Unprecedented Case Points to Need to Make Open the Default for Communicating Research

Unprecedented Case Points to Need to Make Open the Default for Communicating Research

The State of Open Access: An Interview with Peter Suber of Harvard University

This is an edited transcript of a webcast featuring Dr. Peter Suber, Director of the Harvard Office for Scholarly Communication and Senior Researcher at the Berkman Klein Center for Internet & Society. The webcast occurred live on October 18, 2016, and would not have been possible without generous sponsorship from the Public Library of Science.

SPARC and the Right to Research Coalition are grateful to Dr. Suber for his time, energy, and expertise.

Interviewer: Joe McArthur (Right to Research Coalition). Transcriber and Editor: Scott St. Louis (Grand Valley State University)

Next February, the Budapest Open Access Initiative (BOAI) will turn fifteen. What do you see as the greatest obstacles to Open Access that the movement has overcome, and the greatest obstacles that it has yet to overcome?

Thanks, Joe; I’ll be happy to dive into that. First, let me just say how happy I am to be here, with one qualification: I wish I were not doing it remotely! I wish I were live in front of the audiences that are tuning in. OpenCon is the biggest of the conferences I’ve never been able to attend in person, so I wish I could be there.

But you’re asking a good question: not just what are the obstacles today, but what are the obstacles we’ve overcome in the past fifteen years? You have to remember that in 2002, when we issued the first statement from the BOAI, Open Access was at least a fringe movement. It was not mainstream, and so one of the first obstacles we overcame – gradually, not quickly – was to make it mainstream. Open Access is clearly now mainstream. Roughly thirty percent of peer-reviewed journals are Open Access. Roughly half of all new research articles become Open Access one way or another within the first two years of their publication. Major universities and major funding agencies require it; major bills requiring Open Access have been introduced in the U.S. Congress and have been adopted in other countries. It’s moved from the periphery to the mainstream. That was big. If you want to go back to how it moved from the periphery to the mainstream, that’s interesting. We can get into that, but it helped that major institutions, and not just fringe activists, endorsed the idea.

At the time, the people brought into Budapest to draft the Initiative were all working on what we now call Open Access, but the question for us was, are we all working on the same thing? Are our efforts convergent? Could we all cooperate? Could we amplify each other’s efforts? Could we unify? Could we make some kind of movement out of this? At the time, the answer was unclear, which is why we had to raise the question. Now, I think the answer is clear. It [became] clear at the meeting: the answer is yes … By the way, Open Access didn’t have a name at that time! I was writing about it, but I called it free online scholarship. We agreed to call it Open Access, so we introduced the term. We introduced what I still think are the two primary strategies for Open Access, namely Green and Gold, or Open Access through repositories and Open Access through journals; those remain the primary channels for Open Access to journal articles, and we got the first significant funding for Open Access from the Open Society Institute, now called Open Society Foundations.

We overcame these obstacles in part because our ideas were good. It was an important moment in the middle of the period between the launch of BOAI and today when publishers hired a very aggressive lobbyist to counter our efforts in Congress, and he was notorious for his aggression, even for his deceptive tactics. But he had to tell his clients – which we know from leaked papers – that the problem of fighting against the Open Access movement is that our arguments are better than the publisher arguments! So, part of what enabled us to become mainstream was that we had good arguments, and even though we were just individuals who didn’t represent important institutions, the arguments persuaded important institutions. We didn’t need institutional buy-in if we got wide enough individual buy-in, but we got the institutional buy-in, which in turn helped persuade individuals … and all of that helped. We have good arguments because Open Access, in my humble opinion, is unambiguously good. I can’t think of a single downside to Open Access, even after all this time.

At the BOAI, those two strategies [Green and Gold Open Access] were really enshrined as the way forward … Was there a third in mind then, or others that didn’t make the cut, so to speak?

No, but the cynical way to read what we did is that there were people in the room who favored Green over Gold, and other people in the room who favored Gold over Green, and we ended up endorsing both because it was the best way to reach consensus.

I don’t take that cynical reading. I endorse both because I endorse both. I actually think they’re compatible and even synergistic, and I’ve been arguing that ever since, but not everybody in the room thought that. So, it was a big-tent kind of document that embraced what at the time were the different factions of a very, very small movement.

But since then, we’ve discovered other channels that could be used, and sometimes are used, for distributing Open Access research. Some of them are very familiar: blogs, wikis, listservs, peer-to-peer networks. They’re all valid channels; that is, what they contain could well be Open Access, and sometimes is Open Access by default. They may not be the best containers for Open Access research, and by the way, journals may not be, either. Repositories can host arbitrary file formats, and so they’re more adaptable. Journals are evolving – maybe we’re already in an era of post-journals – but we should acknowledge that there are far more than two channels.

It’s just that even today, after all this experimentation, most of the OA literature is in one or both of those two channels – repositories and journals – and most of the discussion is about those two. But every now and then there was a serious experiment with some additional channel. There was an attempt to share very large datasets through BitTorrent, because the files were too large to share any other way. There were experiments to post journal articles to wikis, so the version published by the journal froze at the moment of peer review approval – you could call it a quality ratchet – but another version went to a wiki, where the public could revise it, enhance it; and no matter what they did to it – whether they improved it or degraded it – the journal version remained unaltered. That was a nice experiment. It didn’t last; I don’t know of any journal that still does that, but I’d like to encourage more of that experimentation. I think one of the exciting, uncharted frontiers for Open Access is to think about new containers, or structures for containing research, other than the journal article, other than the dataset, other than the conference presentation, other than the dissertation. The digital world gives us the flexibility to try new formats, and even if we came up with one that was very good, we’d have the usual problems of inertia in getting people to adopt it, but we now have a chance to experiment that we didn’t have before.

I think you’ve started with a really excellent overview of what the past fifteen years have looked like, and the strategies that we’ve adopted, and some of the big wins we’ve had as a community, as a movement. Looking back, is there anything you wish had gone differently, or that the movement could have tried to do differently to be more effective?

We’ve made nothing but progress since then. The progress has not been fast as we wanted. It’s not been as fast as we hoped. In my opinion, it’s not been as fast as the opportunities allowed. There were chances to make faster progress that we just for some reason couldn’t seize, but I don’t think we took any seriously wrong turns.

The Open Access movement has grown large, and as it grows large, it contains more internal disagreement. Everybody acknowledges that, and by the way, it goes back to the beginning. The Budapest meeting, which was very small – about a dozen people – contained some of this disagreement already. But the fact that individual Open Access proponents, or even some cultures within the Open Access movement disagree with one another is not a failing; it’s not a mistake. It’s not something we could have avoided, even if we had wanted to avoid it. I think it’s the price to pay for scaling up, and including many people of many different kinds.

You have to remember that Open Access is proceeding at different rates in different disciplines, because the disciplines are themselves different in relevant ways. They have different funding mechanisms, different levels of levels of wealth, different sizes, different customs and practices for distributing research. So, even if we all agreed on the principles of Open Access, we would still have differences arising from our fields, from our disciplinary perspectives. So, I don’t see those internal differences as a mistake that we could have avoided. I think it might have helped if we had anticipated them more, but are there any mistakes? Can anybody allege a mistake that was made? It’s very hard to say, in part because there was no top-down leadership of the Open Access movement, so we can’t point to a history of decisionmaking by them. I include myself there. Obviously, I made my own decisions, but I didn’t speak for everybody.

I think we could have moved faster. I still think we can move faster, but when I think about the reasons why we’re not moving faster, they all seem to be deeply entrenched in institutional practices which are very, very hard to change. I think one thing we could have appreciated at the beginning was that that’s what we’re up against. It looked to most of us at the time as if Open Access were a matter of taking advantage of this cool new technology … and nothing more. We had the Internet. We had authors who were fortunately situated, so they didn’t have to be paid for their work. It looked like if you put the two together, suddenly you had Open Access, and it’s true that in a sense you did, but only when you have willing authors. How do you get willing authors? Authors follow the incentives that exist in their time and place, in their institutions, and those incentives are based on long practices that antedate the Internet. So, it’s cultural change; it’s institutional inertia, much more than technology and economics, that dictates the rate of progress. I don’t think we appreciated that at the time.

If we had known that the long-term prospect was to change the culture of academic publishing, then I think we would have been less surprised by the slow rate of growth. But in light of the fact that it is cultural change that we’re talking about – making deep changes to institutional practices – I think it’s actually gone faster than other kinds of deep institutional change. So, I’ve learned to live with our slow rate of progress, but I still see lots of allies and colleagues who mistake the slowly rising curve of progress for a flat curve – as if we were making no progress – and sometimes even for a regress. The progress has been steadily up. The slope of the curve is up. It always has been up, ever since we started. It’s just not up as steeply as we would all like.

So many questions have been submitted about research evaluation, it’s actually hard to know which specific one to pick! Do you have any reactions about the topic of research evaluation and how we should be focusing on it now?

Promotion and tenure committees are a major obstacle. They are a bottleneck. They are in a position to create good incentives or bad incentives; mostly, until now, they’ve created bad incentives. They are staffed by us! That is, they are staffed by academics, and these are academics who’ve published research. Some of them support Open Access. They see themselves – correctly, I think – as the gatekeepers of quality. They want to hire good people who have written good work, but if you look at the customs that guide their practices, how do they judge whether someone has written good work? They look at the impact factor of the journals in which they’ve published, which is a very bad metric for that, and it’s a scandal that PhDs don’t understand that. It only takes a few minutes to explain it, and when they get past that and say, “Okay, let’s not use impact factor. What other metrics can we use?” The short answer is, read the darn articles! If you’re experts in the field, if you’re evaluating people from your own field, read them! Now, your judgment about them will not necessarily be objective. It will be your judgment, and there will be other judgments by other members of the committee, and all of our disciplines contain internal disagreements, so even people who have read these articles carefully, firsthand, might disagree with each other, but that comes much closer to an actual evaluation of quality than to pretend that an impact metric is a quality metric. Again, that’s a fundamental mistake that smart people should not make.

In the very beginning; that is, when Open Access was brand new – in fact, when digital anything was brand new – promotion and tenure committees didn’t give credit for digital journals of any kind, whether they were Open Access or not. The idea seemed to be that real scholarship isn’t digital; real scholarship is ink on paper, which, of course, is silly! In this case, it’s easy to tell that it’s silly because we have overcome that. We don’t think that way anymore; it only took us a few years, but the idea that it took years instead of minutes is a scandal! It’s like saying that real scholarship is inscribed on sheepskin. It obviously has nothing to do with the quality of research. But it held us back, because smart people were being very conservative about the traditions of their institutions. That’s what I mean by a cultural barrier.

So, sometimes, good people – namely, ourselves – are on these committees enforcing old criteria. But sometimes, they are people who aren’t even clued into the controversies, or aren’t sensitive to the incentives and disincentives that they’re creating. One of the worst incentives which accompanies the impact factor, but isn’t the same thing, is to give career points to people who publish in venerable, old, well-established, high-prestige journals, and either no career points or negative career points for publishing in anything else. Again, taking the prestige of the journal as the metric for quality of articles is a simple mistake; it has nothing to do with the quality of the articles. It’s a way of outsourcing judgments of quality to publishers, which is a mistake. Academics serving on these committees ought to be the judges of quality themselves.

Impact factors, or getting your work accepted into a high-prestige journal, is a kind of endorsement from an important source. But that’s all it is, and it might be that the article that you’re looking at from a candidate might be the worst one in the journal that year, dragging the average down; it might be the best one in the journal that year, dragging the average up. You won’t know unless you read it and apply your own expert standards to the piece. So, I think promotion and tenure committees have to stop using publisher prestige as a surrogate for author quality.

The Public Library of Science was pioneering in looking at article-level metrics as opposed to journal-level metrics; that’s good. At the beginning, most article-level metrics were citation-based (or, they were impact-based). That’s understandable, because we don’t have quality metrics! But we can get closer and closer to quality metrics – and further and further from crude, oversimplifying citation metrics – as we add altmetrics, as we add other measurements to the mix. So, I support the altmetrics movement, but so far I don’t see any of them coming very close to judgments of quality … and to evaluate somebody for hiring, promotion, or tenure, we really have to judge quality, not just impact.

…

Now, after they have read them, and have some estimate of the quality of the article, and some estimate of the quality of the person, they can supplement that with citation metrics. I had a colleague who came up for tenure in the early days of the Open Access movement, when Google was new, and he was trying to give his committee some idea of the influence of his work, so he gave the Google rank of some of his papers. Committees didn’t ask for that then; they still don’t ask for that … He wasn’t pretending that it replaced anybody’s evaluation of his quality, but it was supplementary. So, as long as committees are trying to look at quality, let them also look at all kinds of supplementary impact measurements.

One thing I didn’t add before is that one disincentive created by these committees – especially when they ask for their authors to publish in high-prestige, venerable journals – is that they are steering them away from Open Access journals, regardless of the quality of the Open Access journal. The reason is that, on average, Open Access journals are much younger than non-Open Access journals, and the younger they are, the less time they’ve had to develop prestige in proportion to their quality. So, they could be born excellent, but they take time to acquire a reputation for excellence … [When you ignore evaluations of] quality and jump to prestige, you automatically favor the older journals, regardless of their access policies. That’s a disincentive to publish in Open Access journals. It still exists, and many young scholars feel that they cannot publish in (at least most) Open Access journals, because they don’t meet the standards set by their promotion and tenure committees. That’s not a problem with the journals; it’s a problem with the committees.

For people who are trying to change the tenure and promotion system, do you have any strong feelings about how that could be done, about what the strategies are to do that? It seems like a hard problem to solve, but one that has really come into its time.

It’s a very hard problem. I’ve tried to do it at more than one institution. One reason it’s hard is that promotion and tenure committees tend not to be governed from the top, because the members would resent that; faculty would resent that. The committees make their own decisions, and then those decisions might be advisory to somebody above them, like a president – and that’s when the presidential criteria come into play – but the committee gets to use its own criteria, so they don’t want anybody to tell them what their criteria should be. There’s also a pull-up-the-ladder sense, that “I got my job because I complied with the old standards, and I’m not going to let anybody in who doesn’t comply with the same standards I did.” But at most schools, including the school where I am now, every promotion and tenure committee runs itself. For example, I’m at Harvard, and every department has its own promotion and tenure committee, and they all do it their own way. There’s no uniform set of guidelines for promotion and tenure committees, so even if you could persuade one to change, it’s not the same as persuading them all to change. This might be different at some institutions that are more uniform across their departments and schools – Harvard is notoriously disunified that way – but you can’t change them from the top down, and if you can’t do that, you have to persuade them one at a time, and their membership keeps turning over.

So, practically the only recommendation I have is that if you’re a faculty member, and you’re asked to serve on the committee, or you ever have a chance to volunteer to serve on a committee, take it! Say yes, and be one of those people who can help change it from the inside.

Institutional repositories have always struggled to get researcher buy-in, and now many subject repositories, or preprint repositories, are popping up. What do you think is the future of institutional repositories, and how can we take our experience with institutional repositories into the realm of preprints as we make big pushes for those?

I’m a strong proponent of institutional repositories, and unlike some of my allies, I also support disciplinary repositories … I support them both … They both support Green Open Access, and I support Green Open Access. I don’t support it to the exclusion of Gold, but I support Green as well, and I support any platform, and any policy, and any practice that gives us more Green that we used to have.

Institutional repositories are growing. There are four thousand Open Access repositories around the world, according to the Registry of Open Access Repositories. I don’t know how many of them belong to universities as opposed to other kinds of institutions. It’s a large number. I wouldn’t be surprised if a large fraction of them were experiments that died off and are not active, but the active ones are providing a very good service. At every school that has one, at every school that has an ongoing effort to fill it, they’ll find that some faculty would rather deposit in a disciplinary repository, especially the popular ones, like the physics arXiv at Cornell, PubMed Central, and SSRN. I see no problem with that. Again, the work is being made Open Access. If the institution has its own reasons to want a copy in the institutional repository, they can try to get one. They can either get it from the author, or they can harvest it back from the [disciplinary] repository.

One way that we could make further progress is if all Open Access repositories were better than they are today at permitting this sort of mutual harvesting. Believe it or not, they don’t all cooperate with this. Some do, some don’t. But there are reasons why institutions might want their own copy. All those reasons, in my view, are less weighty than the reason to make the work open in the first place. So, if a work is open through the arXiv, and not through the Harvard repository, and it happens to be by a Harvard author, I feel that the pressure is off; at least it’s Open Access. If we can also get a copy for the Harvard repository, good for us. We can do good things with it, including preserve it in the Harvard Library, which is something the arXiv cannot do. But I’d rather have the work be open through the arXiv than not be open at all.

…

The question seemed to assume that disciplinary repositories grew up after institutional repositories, but they didn’t. The earliest Open Access repositories were disciplinary, not institutional. I’m thinking primarily of the arXiv, which was originally at the Los Alamos National Laboratory. That was launched in 1991, which is ancient in Internet time! There’s also the Department of Education’s Open Access repository for educational work. Both of these were pre-Web. Institutional repositories came later. They are proliferating in numbers, and so are the disciplinary repositories, but if we have to say which came first, I think we’d have to say it was the disciplinary repositories.

It seems like there’s been an increase in the number of preprint repositories, especially in the fallout from SSRN being acquired by Elsevier and things like that. Do you have any opinions on that trend?

I apologize for leaving that out of my previous answer, because it was part of the question. First, institutional repositories – at least, like ours here [at Harvard] – accept preprints. So, it’s not as if you have to go somewhere else to provide Open Access to your preprint. There may be institutional repositories that don’t. We focus on post-prints, like most of them, but we accept preprints. I hope that’s a common practice, and if it is a practice at your institution, I hope the institution tells people about it. We tell our faculty about it.

On the other hand, if you’re in a field like physics, mathematics, or computer science, and you want to share your preprint, I understand perfectly why you’d rather do that through the arXiv than through the Harvard repository. The eyeballs for that content go to the arXiv; they don’t come here. By the way, to find stuff in the Harvard repository, you don’t have to come to Harvard and run a local search. You can run a search on practically any search engine, including Google and Google Scholar, and that’s true with every well-configured repository. But many repositories do receive visitors who run local searches, and the biggest ones – like the arXiv, SSRN, and PubMed Central – do that. I can understand why you’d want to put your work there, where people go to look for it. No problem with that.

I completely support Open Access to preprints, just as I support Open Access to post-prints. There’s a lot of value in it. One interesting question for cultural historians one day will be, why did preprint Open Access take off faster in physics, mathematics, and computer science than it did in other fields? Not just other natural science fields, like chemistry, but also fields in the humanities? The benefits they bring to the researchers are roughly the same in every field, so it’s a nice question.

But one of the benefits is the early timestamp on your work – and this might apply more in the natural sciences than in the humanities – but if you put something in the arXiv, which you can do just a few minutes after you’re ready to do it, you get a public timestamp that minute … [acknowledging] that you made this discovery, or you came up with these findings, perhaps earlier than someone else working on the same problem. For the sake of your career – hiring, promotion, and tenure – that timestamp is more important than the name of the journal in which you eventually publish.

So, preprint archives give the earliest possible timestamp. That’s valuable. They also give the author feedback on their work before they submit a version to a journal for formal publication. That can be valuable. So, you’ve got a draft of an idea, or you’ve got findings from an experiment: you write them up in some preliminary version, post them for comment, that allows you to submit a better version for publication, and that value, that benefit, should apply to the humanities as well as the sciences.

When talking to researchers, they often will only be aware of making their work Open Access by way of paying article processing charges (APCs), and generally consider Open Access to be too expensive for them, unless they work at a very well-resourced institution. Do you have any insights or thoughts on why that route gets so much more attention from researchers than self-archiving and preprints?

It’s not only getting more attention than Green Open Access; it’s getting more attention than no-fee Gold Open Access. Fee-based Gold – the APC-based Gold – is one business model for publishing an Open Access journal. It’s a valid business model. I support it. I administer a fund here at Harvard to pay the fees of journals like that. I don’t think it’s a failure of openness, as some people say. On the other hand, it’s a minority model. Most people think it’s the only way to support an Open Access journal, the only way to get a revenue stream to pay the bills. It’s not true! It’s not even true that it’s the dominant way.

Every estimate for the past ten years or so has shown that the fee-based model only covers about thirty percent of peer-reviewed Open Access journals. Every study shows that a vast majority – roughly seventy percent – charge no fees at all. Some people call these Platinum or Diamond Open Access. I don’t, just because I don’t want to get into terminology wars, and I want the term Gold Open Access to refer to journal-based Open Access, regardless of the journal’s business model. But apart from terminology, we just have to recognize the substance. Most Open Access journals – most peer-reviewed Open Access journals – don’t charge any fees at all.

So, it’s one thing to be deterred by fees, and say “I can’t afford that,” “My institution has no fund for that,” “My grant doesn’t let me pay for that,” or “I have no grant.” But it’s another thing to mistake fee-based Gold Open Access for all Gold Open Access. Again, that’s a failure of information. We can help illuminate that by informing people that the fee-based model is not only just one among many, but it’s the minority model.

Then there’s the fallacy that says Gold Open Access is the only kind, or all Open Access is Gold, and there isn’t even such a thing as Green. That’s also a lack of information. Call it unfamiliarity, call it ignorance. It’s our fault because we have to do better to inform people who are not closely following this movement that we follow very closely. But we do have to inform people that if you publish in a traditional (that is, non-open) journal, much more often than not, you have permission – standing permission from the contract – to put a version of the article in an Open Access repository. Most people don’t know that. People who know it have been saying it, endlessly, for years and years. We’ve been doing our best to spread the word about it. The word has not gotten out very far. Going back to the obstacles that remain even fifteen years from Budapest, we still haven’t taught people that there is this Green alternative next to Gold. I think people who have not been following closely understand journal-based Open Access because they understand journals, or at least they understand them more or less. They don’t understand repository-based Open Access because that’s something new in the landscape. They didn’t exist before the Internet. It wasn’t a serious alternative, and the idea of publishing an article, and then doing something else with it to improve its access, was unknown.

So, we have an education job to do. It’s not as if these options don’t exist. It’s not as if people who don’t have money to pay the fees have no choices, and they just have to suffer. They do have choices. It’s just that the choices are not well known, and our job is to help publicize those options, make them known to everybody.

A lot of the questions that we received were about APCs and corporate publishing models. Are there less traditional corporate publishing models that we should be adopting or moving toward? How can libraries better fit into these new models for Open Access? How do these questions connect to your earlier remarks about containers for Open Access research?

There are several questions in there; let me disentangle them. You asked how libraries could involve themselves. When an Open Access resource comes along, whether we call it a journal or something else, it needs some funding. Its work is not altogether provided by volunteers, even if quite a bit of it is (and by the way, the labor behind a conventional subscription-based journal is also largely provided by volunteers). But it needs some revenue, so many of them turn to libraries, and they say, “You’ve been paying for conventional, peer-reviewed research through journal subscriptions. Why not pay for peer-reviewed research this other way? The advantage of paying for what we do is that the result is Open Access to everybody.”

That’s a pretty good pitch, and many libraries pay for it. I’m inside a library now, and I know that many libraries are not used to paying for this; that is, they know what it’s like to pay a subscription or to buy a book, but they don’t know what it’s like to buy a membership fee, or to provide start-up funding for some good idea. So, they don’t have standing policies for this. They may have the money, but they don’t have a decisionmaking procedure. This is something that we’ll get over. The world is changing; libraries are changing, too. Some libraries are ready to do this, and already do it on a large scale. Some don’t. But it’s a good alternative for journals that don’t have subscribers to give them revenue to try libraries.

By the way, one of the pioneers here … the Stanford Encyclopedia of Philosophy. It’s not a journal; it’s an encyclopedia, but it was Open Access. It’s very large; it’s very successful; it’s very high in quality, but it had a funding problem, and after a while, it realized it needed a new source of funding, and it went to libraries and said, “Our encyclopedia is so good that if it were toll-access, you would have paid for it. You would have bought a subscription because your people use it, and by the way, we know that because we track our traffic. So, instead of changing to a subscription model, what if we just said to you that we can keep this Open Access if you paid us the equivalent of three years’ worth of subscriptions, and then after those three years – or even before, if you make this commitment – the work can be Open Access?” Enough libraries saw the deal there, saw the bargain, that they committed. What those libraries committed was matched by the National Endowment for the Humanities in this country, and the result was an endowment for the Stanford Encyclopedia of Philosophy. So, libraries are willing to chip in to support an Open Access resource – even to chip in to the endowment to support one in perpetuity – because it saves them money on the other end, that they don’t have to pay out through subscription. I hope our libraries have become more flexible in their willingness to consider models like that.

…

The libraries that are willing to do this are willing to give money to Open Access resources that they could have given to subscription journals, or to non-Open Access books. In other words, all library budgets are tight – that’s notoriously true – but a growing number of libraries are willing to [offer] money from their collections budgets to invest in what they see as a superior alternative. I think we can take this as a vote of confidence in the superiority of this alternative, because the money is very tight. It’d be very easy for these libraries to say “No, simply because we’re strapped.” If they said they were completely strapped, everybody would know that’s true; it’s not an excuse. But despite being strapped, they’re finding the money – sometimes within severe limits – to pay for these alternatives, because they want to pay for that better future.

Do you have any advice for younger librarians in talking to senior faculty and staff in their institutions who think they know a lot about Open Access, but may have adopted a few of the widespread fallacies we discussed earlier? 

That’s a good one. It’s been true since the beginning – I think it’s still true – that as a class, librarians know more about Open Access than faculty. I’m not saying that’s a shame. It’s good that librarians know so much, but it’s a shame that faculty don’t. They’re the authors and the readers, for the most part, of this research. They ought to know at least as much as any other category, but they don’t. Sometimes, I try to be charitable in understanding why they don’t. When I’m trying to be charitable, I think it’s because they’re good at what they do. That is, they’re good at keeping their head down and focusing on whatever their research topic is and excluding everything else. That is what they do, in general, and it does make them good at what they do, which means that most of them still don’t know a lot about Open Access. They may think they do, partly because they might think it’s very simple, or that there’s just one or two options open to them, when in fact there are many. So, they have to be enlightened, gradually and delicately.

One fallacy that most publishing scholars have is that by publishing in a journal – especially an eminent journal that’s widely subscribed – they reach all the readers who care to read their work. That’s not true. One of the things I really enjoy doing here is showing faculty authors the traffic numbers on the articles in our repository. They’re always astonished. That is, they’re astonished because they didn’t expect it. They thought there would be zero downloads, because the work has already been published in an important journal. Why would anybody go to the Harvard repository when they can go to the journal? Again, they’re a little oblivious to the fact that not everybody has access to the journal.

So, it helps if the young librarian can go to the senior faculty member with some data and say, “By the way, your article was downloaded four hundred times just last month, and by the way, it was downloaded from sixty different countries!” Now, you only have those data when the article is already on deposit in the repository – so, there’s kind of a chicken-and-egg problem here – but when you can get something in the repository, then you can get a foothold in making this argument with faculty.

When you don’t have that foothold, you can try other things. You can say, “Do you know that we have a repository?” or “Congratulations on your latest publication. Can I help you put a copy of it in our repository?” Then, they might ask, “Why should we do that? What’s the point? It’s already been published. It’s already out there. Everybody who wants it can find it.” And that’s when you can gently say, “Yes, it’s out there – and, again, congratulations – but you can reach more people through the repository, people who don’t have access to the journal. Even if you think they’re few in number and unimportant, let me give it a try.” By the way, they’re not few in a number, and they’re not unimportant, but you don’t have to tell them that at the time. Just get the work into the repository, and let the results speak for themselves.

Many schools adopt Open Access policies by faculty vote because the librarians educated the faculty about the benefits of that. By the way, if your school doesn’t have a policy, and you want one, and you’re a librarian, try this. Eventually, you’ll want to pass the leadership onto faculty. Faculty should take the lead in proposing the policy, and getting it adopted, but librarians can take the step prior to that by educating faculty about the problem to be solved and about the elegance of the solution. You’d be surprised by how many schools have followed exactly that recipe. Librarians deserve a lot of credit for the policies that already exist at institutions.

What do you think the role of the next generation is in advancing Open Access to research, and openness in education generally? If there was one action that people could take this Open Access Week, what would you suggest it be?

I argued in my book, and I’ve argued elsewhere, that generational change is on the side of Open Access. We have everything to gain from it. People who were born and raised with the Internet – digital natives – expect to find everything they need online, and they expect to put all of their own valuable work online. Those two expectations will change the world. They will create an Open Access universe. They will make Open Access the default. They will push against any disincentives to do that, or any obstacles in the way of doing that.

A couple of years ago, it was ironic but true that senior faculty supported Open Access more than junior faculty. Again, it wasn’t because this generational change had no impact; it was because junior faculty felt constrained to avoid it, because they were looking for promotion and tenure. By the way, they still feel constrained to avoid it. Nevertheless, they support it, and when they’re surveyed, they say that they support it. They wish that they were not constrained.

Follow your thoughtful preferences. If you see the point of making your work open, then make it open. If you want to argue with your promotion and tenure committee, or with your department chair, or your dean, do that. If you want to persuade your colleagues to make their work open, do that. These constraints are themselves temporary. These constraints are customs. They don’t have to last forever, and they are made by people, like you, who will be in a position to change them once they have careers. So, one of my pieces of advice for junior faculty and early-career researchers is, first of all, if you have to publish in a non-Open Access journal to get tenure, then do it, because I want you to get tenure. I want you to be in a position to help change some of these customs from the inside. By the way, if you do publish in a subscription-based journal, you can still make a copy of your work Open Access through a repository, and you’d better do that!

You can tell I’m drifting into the second question. What can people do? Or, if I had to give a short list of what they should do, what would be on the list? The first thing is, make your own work Open Access, even if you don’t have time to be an advocate, a crusader. If you don’t have time to write extensively about Open Access itself, at least make your own work open. Again, you can do that through Open Access journals; you can do it through Open Access repositories. Do it one way or the other. Don’t feel that you’re being impure if you publish in a subscription journal and make your work Green Open Access. You’re still making it open. You’re only slowing things down if you don’t make it open at all.

One of the other things you can do is refuse to peer review for journals that are not Open Access. This subtracts your free or donated labor from journals that are working against our interests. Or, conversely – as a rhetorical question – why should you donate your labor to a journal that’s working against you and your colleagues? You can explain this politely. I have a boilerplate letter that I use when I’m asked to referee an article in a non-open journal. You’d be surprised how often you get a friendly response from the journal. The reason you get friendly responses is that the person receiving the letter is an editor, not a publisher. They didn’t set the business model for the journal. They’re a fellow scholar and academic like you, and sometimes they’re unaware that people feel this way, but sometimes they’re sympathetic, and they say “I too support Open Access, but I’m an editor at an important journal, so I’m keeping my job.” But even if you don’t get a friendly response, you can avoid donating your labor to journals and publishers that work against you. It’s kind of painless to withhold your labor. It actually helps you by giving you time that you can devote to other good causes, so do that. By the way, when you do it, if you have a letter that you send to the journal in cases like that, blog it! Spread the word that this is a possibility for other scholars.

The State of Open Access: An Interview with Peter Suber of Harvard University

This is an edited transcript of a webcast featuring Dr. Peter Suber, Director of the Harvard Office for Scholarly Communication and Senior Researcher at the Berkman Klein Center for Internet & Society. The webcast occurred live on October 18, 2016, and would not have been possible without generous sponsorship from the Public Library of Science.

SPARC and the Right to Research Coalition are grateful to Dr. Suber for his time, energy, and expertise.

Interviewer: Joe McArthur (Right to Research Coalition). Transcriber and Editor: Scott St. Louis (Grand Valley State University)

Next February, the Budapest Open Access Initiative (BOAI) will turn fifteen. What do you see as the greatest obstacles to Open Access that the movement has overcome, and the greatest obstacles that it has yet to overcome?

Thanks, Joe; I’ll be happy to dive into that. First, let me just say how happy I am to be here, with one qualification: I wish I were not doing it remotely! I wish I were live in front of the audiences that are tuning in. OpenCon is the biggest of the conferences I’ve never been able to attend in person, so I wish I could be there.

But you’re asking a good question: not just what are the obstacles today, but what are the obstacles we’ve overcome in the past fifteen years? You have to remember that in 2002, when we issued the first statement from the BOAI, Open Access was at least a fringe movement. It was not mainstream, and so one of the first obstacles we overcame – gradually, not quickly – was to make it mainstream. Open Access is clearly now mainstream. Roughly thirty percent of peer-reviewed journals are Open Access. Roughly half of all new research articles become Open Access one way or another within the first two years of their publication. Major universities and major funding agencies require it; major bills requiring Open Access have been introduced in the U.S. Congress and have been adopted in other countries. It’s moved from the periphery to the mainstream. That was big. If you want to go back to how it moved from the periphery to the mainstream, that’s interesting. We can get into that, but it helped that major institutions, and not just fringe activists, endorsed the idea.

At the time, the people brought into Budapest to draft the Initiative were all working on what we now call Open Access, but the question for us was, are we all working on the same thing? Are our efforts convergent? Could we all cooperate? Could we amplify each other’s efforts? Could we unify? Could we make some kind of movement out of this? At the time, the answer was unclear, which is why we had to raise the question. Now, I think the answer is clear. It [became] clear at the meeting: the answer is yes … By the way, Open Access didn’t have a name at that time! I was writing about it, but I called it free online scholarship. We agreed to call it Open Access, so we introduced the term. We introduced what I still think are the two primary strategies for Open Access, namely Green and Gold, or Open Access through repositories and Open Access through journals; those remain the primary channels for Open Access to journal articles, and we got the first significant funding for Open Access from the Open Society Institute, now called Open Society Foundations.

We overcame these obstacles in part because our ideas were good. It was an important moment in the middle of the period between the launch of BOAI and today when publishers hired a very aggressive lobbyist to counter our efforts in Congress, and he was notorious for his aggression, even for his deceptive tactics. But he had to tell his clients – which we know from leaked papers – that the problem of fighting against the Open Access movement is that our arguments are better than the publisher arguments! So, part of what enabled us to become mainstream was that we had good arguments, and even though we were just individuals who didn’t represent important institutions, the arguments persuaded important institutions. We didn’t need institutional buy-in if we got wide enough individual buy-in, but we got the institutional buy-in, which in turn helped persuade individuals … and all of that helped. We have good arguments because Open Access, in my humble opinion, is unambiguously good. I can’t think of a single downside to Open Access, even after all this time.

At the BOAI, those two strategies [Green and Gold Open Access] were really enshrined as the way forward … Was there a third in mind then, or others that didn’t make the cut, so to speak?

No, but the cynical way to read what we did is that there were people in the room who favored Green over Gold, and other people in the room who favored Gold over Green, and we ended up endorsing both because it was the best way to reach consensus.

I don’t take that cynical reading. I endorse both because I endorse both. I actually think they’re compatible and even synergistic, and I’ve been arguing that ever since, but not everybody in the room thought that. So, it was a big-tent kind of document that embraced what at the time were the different factions of a very, very small movement.

But since then, we’ve discovered other channels that could be used, and sometimes are used, for distributing Open Access research. Some of them are very familiar: blogs, wikis, listservs, peer-to-peer networks. They’re all valid channels; that is, what they contain could well be Open Access, and sometimes is Open Access by default. They may not be the best containers for Open Access research, and by the way, journals may not be, either. Repositories can host arbitrary file formats, and so they’re more adaptable. Journals are evolving – maybe we’re already in an era of post-journals – but we should acknowledge that there are far more than two channels.

It’s just that even today, after all this experimentation, most of the OA literature is in one or both of those two channels – repositories and journals – and most of the discussion is about those two. But every now and then there was a serious experiment with some additional channel. There was an attempt to share very large datasets through BitTorrent, because the files were too large to share any other way. There were experiments to post journal articles to wikis, so the version published by the journal froze at the moment of peer review approval – you could call it a quality ratchet – but another version went to a wiki, where the public could revise it, enhance it; and no matter what they did to it – whether they improved it or degraded it – the journal version remained unaltered. That was a nice experiment. It didn’t last; I don’t know of any journal that still does that, but I’d like to encourage more of that experimentation. I think one of the exciting, uncharted frontiers for Open Access is to think about new containers, or structures for containing research, other than the journal article, other than the dataset, other than the conference presentation, other than the dissertation. The digital world gives us the flexibility to try new formats, and even if we came up with one that was very good, we’d have the usual problems of inertia in getting people to adopt it, but we now have a chance to experiment that we didn’t have before.

I think you’ve started with a really excellent overview of what the past fifteen years have looked like, and the strategies that we’ve adopted, and some of the big wins we’ve had as a community, as a movement. Looking back, is there anything you wish had gone differently, or that the movement could have tried to do differently to be more effective?

We’ve made nothing but progress since then. The progress has not been fast as we wanted. It’s not been as fast as we hoped. In my opinion, it’s not been as fast as the opportunities allowed. There were chances to make faster progress that we just for some reason couldn’t seize, but I don’t think we took any seriously wrong turns.

The Open Access movement has grown large, and as it grows large, it contains more internal disagreement. Everybody acknowledges that, and by the way, it goes back to the beginning. The Budapest meeting, which was very small – about a dozen people – contained some of this disagreement already. But the fact that individual Open Access proponents, or even some cultures within the Open Access movement disagree with one another is not a failing; it’s not a mistake. It’s not something we could have avoided, even if we had wanted to avoid it. I think it’s the price to pay for scaling up, and including many people of many different kinds.

You have to remember that Open Access is proceeding at different rates in different disciplines, because the disciplines are themselves different in relevant ways. They have different funding mechanisms, different levels of levels of wealth, different sizes, different customs and practices for distributing research. So, even if we all agreed on the principles of Open Access, we would still have differences arising from our fields, from our disciplinary perspectives. So, I don’t see those internal differences as a mistake that we could have avoided. I think it might have helped if we had anticipated them more, but are there any mistakes? Can anybody allege a mistake that was made? It’s very hard to say, in part because there was no top-down leadership of the Open Access movement, so we can’t point to a history of decisionmaking by them. I include myself there. Obviously, I made my own decisions, but I didn’t speak for everybody.

I think we could have moved faster. I still think we can move faster, but when I think about the reasons why we’re not moving faster, they all seem to be deeply entrenched in institutional practices which are very, very hard to change. I think one thing we could have appreciated at the beginning was that that’s what we’re up against. It looked to most of us at the time as if Open Access were a matter of taking advantage of this cool new technology … and nothing more. We had the Internet. We had authors who were fortunately situated, so they didn’t have to be paid for their work. It looked like if you put the two together, suddenly you had Open Access, and it’s true that in a sense you did, but only when you have willing authors. How do you get willing authors? Authors follow the incentives that exist in their time and place, in their institutions, and those incentives are based on long practices that antedate the Internet. So, it’s cultural change; it’s institutional inertia, much more than technology and economics, that dictates the rate of progress. I don’t think we appreciated that at the time.

If we had known that the long-term prospect was to change the culture of academic publishing, then I think we would have been less surprised by the slow rate of growth. But in light of the fact that it is cultural change that we’re talking about – making deep changes to institutional practices – I think it’s actually gone faster than other kinds of deep institutional change. So, I’ve learned to live with our slow rate of progress, but I still see lots of allies and colleagues who mistake the slowly rising curve of progress for a flat curve – as if we were making no progress – and sometimes even for a regress. The progress has been steadily up. The slope of the curve is up. It always has been up, ever since we started. It’s just not up as steeply as we would all like.

So many questions have been submitted about research evaluation, it’s actually hard to know which specific one to pick! Do you have any reactions about the topic of research evaluation and how we should be focusing on it now?

Promotion and tenure committees are a major obstacle. They are a bottleneck. They are in a position to create good incentives or bad incentives; mostly, until now, they’ve created bad incentives. They are staffed by us! That is, they are staffed by academics, and these are academics who’ve published research. Some of them support Open Access. They see themselves – correctly, I think – as the gatekeepers of quality. They want to hire good people who have written good work, but if you look at the customs that guide their practices, how do they judge whether someone has written good work? They look at the impact factor of the journals in which they’ve published, which is a very bad metric for that, and it’s a scandal that PhDs don’t understand that. It only takes a few minutes to explain it, and when they get past that and say, “Okay, let’s not use impact factor. What other metrics can we use?” The short answer is, read the darn articles! If you’re experts in the field, if you’re evaluating people from your own field, read them! Now, your judgment about them will not necessarily be objective. It will be your judgment, and there will be other judgments by other members of the committee, and all of our disciplines contain internal disagreements, so even people who have read these articles carefully, firsthand, might disagree with each other, but that comes much closer to an actual evaluation of quality than to pretend that an impact metric is a quality metric. Again, that’s a fundamental mistake that smart people should not make.

In the very beginning; that is, when Open Access was brand new – in fact, when digital anything was brand new – promotion and tenure committees didn’t give credit for digital journals of any kind, whether they were Open Access or not. The idea seemed to be that real scholarship isn’t digital; real scholarship is ink on paper, which, of course, is silly! In this case, it’s easy to tell that it’s silly because we have overcome that. We don’t think that way anymore; it only took us a few years, but the idea that it took years instead of minutes is a scandal! It’s like saying that real scholarship is inscribed on sheepskin. It obviously has nothing to do with the quality of research. But it held us back, because smart people were being very conservative about the traditions of their institutions. That’s what I mean by a cultural barrier.

So, sometimes, good people – namely, ourselves – are on these committees enforcing old criteria. But sometimes, they are people who aren’t even clued into the controversies, or aren’t sensitive to the incentives and disincentives that they’re creating. One of the worst incentives which accompanies the impact factor, but isn’t the same thing, is to give career points to people who publish in venerable, old, well-established, high-prestige journals, and either no career points or negative career points for publishing in anything else. Again, taking the prestige of the journal as the metric for quality of articles is a simple mistake; it has nothing to do with the quality of the articles. It’s a way of outsourcing judgments of quality to publishers, which is a mistake. Academics serving on these committees ought to be the judges of quality themselves.

Impact factors, or getting your work accepted into a high-prestige journal, is a kind of endorsement from an important source. But that’s all it is, and it might be that the article that you’re looking at from a candidate might be the worst one in the journal that year, dragging the average down; it might be the best one in the journal that year, dragging the average up. You won’t know unless you read it and apply your own expert standards to the piece. So, I think promotion and tenure committees have to stop using publisher prestige as a surrogate for author quality.

The Public Library of Science was pioneering in looking at article-level metrics as opposed to journal-level metrics; that’s good. At the beginning, most article-level metrics were citation-based (or, they were impact-based). That’s understandable, because we don’t have quality metrics! But we can get closer and closer to quality metrics – and further and further from crude, oversimplifying citation metrics – as we add altmetrics, as we add other measurements to the mix. So, I support the altmetrics movement, but so far I don’t see any of them coming very close to judgments of quality … and to evaluate somebody for hiring, promotion, or tenure, we really have to judge quality, not just impact.

…

Now, after they have read them, and have some estimate of the quality of the article, and some estimate of the quality of the person, they can supplement that with citation metrics. I had a colleague who came up for tenure in the early days of the Open Access movement, when Google was new, and he was trying to give his committee some idea of the influence of his work, so he gave the Google rank of some of his papers. Committees didn’t ask for that then; they still don’t ask for that … He wasn’t pretending that it replaced anybody’s evaluation of his quality, but it was supplementary. So, as long as committees are trying to look at quality, let them also look at all kinds of supplementary impact measurements.

One thing I didn’t add before is that one disincentive created by these committees – especially when they ask for their authors to publish in high-prestige, venerable journals – is that they are steering them away from Open Access journals, regardless of the quality of the Open Access journal. The reason is that, on average, Open Access journals are much younger than non-Open Access journals, and the younger they are, the less time they’ve had to develop prestige in proportion to their quality. So, they could be born excellent, but they take time to acquire a reputation for excellence … [When you ignore evaluations of] quality and jump to prestige, you automatically favor the older journals, regardless of their access policies. That’s a disincentive to publish in Open Access journals. It still exists, and many young scholars feel that they cannot publish in (at least most) Open Access journals, because they don’t meet the standards set by their promotion and tenure committees. That’s not a problem with the journals; it’s a problem with the committees.

For people who are trying to change the tenure and promotion system, do you have any strong feelings about how that could be done, about what the strategies are to do that? It seems like a hard problem to solve, but one that has really come into its time.

It’s a very hard problem. I’ve tried to do it at more than one institution. One reason it’s hard is that promotion and tenure committees tend not to be governed from the top, because the members would resent that; faculty would resent that. The committees make their own decisions, and then those decisions might be advisory to somebody above them, like a president – and that’s when the presidential criteria come into play – but the committee gets to use its own criteria, so they don’t want anybody to tell them what their criteria should be. There’s also a pull-up-the-ladder sense, that “I got my job because I complied with the old standards, and I’m not going to let anybody in who doesn’t comply with the same standards I did.” But at most schools, including the school where I am now, every promotion and tenure committee runs itself. For example, I’m at Harvard, and every department has its own promotion and tenure committee, and they all do it their own way. There’s no uniform set of guidelines for promotion and tenure committees, so even if you could persuade one to change, it’s not the same as persuading them all to change. This might be different at some institutions that are more uniform across their departments and schools – Harvard is notoriously disunified that way – but you can’t change them from the top down, and if you can’t do that, you have to persuade them one at a time, and their membership keeps turning over.

So, practically the only recommendation I have is that if you’re a faculty member, and you’re asked to serve on the committee, or you ever have a chance to volunteer to serve on a committee, take it! Say yes, and be one of those people who can help change it from the inside.

Institutional repositories have always struggled to get researcher buy-in, and now many subject repositories, or preprint repositories, are popping up. What do you think is the future of institutional repositories, and how can we take our experience with institutional repositories into the realm of preprints as we make big pushes for those?

I’m a strong proponent of institutional repositories, and unlike some of my allies, I also support disciplinary repositories … I support them both … They both support Green Open Access, and I support Green Open Access. I don’t support it to the exclusion of Gold, but I support Green as well, and I support any platform, and any policy, and any practice that gives us more Green that we used to have.

Institutional repositories are growing. There are four thousand Open Access repositories around the world, according to the Registry of Open Access Repositories. I don’t know how many of them belong to universities as opposed to other kinds of institutions. It’s a large number. I wouldn’t be surprised if a large fraction of them were experiments that died off and are not active, but the active ones are providing a very good service. At every school that has one, at every school that has an ongoing effort to fill it, they’ll find that some faculty would rather deposit in a disciplinary repository, especially the popular ones, like the physics arXiv at Cornell, PubMed Central, and SSRN. I see no problem with that. Again, the work is being made Open Access. If the institution has its own reasons to want a copy in the institutional repository, they can try to get one. They can either get it from the author, or they can harvest it back from the [disciplinary] repository.

One way that we could make further progress is if all Open Access repositories were better than they are today at permitting this sort of mutual harvesting. Believe it or not, they don’t all cooperate with this. Some do, some don’t. But there are reasons why institutions might want their own copy. All those reasons, in my view, are less weighty than the reason to make the work open in the first place. So, if a work is open through the arXiv, and not through the Harvard repository, and it happens to be by a Harvard author, I feel that the pressure is off; at least it’s Open Access. If we can also get a copy for the Harvard repository, good for us. We can do good things with it, including preserve it in the Harvard Library, which is something the arXiv cannot do. But I’d rather have the work be open through the arXiv than not be open at all.

…

The question seemed to assume that disciplinary repositories grew up after institutional repositories, but they didn’t. The earliest Open Access repositories were disciplinary, not institutional. I’m thinking primarily of the arXiv, which was originally at the Los Alamos National Laboratory. That was launched in 1991, which is ancient in Internet time! There’s also the Department of Education’s Open Access repository for educational work. Both of these were pre-Web. Institutional repositories came later. They are proliferating in numbers, and so are the disciplinary repositories, but if we have to say which came first, I think we’d have to say it was the disciplinary repositories.

It seems like there’s been an increase in the number of preprint repositories, especially in the fallout from SSRN being acquired by Elsevier and things like that. Do you have any opinions on that trend?

I apologize for leaving that out of my previous answer, because it was part of the question. First, institutional repositories – at least, like ours here [at Harvard] – accept preprints. So, it’s not as if you have to go somewhere else to provide Open Access to your preprint. There may be institutional repositories that don’t. We focus on post-prints, like most of them, but we accept preprints. I hope that’s a common practice, and if it is a practice at your institution, I hope the institution tells people about it. We tell our faculty about it.

On the other hand, if you’re in a field like physics, mathematics, or computer science, and you want to share your preprint, I understand perfectly why you’d rather do that through the arXiv than through the Harvard repository. The eyeballs for that content go to the arXiv; they don’t come here. By the way, to find stuff in the Harvard repository, you don’t have to come to Harvard and run a local search. You can run a search on practically any search engine, including Google and Google Scholar, and that’s true with every well-configured repository. But many repositories do receive visitors who run local searches, and the biggest ones – like the arXiv, SSRN, and PubMed Central – do that. I can understand why you’d want to put your work there, where people go to look for it. No problem with that.

I completely support Open Access to preprints, just as I support Open Access to post-prints. There’s a lot of value in it. One interesting question for cultural historians one day will be, why did preprint Open Access take off faster in physics, mathematics, and computer science than it did in other fields? Not just other natural science fields, like chemistry, but also fields in the humanities? The benefits they bring to the researchers are roughly the same in every field, so it’s a nice question.

But one of the benefits is the early timestamp on your work – and this might apply more in the natural sciences than in the humanities – but if you put something in the arXiv, which you can do just a few minutes after you’re ready to do it, you get a public timestamp that minute … [acknowledging] that you made this discovery, or you came up with these findings, perhaps earlier than someone else working on the same problem. For the sake of your career – hiring, promotion, and tenure – that timestamp is more important than the name of the journal in which you eventually publish.

So, preprint archives give the earliest possible timestamp. That’s valuable. They also give the author feedback on their work before they submit a version to a journal for formal publication. That can be valuable. So, you’ve got a draft of an idea, or you’ve got findings from an experiment: you write them up in some preliminary version, post them for comment, that allows you to submit a better version for publication, and that value, that benefit, should apply to the humanities as well as the sciences.

When talking to researchers, they often will only be aware of making their work Open Access by way of paying article processing charges (APCs), and generally consider Open Access to be too expensive for them, unless they work at a very well-resourced institution. Do you have any insights or thoughts on why that route gets so much more attention from researchers than self-archiving and preprints?

It’s not only getting more attention than Green Open Access; it’s getting more attention than no-fee Gold Open Access. Fee-based Gold – the APC-based Gold – is one business model for publishing an Open Access journal. It’s a valid business model. I support it. I administer a fund here at Harvard to pay the fees of journals like that. I don’t think it’s a failure of openness, as some people say. On the other hand, it’s a minority model. Most people think it’s the only way to support an Open Access journal, the only way to get a revenue stream to pay the bills. It’s not true! It’s not even true that it’s the dominant way.

Every estimate for the past ten years or so has shown that the fee-based model only covers about thirty percent of peer-reviewed Open Access journals. Every study shows that a vast majority – roughly seventy percent – charge no fees at all. Some people call these Platinum or Diamond Open Access. I don’t, just because I don’t want to get into terminology wars, and I want the term Gold Open Access to refer to journal-based Open Access, regardless of the journal’s business model. But apart from terminology, we just have to recognize the substance. Most Open Access journals – most peer-reviewed Open Access journals – don’t charge any fees at all.

So, it’s one thing to be deterred by fees, and say “I can’t afford that,” “My institution has no fund for that,” “My grant doesn’t let me pay for that,” or “I have no grant.” But it’s another thing to mistake fee-based Gold Open Access for all Gold Open Access. Again, that’s a failure of information. We can help illuminate that by informing people that the fee-based model is not only just one among many, but it’s the minority model.

Then there’s the fallacy that says Gold Open Access is the only kind, or all Open Access is Gold, and there isn’t even such a thing as Green. That’s also a lack of information. Call it unfamiliarity, call it ignorance. It’s our fault because we have to do better to inform people who are not closely following this movement that we follow very closely. But we do have to inform people that if you publish in a traditional (that is, non-open) journal, much more often than not, you have permission – standing permission from the contract – to put a version of the article in an Open Access repository. Most people don’t know that. People who know it have been saying it, endlessly, for years and years. We’ve been doing our best to spread the word about it. The word has not gotten out very far. Going back to the obstacles that remain even fifteen years from Budapest, we still haven’t taught people that there is this Green alternative next to Gold. I think people who have not been following closely understand journal-based Open Access because they understand journals, or at least they understand them more or less. They don’t understand repository-based Open Access because that’s something new in the landscape. They didn’t exist before the Internet. It wasn’t a serious alternative, and the idea of publishing an article, and then doing something else with it to improve its access, was unknown.

So, we have an education job to do. It’s not as if these options don’t exist. It’s not as if people who don’t have money to pay the fees have no choices, and they just have to suffer. They do have choices. It’s just that the choices are not well known, and our job is to help publicize those options, make them known to everybody.

A lot of the questions that we received were about APCs and corporate publishing models. Are there less traditional corporate publishing models that we should be adopting or moving toward? How can libraries better fit into these new models for Open Access? How do these questions connect to your earlier remarks about containers for Open Access research?

There are several questions in there; let me disentangle them. You asked how libraries could involve themselves. When an Open Access resource comes along, whether we call it a journal or something else, it needs some funding. Its work is not altogether provided by volunteers, even if quite a bit of it is (and by the way, the labor behind a conventional subscription-based journal is also largely provided by volunteers). But it needs some revenue, so many of them turn to libraries, and they say, “You’ve been paying for conventional, peer-reviewed research through journal subscriptions. Why not pay for peer-reviewed research this other way? The advantage of paying for what we do is that the result is Open Access to everybody.”

That’s a pretty good pitch, and many libraries pay for it. I’m inside a library now, and I know that many libraries are not used to paying for this; that is, they know what it’s like to pay a subscription or to buy a book, but they don’t know what it’s like to buy a membership fee, or to provide start-up funding for some good idea. So, they don’t have standing policies for this. They may have the money, but they don’t have a decisionmaking procedure. This is something that we’ll get over. The world is changing; libraries are changing, too. Some libraries are ready to do this, and already do it on a large scale. Some don’t. But it’s a good alternative for journals that don’t have subscribers to give them revenue to try libraries.

By the way, one of the pioneers here … the Stanford Encyclopedia of Philosophy. It’s not a journal; it’s an encyclopedia, but it was Open Access. It’s very large; it’s very successful; it’s very high in quality, but it had a funding problem, and after a while, it realized it needed a new source of funding, and it went to libraries and said, “Our encyclopedia is so good that if it were toll-access, you would have paid for it. You would have bought a subscription because your people use it, and by the way, we know that because we track our traffic. So, instead of changing to a subscription model, what if we just said to you that we can keep this Open Access if you paid us the equivalent of three years’ worth of subscriptions, and then after those three years – or even before, if you make this commitment – the work can be Open Access?” Enough libraries saw the deal there, saw the bargain, that they committed. What those libraries committed was matched by the National Endowment for the Humanities in this country, and the result was an endowment for the Stanford Encyclopedia of Philosophy. So, libraries are willing to chip in to support an Open Access resource – even to chip in to the endowment to support one in perpetuity – because it saves them money on the other end, that they don’t have to pay out through subscription. I hope our libraries have become more flexible in their willingness to consider models like that.

…

The libraries that are willing to do this are willing to give money to Open Access resources that they could have given to subscription journals, or to non-Open Access books. In other words, all library budgets are tight – that’s notoriously true – but a growing number of libraries are willing to [offer] money from their collections budgets to invest in what they see as a superior alternative. I think we can take this as a vote of confidence in the superiority of this alternative, because the money is very tight. It’d be very easy for these libraries to say “No, simply because we’re strapped.” If they said they were completely strapped, everybody would know that’s true; it’s not an excuse. But despite being strapped, they’re finding the money – sometimes within severe limits – to pay for these alternatives, because they want to pay for that better future.

Do you have any advice for younger librarians in talking to senior faculty and staff in their institutions who think they know a lot about Open Access, but may have adopted a few of the widespread fallacies we discussed earlier? 

That’s a good one. It’s been true since the beginning – I think it’s still true – that as a class, librarians know more about Open Access than faculty. I’m not saying that’s a shame. It’s good that librarians know so much, but it’s a shame that faculty don’t. They’re the authors and the readers, for the most part, of this research. They ought to know at least as much as any other category, but they don’t. Sometimes, I try to be charitable in understanding why they don’t. When I’m trying to be charitable, I think it’s because they’re good at what they do. That is, they’re good at keeping their head down and focusing on whatever their research topic is and excluding everything else. That is what they do, in general, and it does make them good at what they do, which means that most of them still don’t know a lot about Open Access. They may think they do, partly because they might think it’s very simple, or that there’s just one or two options open to them, when in fact there are many. So, they have to be enlightened, gradually and delicately.

One fallacy that most publishing scholars have is that by publishing in a journal – especially an eminent journal that’s widely subscribed – they reach all the readers who care to read their work. That’s not true. One of the things I really enjoy doing here is showing faculty authors the traffic numbers on the articles in our repository. They’re always astonished. That is, they’re astonished because they didn’t expect it. They thought there would be zero downloads, because the work has already been published in an important journal. Why would anybody go to the Harvard repository when they can go to the journal? Again, they’re a little oblivious to the fact that not everybody has access to the journal.

So, it helps if the young librarian can go to the senior faculty member with some data and say, “By the way, your article was downloaded four hundred times just last month, and by the way, it was downloaded from sixty different countries!” Now, you only have those data when the article is already on deposit in the repository – so, there’s kind of a chicken-and-egg problem here – but when you can get something in the repository, then you can get a foothold in making this argument with faculty.

When you don’t have that foothold, you can try other things. You can say, “Do you know that we have a repository?” or “Congratulations on your latest publication. Can I help you put a copy of it in our repository?” Then, they might ask, “Why should we do that? What’s the point? It’s already been published. It’s already out there. Everybody who wants it can find it.” And that’s when you can gently say, “Yes, it’s out there – and, again, congratulations – but you can reach more people through the repository, people who don’t have access to the journal. Even if you think they’re few in number and unimportant, let me give it a try.” By the way, they’re not few in a number, and they’re not unimportant, but you don’t have to tell them that at the time. Just get the work into the repository, and let the results speak for themselves.

Many schools adopt Open Access policies by faculty vote because the librarians educated the faculty about the benefits of that. By the way, if your school doesn’t have a policy, and you want one, and you’re a librarian, try this. Eventually, you’ll want to pass the leadership onto faculty. Faculty should take the lead in proposing the policy, and getting it adopted, but librarians can take the step prior to that by educating faculty about the problem to be solved and about the elegance of the solution. You’d be surprised by how many schools have followed exactly that recipe. Librarians deserve a lot of credit for the policies that already exist at institutions.

What do you think the role of the next generation is in advancing Open Access to research, and openness in education generally? If there was one action that people could take this Open Access Week, what would you suggest it be?

I argued in my book, and I’ve argued elsewhere, that generational change is on the side of Open Access. We have everything to gain from it. People who were born and raised with the Internet – digital natives – expect to find everything they need online, and they expect to put all of their own valuable work online. Those two expectations will change the world. They will create an Open Access universe. They will make Open Access the default. They will push against any disincentives to do that, or any obstacles in the way of doing that.

A couple of years ago, it was ironic but true that senior faculty supported Open Access more than junior faculty. Again, it wasn’t because this generational change had no impact; it was because junior faculty felt constrained to avoid it, because they were looking for promotion and tenure. By the way, they still feel constrained to avoid it. Nevertheless, they support it, and when they’re surveyed, they say that they support it. They wish that they were not constrained.

Follow your thoughtful preferences. If you see the point of making your work open, then make it open. If you want to argue with your promotion and tenure committee, or with your department chair, or your dean, do that. If you want to persuade your colleagues to make their work open, do that. These constraints are themselves temporary. These constraints are customs. They don’t have to last forever, and they are made by people, like you, who will be in a position to change them once they have careers. So, one of my pieces of advice for junior faculty and early-career researchers is, first of all, if you have to publish in a non-Open Access journal to get tenure, then do it, because I want you to get tenure. I want you to be in a position to help change some of these customs from the inside. By the way, if you do publish in a subscription-based journal, you can still make a copy of your work Open Access through a repository, and you’d better do that!

You can tell I’m drifting into the second question. What can people do? Or, if I had to give a short list of what they should do, what would be on the list? The first thing is, make your own work Open Access, even if you don’t have time to be an advocate, a crusader. If you don’t have time to write extensively about Open Access itself, at least make your own work open. Again, you can do that through Open Access journals; you can do it through Open Access repositories. Do it one way or the other. Don’t feel that you’re being impure if you publish in a subscription journal and make your work Green Open Access. You’re still making it open. You’re only slowing things down if you don’t make it open at all.

One of the other things you can do is refuse to peer review for journals that are not Open Access. This subtracts your free or donated labor from journals that are working against our interests. Or, conversely – as a rhetorical question – why should you donate your labor to a journal that’s working against you and your colleagues? You can explain this politely. I have a boilerplate letter that I use when I’m asked to referee an article in a non-open journal. You’d be surprised how often you get a friendly response from the journal. The reason you get friendly responses is that the person receiving the letter is an editor, not a publisher. They didn’t set the business model for the journal. They’re a fellow scholar and academic like you, and sometimes they’re unaware that people feel this way, but sometimes they’re sympathetic, and they say “I too support Open Access, but I’m an editor at an important journal, so I’m keeping my job.” But even if you don’t get a friendly response, you can avoid donating your labor to journals and publishers that work against you. It’s kind of painless to withhold your labor. It actually helps you by giving you time that you can devote to other good causes, so do that. By the way, when you do it, if you have a letter that you send to the journal in cases like that, blog it! Spread the word that this is a possibility for other scholars.