than for articles, data, and code.

Do the same for your preprints, if you won’t be punished for it. If you might be punished for it, that is, if you might want to publish in a journal that deliberately excludes authors who have already circulated a version of their work, that is, a journal that still follows the Ingelfinger rulethen at least understand the risk and choose your journals with care.

Don’t perform peer review for journals that violate or fall short of your standards for scholarly journals. For example, entirely apart from a journal’s quality, I don’t peer-review for non-OA journals, for hybrid journals, or for any journals, even OA journals, from publishers who lobby against OA policies at universities, funding agencies, or governments.

Write this list your own way, but set your own standards and follow them. Don’t donate your time and labor to publishers who work against the interests of researchers and research. Don’t donate your time and labor to enrich and entrench them. This is one kind of action that doesn’t take time, and even saves time. It’s an action that doesn’t require you to persuade anyone else, just yourself.

Write up a polite version of your “rejection letter” and send it to editors of insupportable journals who ask you to referee a new manuscript. BTW, you’ll be surprised how many editors send sympathetic replies.

If you have a chance to serve on the promotion and tenure committee for your institution or department, say yes. These committees create some of the largest disincentives to make new work OA, and not by insisting on high quality or deliberately trying to thwart OA. They could create some of the largest incentives to make new work OA, and not by lowering standards of quality or deliberately trying to foster OA. The best way to influence them is to join them and work for change from within.

Read your publishing contracts and learn how to understand them. If you’ve followed the practice of signing whatever a publisher puts in front of you, it’s time to stop.

If you’ve never tried to retain key rights, such as the right to make your work OA, it’s time to start. At least know what rights you want so that you can distinguish better from worse contracts.

Work for a rights-retention OA policy at your institution. If others propose one, support it. If others propose a different kind of OA policy, try to make sure it includes rights retention. More on this below, under research institutions.

When you hit a paywall trying to read an article, look for an OA edition using tools like Unpaywall, oaDOI, DOAI, the OA Button, and Google Scholar. Get good at finding OA editions when you need them.

Pass on your tips and skills to students and colleagues. Increase the demand for OA editions. Increase the expectation that research should be OA.

If one of your past articles is paywalled, and a reader asks for a copy by email, send one. But at the same time, deposit a copy in an OA repository. That will help all who need access, not just the tiny subset willing to hunt you down, write, and ask.

You needn’t become an expert in OA, but don’t fall prey to the most common and harmful myths about OA. Don’t let them lead you astray, and correct them when you hear them repeated.

A comprehensive list of these harmful myths would be long, but here are three from the top: that all or most OA is gold OA (OA through journals); that all or most OA journals levy article processing charges (APCs); and that all or most APCs are paid by authors out of pocket.

Or looking at the same myths from the other side: don’t overlook green OA (OA through repositories); don’t overlook no-fee OA journals (which constitute the majority of peer-reviewed OA journals); and don’t overlook the fact that author funders and employers, not authors themselves, pay the majority of APCs for authors who choose to publish in fee-based OA journals.

If you’re an early career researcher, you may feel pressure from your promotion and tenure committee to publish in high-prestige, closed journals. If you haven’t fallen prey to myths about OA, then you know that you can usually accede to your committee and still make your work OA through a repository. You can do this most of the time without retaining any special rights, because most publishers already allow it. You can do it nearly all the time if your institution adopts a rights-retention OA policy.

Do push back against bad incentives from your institution, but remember that this particular bad incentive is an obstacle you can step around. Remember that not all OA is gold OA, that green OA is compatible with publishing in a non-OA journal, and that OA benefits if you get a job, get promoted, and work for OA from within.

Research institutions

Launch and maintain an OA repository, or take part in a shared OA repository.

Adopt an OA policy. Make OA the default at least for scholarly articles. It could be a green-only policy, or green-gold agnostic, that is, satisfiable by green or gold OA at the author’s choice. But it should respect faculty freedom to submit new work to the journals of their choice, and therefore should not be gold-only.

Even if you think OA is more important than academic freedom, at least bring some realpolitik to bear here. Many or most faculty will disagree with you on that point, and you will need their support to adopt the OA policy.

Make sure the OA policy includes rights retention. There’s little point encouraging or requiring deposit in the repository if the repository doesn’t have permission to distribute most of the works it receives or if it must spend a lot of staff time seeking permission.

There are more than 70 examples of rights-retention policies in North America, Europe, Africa, and Asia, and more coming all the time. In the UK, you can support rights retention by supporting the UK Scholarly Communications License.

Launch an office to maintain the OA repository and implement the OA policy. Fund it adequately.

Create good incentives for making new work OA. For example, limit the review of journal articles for promotion and tenure to those on deposit in the repository.

Make download stats from the repository public. Publicly recognize papers trending in the repository or receiving press or social-media attention. Reward departments with high deposit rates.

Launch a fund to pay publication fees or APCs for faculty, students, and other affiliates who choose to publish in fee-based OA journals. Adopt strategic restrictions on how the money can be spent. These can make the money go further and create good incentives for authors and journals receiving the money. For example, see the general restrictions recommended by the COPE project or the specific restrictions adopted by Harvard’s HOPE Fund.

Collaborate with other institutions to the extent that anti-trust law will allow. We don’t really know where the anti-trust borderline is, in part because it’s rarely clarified by litigation, and in part because it varies from country to country, time to time, and issue to issue. But there’s little doubt that universities could safely undertake collective action far more often than they do today.

Monitor new and emerging OA infrastructure. First, look for platforms that are non-profit and open-source, in order to resist enclosure and corporate capture.

Second, avoid technological or business-model monocultures, in order to avoid lock-in, fragility, stagnation, and breakage. Look for open code, open standards, modularity, and interoperability. It may be a while before platforms emerge with all these properties plus the features you want. That’s a reason to monitor the scene. It’s also a reason to insert yourselves into the process, help develop the code or influence the specs, in order to increase the odds that at least one of the emerging platforms will meet your needs. 

Research funders

Adopt an OA policy. For work arising from grant-funded research, require OA for articles, data, and code. As with university policies, the OA policy for articles could be green-only or green-gold agnostic, but should not be gold-only.

The OA policy should also require rights retention, at least for the works destined for OA repositories rather than OA journals. Again, the purpose of retaining key rights is to permit the repository to make the deposited work OA without further ado. It avoids cases in which someone must spend time seeking permission, and it avoids cases in which the answer is no.

If you start by permitting embargoes on green OA, plan to reduce permissible embargoes to zero over a few years. Announce the plan in public.

If you start by permitting restrictive licenses on green or gold OA, plan to require CC-BY after a few years. Announce the plan in public.

Pay APCs for grantees who choose to publish in fee-based OA journals. As with university APC funds, put strategic restrictions on how the funds can be spent.

If you support OA by adopting a strong OA policy, covering the results of the research you fund, also consider supporting OA by funding projects to advance OA itself. Many of the most OA-friendly funders cannot do this, because it wouldn’t fit their public mission or charitable purpose. But explore the possibility in your own case. Many OA promising initiatives could use your support.

Politicians and governments

See the section on research funders above. Insure that all public funders adopt OA policies for publicly-funded research. Minimize exceptions, such as classified research or research whose release would violate medical privacy.


Support an OA repository, OA policy, and office to implement the OA policy. See the section above on research institutions.

When negotiating site licenses, don’t sign offset agreements unless the covered journals say they are in the process of converting to OA. When journals do say they are in the process of converting to OA, don’t take them at their word. Insist on putting that promise into the contract. More generally, don’t use your limited budgets to reward hybrid journals for remaining hybrid.

Think about redirecting funds from supporting paywalled journals to supporting OA. Think about a timetable for doing this. Think about doing it in concert with other institutions.

Notice that large-scale cancellations almost never trigger the faculty protests that many people fear and predict.


Here I’m only making recommendations for OA publishers and those conventional publishers transitioning to OA and sharing the goals of the OA movement.

Join the Open Access Scholarly Publishers Association (OASPA) and live up to its code of conduct. Submit your journals to be listed in the Directory of Open Access Journals (DOAJ). When authors are confused about which journals are predatory and which are not, OASPA membership and DOAJ indexing are two ways to flag your professionalism.

Another way to put this: Don’t underestimate how the existence of predatory journals, and the fear of them, are harming legitimate OA journals. Take steps to avoid becoming collateral damage.

Publish your research articles under CC-BY licenses.

Never publish in PDF-only. Always offer at least one other file format friendlier to text mining, visually impaired users, and low-bandwidth parts of the world.

Require authors to provide ORCIDs. Publish their ORCIDs alongside their names and institutional affiliations.

Don’t refuse to consider preprints as submissions. Don’t follow the Ingelfinger rule. If you’re a book publisher, don’t refuse to consider OA theses and dissertations as submissions, especially when they’re revised for publication.

If you charge APCs, then waive or discount them in cases of economic hardship.

Likewise, if you charge APCs, then don’t charge whatever the market will bear. Charge your production costs and a modest surplus to grow the enterprise.

While there’s an obvious market incentive to dismiss this recommendation, at least today, there’s some self-interest on the other side to consider as well. We may never see a market in which APC-based journals compete at the fee level to attract authors. But many organizations are experimenting with incentives to create such a market, in order to keep fees as low as possible. Don’t bet against them, or don’t deliberately step away from an advantageous market position if these incentives should start to take effect. You could even help that cause, and help yourself, by lowering your APCs to test the waters.

Finally, consider flipping subscription and hybrid journals to full OA. Look at the many approaches or scenarios for doing this, and don’t assume that flipping a journal must mean flipping to an APC-based business model.

The full range of these scenarios is documented in the journal-flipping literature review commissioned by Harvard Library in 2016, and conducted by David Solomon, Mikael Laakso, and Bo-Christer Bj√∂rk. Among other things, the study concluded, “Journals that picked a scenario that fit their circumstances were able preserve or enhance their readership, submissions, quality, and financial sustainability.”


Earlier responses to these questions from Danny Kingsley, Lisa Hinchliffe, Richard Fisher, Alison Mudditt and Dominique Babini can be read here, here, here, here and here.

A list of some of the more interesting OA developments during 2017 is available here.


Mike Taylor said...

Unsurprisingly, Peter's advice here is comprehensive, careful and wise. Thanks to Peter for writing it and to Richard for publishing it here.

I have two minor points to make in response:

First: "an OA policy [...] should respect faculty freedom to submit new work to the journals of their choice, and therefore should not be gold-only. Even if you think OA is more important than academic freedom, at least bring some realpolitik to bear here. Many or most faculty will disagree with you on that point, and you will need their support to adopt the OA policy."

I agree with the strategic insight here, but I am surprised to see Peter's wording acquiescing to the legacy-publishing industry's equating "academic freedom" with the right to choose which journal to publish in. That simply isn't what the phrase means ("the conviction that the freedom of inquiry by faculty members is essential to the mission of the academy as well as the principles of academia, and that scholars should have freedom to teach or communicate ideas or facts (including those that are inconvenient to external political groups or to authorities) without being targeted for repression, job loss, or imprisonment."). I wish Peter had more explicitly said what he meant here: "Even if you think OA is more important than the right to choose which journal to publish in ..."

And second: "Require authors to provide ORCIDs. Publish their ORCIDs alongside their names and institutional affiliations."

I am pro-ORCID, but I'm not sure I see how it is relevant to open access. Maybe Peter could elucidate?

Richard Poynder said...

Hi Mike,

I cannot speak for Peter, but I am prompted to make this comment:

I suspect you have taken your definition from Wikipedia. My experience is that there is no universal agreement as to what exactly academic freedom covers. Certainly, some do view it as encompassing – as Rick Anderson has put it – “the right to have some say as to how, where, whether, and by whom one’s work is published.”

Indeed, it may be that this is a common view in the US. The American Association of University Professors has as its first statement on academic freedom the following: “Teachers are entitled to full freedom in research and in the publication of the results, subject to the adequate performance of their other academic duties.”

It may also be a view held by some in the UK and Europe. Recently, for instance, Loughborough-based Elizabeth Gadd pointed out that some see academic freedom as extending to researchers' right "to publish what they want, where they want and when they want.”

We could also note that some academics appear to believe that requiring authors to self-archive their papers (green OA) is an infringement of academic freedom too. It is on this basis that professors at the University of Konstanz have made a legal challenge to the OA policy introduced by the University.

Unknown said...

I agree that academics (either individually or through their associations) tend to interpret the notion of academic freedom very liberally. What surprises me though is that it is almost never invoked (see for an exception) to oppose the rules “publish or perish” or, more to the point, “publish in high-IF journals or perish”, which are strongly entrenched and widely applied for evaluation purposes (grants, hiring, tenure, etc.) Does anyone really think academics can "publish what they want, where they want”? What makes the “publish OA or perish” rule so different?

Peter Suber said...

Hi Mike: Academic freedom has nearly as many fronts as political freedom. I certainly didn't mean that the freedom to submit new work to the journals of one's choice is the only front or even the main front. BTW, to me the main front of academic freedom is the freedom of scholars to teach and publish what they think is true or likely true, without any reprisals except disagreement from other scholars. It looks like we agree about this.

I brought up the "freedom to publish" objection here because it actually arises in the context I described. When faculty think a draft OA policy would require them to publish in OA journals, many will oppose it. Their opposition can be bitter and widespread. This is homegrown faculty opposition and has nothing to do with publisher lobbying. I've seen it firsthand when working with dozens of universities on OA policies. It's especially common among faculty who don't know much about OA and who, for the same reason, don't know much about publisher positions on OA.

I don't think any university has ever proposed a gold-only OA policy. But several have proposed poorly-worded green policies that led faculty to think they were gold policies. Moreover, many policy champions did too little to educate faculty about draft green policies. Many failed to explain that a policy to deposit work in an OA repository is not a policy to submit work to an OA journal. (This advocacy failure combines badly with the the common false assumption that any attempt to assure OA must be an attempt to assure gold OA because, after all, all OA is gold OA.) When weak policies or weak campaigns let faculty believe, even inaccurately, that they will lose even some of their freedom to submit new work to the journals of their choice, you'll see this objection arise spontaneously, and the protesters will object in the name of academic freedom.

Mike Taylor said...

to me the main front of academic freedom is the freedom of scholars to teach and publish what they think is true or likely true, without any reprisals except disagreement from other scholars. It looks like we agree about this.

We certainly do!

I do absolutely agree, as well, with your pragmatic argument for not advocating Gold-only policies. Without a doubt, restricting choice of publication venue pointlessly antagonises potential allies. What I am concerned about here is ceding the meaning of an important term to what our opponent would like it to mean. We have all heard the flagrant lie "open access infringes academic freedom" over and over. I want to reject every aspect of falsehood in that statement.

Anyway: we are, unsurprisingly. very substantially in agreement.