Trespassing (Photo credits:

I recently finished reading a long essay by Daniel Allington, a sociologist, linguist, and book historian living in the UK. He’s been following the debates about open access (OA) in the UK quite closely, and has written a well-informed piece detailing the hopes, limitations, and mandates associated with OA. The essay, entitled, “On open access, and why it’s not the answer,” brings a very careful analytical style to the proceedings, something that we encounter too infrequently, I believe.

His conclusion? OA is not the solution, partially because advocates can’t agree on the problem to be solved, partially because the economics of the OA solution shift financing but don’t solve the basic economic problems of science publishing, partially because OA seems far too disruptive for the purported benefits, and partially because the route to accessibility is only slightly dependent on economics but significantly dependent on expertise.

To give you an idea of what this essay is like, here is a brief excerpt about how Green OA and Gold OA appear to Allington:

One of the forms of open access . . . consists in the creation and use of repositories for research writing: databases, typically run by university libraries, into which ‘pre-prints’ (basically, manuscripts) of journal articles may be uploaded for free download by anyone with access to the internet. This has recently become known as ‘green’ open access. For reasons that I shall come to in section 2, I always considered it to be a good idea. However, in itself, it represents a further drain on university budgets (since repositories are not free to run), so it is hard to see how it can facilitate increased expenditure on monographs, unless libraries adopt the policy that where journal articles are available from repositories, journal subscriptions should be cancelled. But such a policy would clearly be unsustainable: journals would close, and the supply of journal articles for upload would dry up. That is presumably why Darnton has advocated more strongly for what is now known as ‘gold’ open access, which keeps journals open by moving the burden of payment from the reader to the writer. Yet as far as the junior scholars for whom Darnton has so much sympathy are concerned, this simply amounts to giving with one hand while taking with the other: it may make it easier for them to publish monographs, but it will certainly make it harder for them to publish journal articles, unless they are wealthy enough to pay for this themselves. Many of them, of course, can barely afford to eat.

Allington writes with clarity and panache, and voices some of the limitations of OA careful readers of this blog will likely have encountered before:

  • Research reports hosted on public web sites are the best solution to the claim that taxpayers have a right to see the results of research they’ve supported; however, requirements for grant recipients to produce these reports are not enforced and the sites hosting what is written are far from comprehensive and difficult to use
  • Most journals are affordable, and citing only extreme examples to drive the conversation isn’t fair
  • OA advocates tend to conflate problems (e.g., library access with subscription prices with domain expertise with taxpayer status), which makes each problem harder to solve or address in a practical way
  • Lack of access has always been a problem, and there are good solutions that are not as disruptive as OA, like authors emailing PDFs to interested people
  • Making scientific content comprehensible for a lay audience takes a lot of time and effort, which goes well beyond mere access issues
  • Price increases for libraries have been driven mainly by the amount of content being generated by scientists, raising the possibility that rather than price increases being too great, the real problem is that library budgets may be unrealistically low
  • Solutions based on technology alone invite the possibility that a single, large, for-profit technology provider will replace a diverse mix of publishers, many of which are society-based or non-profit
  • Gold OA will likely only work for academics at the richest institutions, creating closed access further upstream, a situation that may be harder to repair later and which may be more damaging to young academics than any subscription barriers could possibly be

One reason why these arguments may not be new to our readership is that four Scholarly Kitchen posts are cited by Allington.

Allington makes some other points that made me want to stand up and shout in recognition:

  • The demonization of publishers has not helped the academic community, which needs publishers to give them the journals they rely on
  • Authors are not producing work for publishers, but for other academics; publishers produce finished works for academics; portraying academics as “unpaid labor” for publishers gets the relationship wrong, as publishers are in fact paid labor for academics, who are the ultimate consumers
  • Careers in publishing are getting harder, especially in editorial roles, which is leading to fewer young professionals pursuing these paths, bad news for the future of high-quality scientific communication

His assessment of Gold OA is particularly penetrating, as captured to some extent by this selection:

. . . it’s hard to see how this [Gold OA] will further the cause of public communication of knowledge, espoused by Bell and Fuller and given lip service at least by many open access advocates: the pay-to-say system was devised in order to permit elite academics to continue publishing in the manner to which they had become accustomed, they will be under no obligation to write in a manner more accessible to an audience of non-specialists, and their publishers will be paid in advance even if no-one ever so much as downloads the articles they turn out. Willetts will get what he wants, as will Monbiot, but, as noted above, what they want (i.e. the free online dissemination of research findings) could have been achieved by less disruptive means. Ironically, no money will be saved by the public purse unless the system is shrunk and less research is published.

There are important insights into academic attitudes on offer, as well, such as the ones in this passage about free content on his institutional repository:

Because I don’t want pre-publication copies of my work floating around on the internet (referencing such versions is problematic, and even though I work in disciplines that ignore citation metrics, unquotable articles are less useful to their potential users than quotable ones are), what I have uploaded to ‘my’ institutional repository are electronic copies of the actual published articles, which means that the would-be reader cannot simply click on a button for a download, but must submit a request (by filling out a web form) that I in turn must approve (by clicking a button) – this being legally equivalent to said reader’s phoning me up and asking for an off-print, which I subsequently choose to drop in the post. This in turn has had the interesting side effect of keeping me informed as to who is getting the free copies. Somewhat predictably, it turns out to be the same sorts of people to whom an academic of a previous generation would have been mailing off-prints: postgraduate students, other academics, and the very occasional independent scholar. . . . on the face of it, it seems quite unlikely that a person unwilling to make the effort to fill out such a short form (it has only five compulsory fields, one of which is a dropdown menu and one of which is a CAPTCHA challenge) will be willing to plough through several thousand words of densely-written academic prose, unleavened even by the sort of weak humour in which an academic blogger may indulge from time to time.

I’m going to stop quoting Allington with that. He has written a long, thoughtful essay about OA in the UK that merits careful reading and reflection. He’s well-informed, fair, and considered in his approach — things we should all aspire to.

The comments are worthwhile, as well, such as this selection from the first comment:

. . . my final thought is that recent things I have been hearing suggest that publishers are not anticipating huge changes to their income streams any time soon, which makes me wonder what the point is . . . except perhaps to further privilege the dominance of big funders/ research councils in setting the agenda and reinforcing such hierarchies within universities as well.

Statements like “this is the best thing I’ve ever read about open access” are consistently found in the comments. Let this post add to that praise. Well done, Daniel Allington.

Kent Anderson

Kent Anderson

Kent Anderson is the CEO of RedLink and RedLink Network, a past-President of SSP, and the founder of the Scholarly Kitchen. He has worked as Publisher at AAAS/Science, CEO/Publisher of JBJS, Inc., a publishing executive at the Massachusetts Medical Society, Publishing Director of the New England Journal of Medicine, and Director of Medical Journals at the American Academy of Pediatrics. Opinions on social media or blogs are his own.

View All Posts by Kent Anderson


88 Thoughts on "Not the Answer — An Academic Carefully Assesses the Arguments for Open Access"

Thank you so much for this write-up of my essay, and for your very kind words. I feel I should emphasise that I’m far from being an expert in these matters: I’m just someone who’s concerned that we may be about to remodel the entire academic publishing system with no clear understanding of why we’re doing it in one particular way rather than another or of what the likely consequences will be. There are many problems in academic publishing right now: nobody doubts that. Maybe some solutions currently described as Open Access make more sense than others do. Maybe some of those solutions make more sense in some areas than they do in others. And there is a whole set of problems that Open Access solutions just don’t address. But discussing these things requires us to think in terms of specific solutions to specific problems, and it’s very hard to do that while caught up in highly emotive arguments ‘for’ and ‘against’ a seemingly monolithic Open Access.

What’s happened in the UK ought to be a clear warning to the rest of the world: a government minister, swayed by arguments for something called ‘Open Access’, has implemented an Open Access policy that creates many more problems than it solves (including for ‘green’ Open Access: the parliamentary Business, Innovation, and Skills Committee has warned that repositories are likely to suffer as a result).

I quite agree that there is enormous confusion with these issues and well done for sorting them out some while making that point. But I wonder if the US decision to limit it’s mandate to a delayed access subscription system makes the policy debate rather moot?


I’m reading your blog post, and finding it very interesting BUT: Grey text on a black background? I don’t think there’s any way to read this all in one sitting without squinting my eyes right out of my head. Yeurgh.

I recently found out that the OA movement in Latin America is long established and most OA journals there do not charge article processing charges. My interactions with OA advocates from Latin America leads me to believe that something big is going on there. I recently wrote about this in a post on BMJ Blogs ( OA is not confined to the Anglosphere and the traditional geographies of scholarly publishing, so I wonder why arguments about OA are set out and concluded in this context without even a mention of OA elsewhere.

Ravi, the focus of the discussion is the subscription journal industry so free, no APC journals do not play into it. They might if people proposed that the paid for journals be forced to adopt the free, no APC model. That would be dramatic indeed. The question is how are these free journals paid for?

Thanks Richard. I think many people would not like to see governments directly funding the existing subscription journals. Moreover, given the international character of many it is not clear how this could be done. On the society side the journal is presently the chief membership benefit. I only belong to AAAS in order to get Sciencemag. Having the members pay for a journal that is free to all is an interesting model indeed. But it is not something that governments could mandate.

Earlier this year, I had a two-hour conversation with the librarian in charge of biomedical SciELO in Chile, and she was the first to admit subpar online publishing standards. Of course, the reason was underfunding for the project. SciELO was in the upfront once, but it isn’t there anymore.

How do you fund decent, professional, innovative fully OA journals? If access to knowledge is to be considered a public good, then public concern must be present, meaning that governments will have to engage, especially if trying to jumpstart science in their countries. On the other hand, social entrepreneurs will also have to step in (my case) and devise ways of keeping a journal going. Creativity and commitment here are the key words. And on top of this, learned societies and academics will also have to continue doing what they’ve always done – donate their personal time for the good of science.

If we can bring together all of these parties to work together, to collaborate, we may find a way out. But paywalls and for-profit ventures are certainly not the way, at least not for regions like mine.

So it’s not a case of “for” or “against” OA. It’s a case of how you are able, as a society, to disseminate and expand knowledge without paying the price of lowering standards while you’re at it. Set your priorities, assign your budgets, be creative and aim for the stars.

I think it’s also important to pull at the reasoning that denies “for-profit” ventures a chance to compete and best fulfill the needs of the research community. Some of the most successful leaders in driving OA are for-profit companies, like Springer (BMC), and some of the more interesting experiments are being done by for-profit venture capitalists like PeerJ.

It’s also a misnomer to think that not-for-profits can survive by running at a break-even level. In order to grow, experiment and weather hard times, a not-for-profit must run at a profit, to generate funds. As noted elsewhere in these comments, it’s why a not-for-profit like PLOS needs to be profitable and build substantial savings to continue to grow.

I agree that not-for-profits must generate funds and accrue savings in order to be effective publishers. That alone is good reason to say the government should not be in the publishing business. On the other hand, when we know that some for-profit publishers are returning 30-35% to investors, I think only not-for-profits can possibly offer fair prices, and only the government offers capital at fair rates!

Thanks for mentioning PeerJ — it will be interesting to see how that develops.

OA strictly defined as Gold or Green probably isn’t the answer to the question of how best to reduce or eliminate the barriers to accessing academic work. Telling people that they probably couldn’t make good use of it anyway isn’t helping either. Simply shifting costs is like rearranging the deck chairs on a sinking ship. Yes, we have to get the problem statement right before we jump on our horse and ride off in all directions.
So, is there is a central problem with scholarly publishing? If so, what is it?
I’d say that the central problem is that the status quo is unnecessarily inefficient making the cost/benefit ratio of scholarly publishing far higher than it needs to be. That cost is the root cause of the barriers OA advocated complain about.

Correction: “… cost/benefit ratio far worse than it needs to be.”

There is always a tendency to point to a cost problem, but costs have actually risen fairly little if you look at the right costs (cost to access content, not print subscription costs). The macro cost issue is based on the volume of published research, which is where the true growth has come. Even at lower costs, a much higher volume costs more. All the while, library and departmental budgets for materials acquisition have decreased as a percentage of academic budgets. So, we have growth in supply (more scientists producing more papers) and demand (more scientists reading more papers), but not in funding. Could that be the cause of the crisis? Inadequate funding?

The obvious trouble with the claim that whatever “crisis” we are suffering from is really a crisis of not throwing more and more public funding to publishers is that the suggested solution removes and sort of upper limit on the amount of that more and more public funding being thrown to publishers. it could well be that the world would be a better place if publishers had an unlimited supply of money to work with, but it isn’t obvious.

You’re making the same mistake as noted above, which is that publishers are anything more than hired hands for academics. (We are, but our pay differentiates somewhat along the lines of how well we do what we do, like all trades.) In any event, “throwing money” at researchers in the form of grants, for the training of PhDs, for lab equipment, all to generate studies and research outputs and more scientists, and then to not fund their access to information? That’s where the gap is.

We’re not talking about an “unlimited supply,” but it seems clear that the amount of funding libraries get is unrealistically low given all the pressures to publish and desire to read more peer-reviewed research reports.

I think it’s also important to recognize that there’s a lot more than just “public funding” that goes to support journals. From what I can tell (and librarians, feel free to chime in), library budgets come from a variety of sources, tuition, student fees and in some cases grant overheads (some portion of which comes from “public funds”). But journals have revenue streams beyond subscriptions, including licensing revenue and advertising revenue. So it’s not accurate to think solely in terms of “public funding”, and if one is trying to drive a shift to a Gold OA world where 100% of journal support comes from funding agencies, particularly public funding agencies, that may be a larger hill to climb than imagined.

In support of Ravi’s comment: the journals from Bangladesh, Nepal and Sri Lanka which are hosted on the INASP Journals Online websites ( have been OA (free to read) since the inception of these sites in 2007. The 280 journals on these sites are supported by their Universities or professional societies and the cost of publishing is small. Most of them had relatively few paid subscriptions anyway, so their move to OA had little impact but allowed their research to be accessed by a global audience – there have been 17.7 million downloads of the articles from these sites. The research from these countries is now part of the research communication cycle through OA.

Sioux, as I asked Ravi, are you proposing that the six billion dollar a year subscription industry transition to this free model, forced by government? That the universities take over the journals from the commercial publishers and the societies make their journals free without reducing dues? It is hard to see this as realistic.

David – no, there is no proposal that the entire subscription industry does anything in particular. Rather, this is an observation of how things have been moving in the global south which we at INASP certainly hope will make an increasing contribution to global research. We thought this information was worth sharing in a debate that tends to focus very much on conditions and positions in the north.

Thanks Sue. I agree it is important but this thread is about government action, especially in the UK. Nor is it clear that free journals make a greater contribution to global research than subscription journals. It probably takes serious money to build a top ranked journal in any field. looks interesting. I may do an article on it.

Why don’t you think there’s more of a movement among academic publishers to lobby for increased library budgets? Libraries publish the vast majority of journal subscriptions.

Frank, that the status quo is “unnecessarily inefficient” is a very strong claim with little obvious empirical support. I have seen no evidence for it. So it is not clearly a problem, and not one to be solved by government mandate in any case.

It would be interesting to know the particulars of this claim about inefficiency. As one who worked at a university press before computers were on everyone’s desk, and who witnessed the photocopier as a technological revolution, and later headed a university press as it made the transition from print to electronic in journal publishing, I can cite any number of changes, technological and otherwise, that made publishing a far more efficient industry than it had been before. On the book side, it is those very efficiencies that explain why the prices for monographs have not skyrocketed as the sales declined from an average of over 3,000 in the 1960s to fewer than 500 after 2000.

Not having the time to read Arlington’s “long, thoughtful essay about OA in the UK”, which I note is an OA blog post, I would like to challenge some of the misconceptions quoted in this much shorter review (also an OA blog post). Firstly, Arlington’s assumption that “repositories are not free to run” is unfounded. Many institutional repositories (e.g. ePrints) are open source and quite free. Even if you consider the staff time involved in setting up and running such a system, those staff members are normally already employed by the institution before acquiring the repository. Secondly, the green open access movement is not suggesting that journal subscriptions should be cancelled. The whole point of going green is that authors can publish wherever they like, and then upload their accepted version to the institutional repository for all to access (notwithstanding embargoes and other restrictions – even when there are restrictions, as Arlington himself points out, a copy can be requested anyway from the author). On his argument that “many journals are affordable”, try saying that to African or Asian countries that struggle to purchase access to the most basic of journal collections.

Arlington’s point that “Making scientific content comprehensible for a lay audience takes a lot of time and effort” is quite correct, but well worth the effort, if the academy is at all interested in having an impact beyond itself. It also encourages transparency of output for taxpayer funded research. How gold OA creates “closed access further upstream” is beyond my comprehension. Institutions like Electronic Information for Libraries in the Ukraine are doing some wonderful work in generating open access publishing grants for research institutions in developing countries. The alternative is that they get published for “free” in commercial journals which their own institution cannot afford to subscribe to.

Arlington states that referencing and citation metrics are problematic with pre-prints and versions other than the officially published version. But Arlington is only thinking about traditional referencing and metrics here. The new altmetrics will count online engagement (downloads, social media interactions, etc.) for any version of any kind of research output, including non-traditional outputs such as blogs, research data, presentations, videos, programming code, and news media coverage. The point is that scholarly communication, as it has always done, is using whatever technologies and systems are available to it for communicating its message. And today this includes social media, which is possibly the most open way ever of communicating research findings to the widest possible audience.

The real question is not about the business model of publishers (commercial vs gold OA) or even about the need for institutions to collect their publications in a repository (which has many benefits beyond open access, including the facilitation of internal and external reporting on research output activity). No, the real question is whether or not a scholarly communication model that has been largely monopolized by multinational corporations from whom shareholders expect healthy dividends is sustainable, given the increasing financial difficulties most higher education institutions find themselves in, and whether or not such a model is consistent with the ultimate goal of research, to make the world a better place. An open access peer-review and publishing model may well be at least part of the answer.

“Firstly, Arlington’s assumption that “repositories are not free to run” is unfounded. Many institutional repositories (e.g. ePrints) are open source and quite free. Even if you consider the staff time involved in setting up and running such a system, those staff members are normally already employed by the institution before acquiring the repository.”

And all the development time, technical support and server space is free too, is it?

No, it is well-founded. The “preprint” repository that’s been around longest (arXiv) has had consistent financial woes, and has been tossed between government and academia a couple of times because of it. Studies consistently show that the technology, bandwidth, maintenance, management, and upgrading costs are significant, and steal money from library budgets.

The logic that “online is free or cheap” undergirds many of the assumptions OA activists have lived by for years. The logic is superficial and flawed, while there is ample evidence that it is also not based in actual fact.

It is true that arXive has had financial woes, but when Cornell asked for help a couple of years ago, one foundation and 170 research institutions stepped up to help support it. Whether or not it gives them the most bang for the buck, obviously 170 academic administrators believe it does or will. Similarly, whether institutional repositories are efficient or not, nearly every college in the country decided it had to have one, so now they do. It is true that continuing expenses not only takes away from library acquisitions budgets, but from staffing of library services. Administrators seem to think that is a fair trade, so far, and they won’t reverse those decisions just because you or I don’t approve.

The Association for Research Libraries banded together years ago to create SPARC
because they believe that collaboration and innovation, not larger library budgets are the best way forward. At the Charleston conference last week, the program called “What Provosts Want Librarians to Know (James J. O’Donnell, Georgetown University; Beth Paul, Stetson University; Jeanine Stewart, McDaniel College; John Vaughn, Association of American Universities)” was all about innovation in the “changing information landscape”. When asked about library budgets and increasing costs of library resources, the answers were all about more collaboration and more innovation, not more money.

Are they doing the right thing in the long run? Perhaps not, but if academic institutions are your primary market, it would be a good idea to recognize that there already is a changing information landscape, and try to understand it, rather than pooh pooh it. There is still a place for subscription journals in that landscape and there is no logical reason to think it will disappear because of green OA. On the other hand, it might contract some, and the fact that there is more research being done, may be irrelevant.

Saying “the logic that “online is free or cheap” undergirds many of the assumptions OA activists have lived by for years” only insults people who otherwise might want to help you understand what is changing and how. In my experience, the only professional on my campus who ever said anything like that in public was a Dean who was not at all interested in OA. He thought it unnecessary. That was in 2005. Now he is back in the classroom, where the only thing he needs to do online is use Blackboard.

P.S. Daniel Allington’s article was indeed very interesting, thanks for linking to it!

It depends on what definition of “free” you are using. If “free” means that institutional repositories are run as part of the normal IT workflows of the university, then yes, many of them are free! These are the very kinds of things our good IT and systems folk are expected to support as part of their normal workload.

If you are paying staff to devote paid, working hours to a project, then no, the project cannot be considered “free”. You’re paying for staff time and certainly paying opportunity costs–what other things could that staff be doing during that time? Will you have to hire additional staff if new projects come along requiring the staff time that’s currently devoted to your repository? What happens if your server goes down after hours or during a holiday? How many backups of your repository are you creating and where are you storing them?

Running a repository goes beyond merely providing server space. A good repository needs curation, it needs organization and it needs discovery tools. What submission system are you using? Who is checking the articles that are deposited to make sure they’re compliant with your repository’s rules? Who is constantly checking articles to make sure they’re legitimate articles and not spam or other unsavory types of content? Who is monitoring the content to watch out for illegal content and who responds to legal notices of copyright violation and the like? If you’re talking about something that’s required by a university or a funding agency, you have to devote time and effort to checking and enforcing compliance. All of these things (and many more) require time and effort from trained staff members, which costs money.

There’s no free lunch here.

By this definition of “free,” then anything that gets to be part of the normal workflow of an institution somehow magically doesn’t cost anything. So, for instance, when Amherst College decides to set up a university press from within the library, this becomes “free” once it gets to be part of the library’s normal responsibilities? I suppose that all athletic teams, as part of the normal operations of a college, also are run for “free” because they are just what a college normally does, right? Really, you should listen to what you are saying here. It has ludicrous consequences.

So, first, some points of order. You admit not having read the entire essay, and you misquote Allington’s name routinely. (It’s Allington, not Arlington.) Let’s contemplate this a moment before taking on your other arguments. You did not do your research, and you were not able to accurately translate the name of the author you quote. From an academic point of view, both of these are not very commendable upfront indications.

Nevertheless, you then espouse a whole set of unrealistic claims and directions, which only goes to once again show that OA is far too broad in its aims and scope, and far too confused internally about its goals.

The Green OA movement may not have as its aim the cancellation of subscriptions, but it is a clear side-effect, and one that many Green OA advocates are fine with, at least in their public declarations. Repositories are not free to run. I dare you to start one that has robust content, a useful search interface, and a steady stream of qualified content, and then tell me how it went. Many journals are affordable for the audiences they’re intended to serve. You can claim that anything is overpriced if you play the game you’re playing here — lab equipment is overpriced compared to a toaster, but they are very different things aimed at very different markets. Journals are reasonably priced (really, $75-$200 per year are common prices, or tied to membership dues) in most situations. As Allington states, using extreme examples isn’t fair.

Creating lay content is difficult and expensive work. A lot of companies do it well. Why should the academy adopt this as a role?

Gold OA creates access problems further upstream because young academics or those without access to a large grant or university budget can’t publish. You can’t access research that isn’t published in the first place. This could derail a lot of young academics early in their careers, while also suppressing a lot of content. That’s what it means.

Altmetrics have a lot to prove. They are so far mainly about counting sharing events, and not necessarily meaningful sharing events. They are not widely accepted as valid substitutes or additions to current metrics. Please keep this real.

Most scholarly publishers are small and non-profit. In fact, 81% of them have less than $25 million in annual revenues. The market is not overrun with large publishers, but they are poised to win big in the OA world, which favors scale. If you think OA provides a route to solving the market concentration problem, you are dead wrong, and pretty much playing into the hands of the large publishers.

Your defensive position betrays your bias as a commercial journal publisher. Whether the author’s name is Allington or Anderson, you guys need to get with the program. OA is reaching critical mass and is here to stay. Research institutions are interested in creating and communicating knowledge to the widest possible audience, to achieve the widest possible good. Green open access is the best way of achieving this goal and represents a win-win solution for both authors and publishers. Publishers who consistently refuse to include green-friendly clauses in author agreements will find that authors will take their business elsewhere, due to a growing number of institutional and funding agency open access mandates. I think the fact that the OA movement is not just championed by a few enthusiastic individuals any more, and has taken on a much more systematic approach across the research sector, has spooked publishers. But there is no need to be afraid. All of us want the same goal, which is for good research to reach as wide an audience as possible.

Also, your assertion that altmetrics “are not widely accepted as valid substitutes or additions to current metrics” is complete conjecture. In particular, early career researchers are using them as measures of public impact to compliment traditional metrics of scholarly impact with great success in applications for grants and tenure, as well as for performance reviews. We are in the 21st century now, the century in which shared information and the impact it enjoys has taken on a much more interactive and democratic approach than ever before.

But this is merely an opinion on a blog post, as I’m sure Allington’s is on his.

Again, poor research on your part. I’m not a commercial journal publisher, but a non-profit publisher. I’ve been a non-profit publisher for nearly 20 years.

OA is about 3-10% of the market, depending on which estimate you believe. It has not achieved critical mass. You are exaggerating. It may be here to stay, but seems to be part of the system, not THE system.

OA is being held up to the light with a great deal of caution now. The OSTP memo gave it no funding. The UK is now quite out of step. Green OA threatens the very industry it needs to survive, so there is a vicious cycle at its heart — the better it works, the more at-risk it is.

My point on altmetrics is not “conjecture” but rather quite factual.

I apologize for my assumption that you were a commercial journal publisher. I am a great supporter of not-for-profit publishing. You are, nonetheless, a publisher! Neither of us have provided any evidence to support our statements here. We are both guilty of stating our heartfelt opinions. I believe personal opinions are significant in transitional periods of history, and I also believe that scholarly communication is currently going through such a period, with the advent of the internet and the rise of social media, with more available channels than ever before for researchers to share their data and findings with the world.

I do not believe your statistics of OA being 3-10% have taken into account green OA, with accepted author versions of papers to non-OA papers made available to the public. If they were taken into account, I suggest the percentage of OA papers would be considerably higher. Besides, there are many more ways for researchers to openly share findings now beyond the published paper, for example research data sharing platforms, blogs and social networking sites. It’s the content scholars are interested in, not the container. There is no “vicious cycle” here. Simply an aspirational desire to widely communicate knowledge. This is the universe I prefer living in!

Well, it won’t surprise you that I disagree about your notion that it’s “the content scholars are interested in, not the container.” Most scholars use brands as proxies for quality, and also prefer the formatted, final version of papers over manuscript versions. Containers matter a great deal. Even OA publishers know this. Look at the millions PLOS spent establishing their brand, and the splashy brand launches of PeerJ and eLife, as well as their care with site design.

There is a vicious cycle, which Joe Esposito recently described here: Essentially, the more Green OA succeeds, the fewer journals will exist, which means the fewer papers there will be for Green OA to work from. Vicious cycle.

Your “apology” is insincere, as you undercut it by calling me with obvious disdain “a publisher!” as if this makes me unable to process evidence or think logically about market realities or trends in readership, usage, or “publishing!” You’re reading this on a blog I started and ran for 5.5 years, one which has become a daily touchstone for many people in scientific and scholarly publishing because it is a fount of evidence, clear thinking, and challenging discourse. Pretty good for a “publisher!” Oh, did I mention I’m also an “editor!”, “peer reviewer!”, and “author!”

Best of luck in that universe of yours.

Touché! One would think you an advocate for the free and open transfer of knowledge!! For the record, I am an open access agnostic. As someone else has mooted in this discussion, it’s all about diversity. No disdain intended. I like (most) publishers. On the whole, publishers are cool (as are most editors). I just wish some of them would be a little more open-minded sometimes. Green will not kill journals. If anything, green OA supports the scholarly journal market, as it allows authors to publish wherever they like. I do agree that mandates will have the effect of pushing authors towards green-friendly publishers. I also agree that authors prefer the version of record.

But instead of spending so much energy trying to convince higher education institutions that green is the devil, a more balanced approach might also encourage publishers to support green. In cooperative game theory, players can coordinate their strategies and share the payoff.

You really don’t get it. Green OA is competitive with journal subscriptions, so there is no “cooperative game theory” at work here. It’s pure win-lose.

And with that, I bid you adieu.

Reblogged this on and commented:
Very interesting…wanted to share with INANE members. Leslie

Having just attended one of the most intensive and informative OA workshops I’ve ever participated in with David Hoole and Della Sar of Nature and Peter Binfield of PeerJ I can say I’m convinced major publishers ARE anticipating significant shifts in income streams on a fairly short time horizon given the dramatic growth in sheer numbers of OA articles. For those planning and creating new products and service models and new revenue streams it looks to be an exciting time in publishing.

I think Daniel’s point was more about publishers not seeing their income streams diminish much in the foreseeable future. Just look at PLOS, which has gone from a $9 million grant a decade ago to a $34 million company now, with a 20% profit margin. Springer acquired BioMed Central, and it’s throwing off cash. The list goes on.

There is also the problem of still separating OA exaggeration from OA reality. It’s not clear that PeerJ will survive, and it’s not clear that any but the largest (e.g., Nature) will benefit. Note that Allington said MAJOR publishers. OA rewards scale. With 81% of the publishers having less than $25 million in revenues, a large swath of the market is at risk. The MAJOR publishers are poised to do well. What about all the smaller, non-profit publishers aligned with professional societies?

The discussion above illustrates very nicely Daniel’s point about OA being a sea of confusion. It is a mush of inconsistent concepts and proposals with Poorly defined goals.

But then social movements typically support strange bedfellows. I count at least seven different basic kinds of OA, all inconsistent with one another, with in most cases a range of inconsistent subtypes within each kind as well. What a mess! No wonder there is so much confusion. Yet funders and governments are moving forward, but in all directions.

I see a parallel between the many types of OA and the many distributions of Linux. When there is more freedom to go in new directions, perhaps there are more varieties. And maybe they can all thrive. In the Linux world two of the major distributions – RedHat and Ubuntu – are technically different. They both have their proponents and detractors. But the bottomline is the user has choice. The diversity in the Linux world is often confusing to people used to Windows and Apple. Maybe there’s an analogy here.

But we are talking about funder and government policies, not software systems. Are you suggesting that researchers choose where to get money and/or publish based on the funder’s mandate? And how are journals supposed to comply with many different inconsistent mandates? In general inconsistency is a bad policy at the policy level. It is not good at the software level for that matter.

I don’t know enough to talk about funders, governments, and their policies. My point is about diversity. I don’t think we live in a world of single paradigms. At least, multiple paradigms seem to be emerging. I would like to understand what these are in scholarly publishing, and I’m troubled when I see commentaries on open access that don’t mention anything about what’s happening outside Europe or North America.

There is plenty of diversity already. This discussion is mostly about forced restructuring of the publishing industry, especially in the US, Britain and the EU. If Brazil or Bangledesh are doing something by force we would like to hear about it. To my knowledge they are not.

You know I just get plain frustrated at people who think know something, but are just plain incorrect!! The author of this blog post and owner of the blog (which I am now seriously considering unsubscribing from) has consistently argued against me that “Green OA is competitive with journal subscriptions”. As an academic librarian, I can tell you and the world that you are factually incorrect and just plain wrong!! As an academic institution, we would never ever stop subscribing to a single title just because there is an OA copy on some institutional repository somewhere. You are just wrong. The fact that you have now censored me by preventing me from continuing the online discussion with you, and the fact that you will probably never let this post see the light of day, just goes to show what an information dictator you in fact are. Good riddance.

Kent Anderson is not “the owner of the blog”. It is run through the good graces of the Society for Scholarly Publishing, though with editorial independence from the society and Kent is a contributing author, not the Editor in Chief, nor the “owner”.

I think there’s a continuing failure to communicate that seems to happen between different factions regarding Green OA. The notion seems to be that if Green OA remains in its current state, where it’s somewhat randomly practiced and inconsistent, offering poor versions of articles sometimes in hard to find places, then it should not be seen as a threat to subscriptions. That may be the case at the moment, but funding agency policies, like that of the US OSTP and the Australian Research Council are deliberate mechanisms to make Green OA organized, regular and rapid. These policies call for freely available versions of articles to consistently appear in easy to find places. If they are successful, then most librarians will have to factor this into their purchasing decisions–if you can get much of what the journal offers for free, why would you continue to pay for it? The real crux of the issue is going to come down to embargo periods, and setting them at appropriate lengths to help balance that decision toward continuing your subscription. We’ll have to see how it plays out in practice.

And no censorship here, and Kent is not the person in charge of moderating the comments, that would be me. I tend to edit to remove harsh language (you’ll note I edited down one of Kent’s comments below) in order to keep our comments section a fairly civil place to have a discussion.

Thank you David. That is very decent of you. “[I]f you can get much of what the journal offers for free, why would you continue to pay for it?” The answer is that having an author’s accepted copy does not actually provide us with what the journal offers: quality editing and peer-review; indexing in subject databases; provenance or scholarly copy of record; digital preservation; facilitating standard citation practices; better formatting; academic credit; copyright control; and so on. On the other hand, providing a green OA version enables: open access to an institution’s scholarly output; satisfies funding body open access mandates; broadens potential readership to to the public (including the media); enables collaboration with research partners who may not have access to the journal; makes the knowledge available to countries and regions that cannot afford access to the journal. Surely there is room for both to coexist in the brave new world of Web 2.0.

It may very much depend on the journal and the subject area–each field has different needs, and for some journals, which may be more peripheral to your local researchers, the author’s accepted manuscript may suffice.

But that said, one of the proposed solutions to meet the US requirement is a technology called CHORUS ( which would provide the green copy of the article in the journal itself. It’s a system that’s designed to alleviate the burden placed on the researcher to provide the green copy, to offer permanent archiving, increased discovery, and really to automate the whole process, ensuring compliance both by the researcher and the publisher.

This would alleviate some of the needs you list above. Some publishers are going as far as using the version of record (VoR) as the publicly available version via CHORUS. Regardless of whether the VoR is used or the Author’s Accepted Manuscript (AAM), the green version would have completed peer review, and because it’s presented in the journal alongside the html and pdf versions of the VoR, it won’t lack for provenance, proper citation information, academic credit and should have no effect on copyright control as the publicly available version will still be subject to the same copyright rules as the journal’s other articles. Part of the CHORUS system is to provide several separate archives for preservation of all articles.

Given all that, if this solution is the one chosen by major funding agencies, I don’t think it’s unreasonable to think that librarians are going to consider what’s now offered as they make their difficult subscription decisions. Do we really need to subscribe to the full version of this journal that we rarely use, or can we get by with the AAM?

And I think that’s where the careful, evidence-based setting of embargo periods is going to be hugely important. Can we get by with information that’s a year old? Two years old? Three? Likely the answer will vary quite a bit depending on the field. But I think if this is done carefully and properly, then a fair balance can be struck that will keep journal publishing a sustainable business, and as you suggest, there’s room for both to coexist.

(full disclosure, I’m a recent addition to the CHORUS board)

Indeed David, the argument that green OA is valuable but harmless is self contradictory. Joe has pointed out that green is only harmless when it does not work very well. Suppose every major funder mandates a 12 month embargo. Then every journal is now selling just 12 months worth of access. This is a major devaluation. Basic economics says sales have to go down, maybe a lot in some disciplines, depending on the time profile of usage.

Factually incorrect? What does that mean? Another librarian, who is one of the main bloggers, Rick Anderson, has said that his library would take into consideration in cancellation decisions whether a journal’s content is available in large part through Green OA. So, clearly, you do NOT speak for all librarians.

You all seem to have missed 30 years of journal price increases that far outpaced inflation every year until the crash and are growing again now at around 6% while the economy is barely holding its own. Even where library budgets have been expanded generously, libraries have had to cancel expensive journals, which reduced publisher revenue and in an all-subscription market simply brought more price increases to the universities that could hold on. That is what has been broken and what OA can help to fix.

BTW, I agree that $75-$200 journals are reasonably priced(!) but those must be prices for individual subscriptions. The prices charged university libraries are more like $1,500-$4,000.

Why is it unfair to quote extreme prices? How is that incompatible with saying nobody wants to do away with subscriptions altogether? Libraries pay $600 a year for some journals and $6,000 for others. Obviously, the $600 journals are good buys for our students as well as faculty and it is the $6000 subscriptions that are inefficient. There is never 10 times more or better content in more expensive journals. Now that libraries have uniform usage statistics, we can measure cost per download and prove that where gold OA affects the market, cost per use decreases. Where costs are increasing it is because for-profit publishers must provide ever-larger profits to investors and maintain active marketing departments to help convince scholars to continue to publish in their journals so that libraries will keep buying them. If OA allows access to enough articles to allow some libraries to cancel some very expensive journals but still provide pretty good access for patrons, how is that bad? How would it do away with subscriptions? A more efficient publisher will come into the market.

–A librarian

You seem to have missed the recent study suggesting that the notion of steep price increases has been driven to a large extent by measurements of print subscription prices, while online pricing has actually led to deflation of prices. Also, the volume of research being published has also put a strain on library budgets, offsetting these counterbalancing efficiencies. See for our coverage of this study and a link to the entire study.

Libraries pay more because you serve more people per title. Is $1,500 reasonable for an entire campus? From a publisher’s perspective, we’re risking or foregoing a lot of revenue. You can’t have it both ways — serve a lot of people for the price of serving one.

It’s unfair to quote extreme prices as your only examples because it’s misleading and often used as a cudgel in public discussions, to tar publishers as exploitative. Studies of the value of high-priced journals have routinely shown they come in as reasonably priced. In addition, most journals are affordable given what the purchaser is getting.

Comparing prices is just the start for comparing value. The pressure is on high-priced journals to prove their worth. If they do not, cancel them. If they do, they are worth it. Quoting prices is not comparing value, I agree. It’s just the start.

As for efficiency of publishers, they are already quite efficient, given everything they do.

My comments have nothing to do with the pricing of print journals. Nor of “notions” of steep price increases.The 6% increases being demanded this year by a few large publishers are for online-only subscriptions, and are similar whether for the “a la carte” subscriptions, we select title by title or the “big deal” publisher packages where we get a lower rate per title because low-use titles cannot be cut.

My library converted to electronic subscriptions to realize those deflated online prices years ago and those decreases in subscription prices were balanced within a year or two with increases that continued to outpace inflation. Most years the only way to balance a serials budget is by canceling a few high-priced journals, and although we do it on a cost per use basis, we have long since canceled all the journals that consistently had very little use on our campus.

How about a link to those “studies of the value of high-priced journals [that] have routinely shown they come in as reasonably priced?” I have never read one of those. Or do you mean that all journals come in as reasonably priced and I am being unfair again to single out high-priced journals?

As for saying, “most journals are affordable given what the purchaser is getting.” You reminded me of the publisher who objected when I canceled a journal for which the price had nearly doubled, telling me that with the broadened scope this year we would have gotten twice the content. That the added content was irrelevant to our curriculum never occurred to him.

Is the 6% increase due to the volume of the research being published? That’s the major driver of costs in scientific and scholarly publishing. More funding for research, more PhDs and researchers publishing more papers, more journals publishing more articles. It’s what’s really driving the inflation. Allington’s point is that perhaps it’s not the publishers’ fault, but rather that the policymakers have failed to fund libraries adequately and realistically given all the supply and demand they’ve created.

Your experience comports with this. Savings from online lasted for a while, but the sheer volume overwhelmed it a few years later.

It’s fine for you to cancel journals that have no value or that posit value statements you don’t agree with. That’s rational behavior. Keep it up!

To be fair though, you’d first want to look at the rate of inflation (though this can be tricky, and may depend on the country of origin of the journal to get a sense of their increased costs). Then you’d look at the price increase on top of that and question whether it was due to increased pages or other factors. So it’s not a question of 6% being due to increased volume, it’s a question of 6% minus the annual rate of inflation being due to increased volume.

To be fair, I use cost-per-download statistics from my own institution for the past 3 years and compare those within disciplines, as best I can, and try not to leave any department totally bereft because they have fewer majors (and their faculty have been going to nearby consortium partner libraries for their research needs, out of necessity, for many years, so that their activity does not appear in my statistics). In terms of utility, that gives fair feedback to publishers whose mission is to filter research so that the best and most widely useful articles command the greatest academic audience.

As an academic librarian, I am well prepared to provide students with the best (affordable) examples of scholarly publishing in the disciplines we teach, but when it comes to the macroeconomics of academic publishing … to be fair? … lots of important contributions are made by those who teach undergraduates to make a living, and expecting them all to hand over unqualified copyright to publishers, when their compensation is declining, let alone when more and more are contingent workers is not fair. Green Open Access is fair.

Green Open Access will not eliminate journal subscriptions in libraries, but it is beginning to make scholarly articles reasonably accessible for researchers throughout the country, regardless of the tenured/untenured, adjunct/temporary or completely unaffiliated status of the researcher.

The fact that publishers continue to flourish despite increasing use of institutional repositories does not mean OA is not needed. It means that research publication flourishes when pay-walls are not the Alpha and Omega of scholarly communication.

Cost per download is indeed a much better metric to get a sense of the changing economics of things. But where cost per download remains static or decreases, and the number of downloads increases, total costs can go up. Elsevier made this point when justifying their prices in a recent open letter:

First, the cost of downloading an article has never been lower than it is today — on average one fifth of what it was just 10 years ago. As the effective price paid per journal accessed has decreased, the number of journals accessed has increased, and the usage of those journals has grown by over 20% per year.

Your mileage may vary, but they give some actual numbers in this pdf:

-UK example (LISU): 20% decrease in average price paid per journal, ’99-’03 -Effective price paid per Elsevier article downloaded: from €12 to €2 and still falling
(45% annual decrease)

Things get complicated though, particularly when you’re dealing with “Big Deal” collections so your reality is likely not as straightforward as presented here.

My opinion regarding Green OA is perhaps a bit different from that of some of my Scholarly Kitchen colleagues. I think it shows promise as a reasonable approach, and I particularly see the value of it in terms of an institution creating a permanent archive of the work that it has done. I do caution though, that it is important to balance the gains offered by Green OA in a sustainable manner, so as not to “poison the well”. And that means evidence-based, rational approaches to setting proper embargo periods. The goal is to release the information freely as quickly as possible without destroying the journals upon which it relies. With the proper balances in place (and these will vary greatly from field to field) I see this as a useful path forward.

Green OA is a double-edged sword as far as contributing to a university’s reputation. Yes, it shows what research is being conducted, but as one who has seen unedited manuscripts for over 45 years, I can assure you that there is potential for embarrassment also in posting pre-edited versions.

In the case of my library, there is no question about why we are underfunded.We are a state university in a state that has reduced support from about 2/3 cost per student to 1/3 cost per student over the past 30 years, and we are better off than many other states. There are other factors, of course, but in a conversation that assumes government intervention is unwarranted now, I just want to ask where has everyone been?

This statement would suggest your views align well with those of Allington on some subjects. A significant part of the problem is that quantities of information are increasing and the costs needed to support researchers’ needs for that information are similarly increasing. At the same time, financial support for providing for those needs is decreasing. Even where universities have massively increased tuition and other charges, the increased revenue is not being offered to satisfy information needs.

The solution to the problem may indeed involve OA as you suggest. However, blaming everything on publisher price increases is too simple a way to view things.

Judith, I do not see how what you describe justifies government intervention. The obvious solution is to stop buying inefficient journals.

David, I am not advocating government intervention. I am advocating green Open Access.

Often, when I stop buying those inefficient journals, I stop buying a small fraction of the research output of my own faculty. If they have signed over copyright to a publisher, the only way I can provide access to an article in a journal my library has canceled is: (1) to tell the patron to wait a week or so while we find a library to lend it “at no charge” (a transaction with a significant cost to both the lending and borrowing library, btw) and one that cannot be repeated indefinitely without putting us in violation of copyright “fair use” provisions) and/or (2) to purchase that one article from the publisher, over and over, as many times as yet another patron needs to read it.

When the author is on our own campus, couldn’t I just send the borrower to ask for a pre-print? Again, up to a point, yes, and perhaps in the case of a 6 to 12 month embargo, that will fill the need quite well, but the the U.S. government mandate pertains only to federally-funded research. In the humanities, not to mention for institutionally-funded research, copyright is older than Mickey Mouse. (N.B: Daniel Allington is writing from the U.K., where government support of research and university education is a given. I am writing in the U.S. where “government intervention” is suspect.)

The scholarly journals market is incredibly inefficient by nature. Unlike books, which can be purchased based on known attributes of the author, subject, and so on, scholarly journals demand prepayment for an unspecified product every single year. The publishers decide how many articles to publish, how much to charge, and depend on the known attributes of past achievements to sell the lot.

Green Open Access is not government intervention, but it is an intervention that can rationalize what is otherwise a “take it or leave it” market that ill-serves the great variety of American higher education institutions, let alone the variety of researchers who depend on libraries.

Yes, but Rick is trying to explain why the market is inefficient. I agree with him, but I don’t expect a macroeconomic assessment of academic publishing to predict the future of scholarly communication, or even provide a model of the current rate of change.

Scholars will do their best to advance their disciplines, and as long as their methods are effective and the rewards remain stable, they will continue in the same manner. Since peer-reviewed journal publishing is far too slow to promote the advancement of science, disciplinary repositories arose decades ago. Since tenure-track faculty positions are rapidly diminishing, institutional repositories have begun to provide a form of security for the scholarly record that the marketplace does not, so they will expand. Publishers who concentrate on filtering the most important articles and selling them at the best possible prices will also prosper, but the expansion you anticipate will not happen.
The idea that more research is being done, therefore more journals will be sold, and so academic library budgets should be increased accordingly is simply not realistic.

Strange assertions. Journals are too slow to promote the advancement of science? Where is your evidence of this? Institutional repositories provide a form of security for the scholarly record that the marketplace does not? Many repositories use the same “security” as commercial publishers (e.g., CLOCKSS, Portico). Publisher sites are much more discoverable on average, as well. Then you say that publishers who filter the most important articles will prosper, which seems to contradict your prior two statements — these journals will prosper while being too slow to advance science and leaving researchers with insecurities that their papers will be preserved and read?

I also strongly disagree with your last statement. Think about it. Funding for research has increased, funding for science education has grown, and the supply of trained scientists has grown accordingly. They are publishing more, and there are more of them. Yet, library budgets have diminished as a percentage of university budgets. In a logical world, library budgets would have at least kept pace. What is unrealistic to me is that librarians are angry with publishers and seeking to dismantle them to deal with this. The sensible path seems to be to use the incredible power of the library community to get funding back on par with overall academic spending. This will help libraries, librarians, and publishers, as well as scholars and researchers. Why wouldn’t you advocate for this? It seems to be in your own self-interest, and in the interest of the communities you serve. It also accords with the facts.

My evidence is, a repository that is 20 years old now because physicists wanted to collaborate in ways that journal don’t support. By sharing pre-prints, and ultimately finished reports they speed up scholarly communication. That has not wiped out physics journals, but it is evidence that more research does not have to mean more and bigger journals. Not everything every scientist ever does needs its own peer-reviewed journal article to be creditably documented. For junior academics, it’s true that there is no other universally recognized measure of productivity, but that is not a reason my library has to contain more and more every year.

I think the reason you hear from angry librarians is that you keep telling us it is our job to buy your journals. It is much more complicated than that to serve our communities well with the resources we have. As for asking for more money, we have done that for years and years, which is plenty of evidence that the library community has very little power in the matter of university budgets. After 30 years of asking for more than I get and working full-time at providing the best possible access to the best possible mix of resources for my community, it is a bit annoying to be told I don’t have the facts.

As for security, even CLOCKSS says the more copies the better. Visit Harvard and read about DASH please.

Well, the strain is being felt across the system. Publishers keep seeing their submission rates rise, and rejecting articles is expensive. OA isn’t capable at current prices of supporting an entire publishing operation of the kind we’re used to (don’t cite PLOS, as it uses a form of peer review and an approach to editing that is not what we’re used to providing). Librarians are stressed, publishers are stressed, policymakers are stressed, authors are stressed, researchers are stressed. We need to find a better way than fighting, and I was trying to point out that there might be something we agree on — that is, libraries should be adequately funded so they can afford the content their constituencies want. Instead of going it alone, I think there is an opportunity for authors, researchers, publishers, and librarians to show the funding problem and the chaos it is causing.

As for arXiv, it has a troubled history, and requires subsidy from whoever is hosting it, currently Cornell. And it hasn’t changed the journal ecosystem in high-energy physics.

Time will tell if all these repositories are helpful or just another drag on systems, people, and funding. Early analyses and some logical projections suggest they will not be terribly effective, widely trusted, or heavily used. Whether those who are advocating for them and implementing the policies propping them up will ever be held accountable is another matter.

Judith, what do you mean by green open access (without government mandate)? These OA terms are hopelessly vague. What exactly are you advocating?

Not really vague, just complicated. I can start with simple examples, though, then refer you to more comprehensive sources.

Green open access is provided by the deposit of author manuscripts in online repositories which include metadata that is harvested by the major Internet indexing agencies, most notably Google, particularly because of the effectiveness of the search utility called “Google Scholar,”

Disciplinary repositories arose first because journal publishing was too slow to support scholarly communication in the sciences. The “” e-print repository started in 1991 with physics papers and has expanded since to include mathematics, astronomy, computer science, quantitative biology, and more. PubMed and BMJ followed for similar reasons. Seeing the greater impact of open access research reports, NIH and other funders (Welcome Trust, for example) began to require (and fund) open access publishing of research results.

Even in disciplines with longer timelines, it became obvious that tying scholarly communication to publishers’ economics creates barriers to progress. In Europe, where governments invest heavily in research and higher education, investing in institutional repositories was a no-brainer. When faculty are obliged by institutional policy to deposit the last author’s version of their research articles in their institution’s open access repository, they can pass along the remaining rights to publishers and publishers’ contributions can be acknowledged also by linking to the published version (without withdrawing access to the author’s version, hence enabling rapid global access to research findings).

In America, green OA has kicked up a great deal of fuss only recently. As you know from Daniel Allington’s blog, the U.K. has provisions for research reporting that are not being adequately supported by the government at present, and a completely separate government policy has recently made an unfunded commitment to “gold open access.” I maintain that it is the publishing industry’s responsibility to filter research output so that unprofitable articles are not underwritten by library subscriptions to scholarly journals. (If “gold open access” (underwritten by a combination of government grants, research fellowships, institutional investments, charities or whatever) can bring more research to light, that may be a good thing, but it is not good policy for any government.

In the U.S., there is no evidence that green OA has affected any publisher’s profits, as yet, but some publishers are coming out in opposition to “institutional mandates” based on the Harvard model. I can only conclude that this is an obstructionist position intended to bring all research under the for-profit model when demand is insufficient to support it.

For all my efforts, I did not simplify very much at all, did I? It really is complicated, but given all those folks working to explain economics and still disagreeing with one another, that’s not surprising, is it?

Anyway, please see Peter Suber’s Open Access Overview
particularly the section on “OA repositories (“green OA”)”

Also, the SPARC (Scholarly Publishing and Academic Resources Coalition) page on Authors Rights
and their page on Campus-based Publishing

Also, the Harvard Office for Scholarly Communication page.
My institution is far from the Harvard model university, but the Harvard model Open Access Policy would go farther than any subscription or article purchasing policy I could institute to help repair what is broken in scholarly communication in the U.S. today.

I agree with Joe that the institutional repository model is intrinsically self defeating. If all journal articles were to become easily available for free no one would subscribe so there would be no articles. Of course we are a long way from that condition but it is not a viable policy direction.

You continue to argue one side, while ignoring facts on the other side, or logical extensions of trends you support. Green OA hasn’t hurt publishers much yet, but it has the potential to do so, at which point Green OA stops working because there aren’t any final manuscripts to point to, no “pre-publication” version, and no third-party peer review. Harvard’s OA policy looks like it will cost them 2-4x more than subscriptions, but yield less available content for their community. That’s the policy of an ideologue, not a pragmatist. And so forth.

I’m consistently struck by the time lag I’m seeing these days in comments. Some people are moving along with the facts that are emerging, the trends in the arguments, and so forth. Allington’s post sums up many of these nicely. Some people don’t seem to be following these trends or emerging facts, or are unwilling to integrate them into their thinking. I guess change occurs at variable speeds.

If the Harvard model is so costly, why are so many universities adopting the same model? One of the facts that is emerging is that the University of California system has announced they are adopting it.

As for green OA having the potential to hurt publishers, where is that post about the 73 things publishers do? I am not ignoring the other side. Article content never has been the only thing we buy when we buy journals. The best journals from the most efficient publishers will continue to sell subscriptions and if others go under they will be replaced. That’s the American way.

Allington’s comments are well taken. I do not advocate gold OA for some of the reasons he explained. The situation with regard to copyright and government support are different in the U.S. however, as I have been trying to explain from “my side.”

This and the essay it cites are very comprehensive discussions, congratulations to the A’s. I wonder though if anyone can point to an assessment of the impact on researcher behaviour and the nature of the body of literature in any field if there is a substantial switch from reader-pays to author-pays (or nobody-pays) models of publishing?

In my publishing career I have seen the academic journal change in many ways. In the 1980s I was employed to manage peer review and publication of specialist computer science journals for an international publisher. I also had to write a news section each month, and commission book reviews and review articles. By the 1990s I had moved on and so had the journals – all “editorial” content had stopped as had most review articles, and only research articles remained. I haven’t kept up, but I sincerely hope that a large part of the content is now available as uncompiled code or data carefully curated to accompany articles and allow reuse by others in the field. In other words careful thought has been given, at least in the subscription journal world I knew, to what content would be most useful and valuable to the reader.

I worry that publishers will have financial incentives driven by authors not readers. Clearly author-pays will encourage publishers to encourage writing, which matches the interests of many researchers via their funding bodies. However there is no incentive to make articles “readable”. There is barely any incentive to make them discoverable, beyond making sure the author can discover their own article. Everyone is happy if research papers that differ only very slightly are submitted to multiple outlets – more author fees, whoopee; more papers published, hurrah. Already many journals don’t bother editing articles at all and have pared peer review down to the minimum, so there is a question whether some articles add any value to the body of literature. Who cares if journals or articles are read? If there is no prospect of unsubscribing, then publishers certainly won’t care (that is to say, they won’t care about the journals in the long tail beyond the competitive world of the impact factor “top of the pops” charts, and for most publishers that is where their money is made).

In the subscription world there is one rather large incentive to keep quality high and encourage reading, and that is the likelihood of losing subscriptions if usage falls. Even then, given the big deals offered by many publishers, there is a risk that a dreadful valueless journal can limp along under the cover of a massive collection, but at least there is a chance it will be put out of its misery.

In our world there is way too much writing, and not enough reading. You need both to encourage healthy advance in all fields of research. I am concerned at the impact these changes in the publishing model will have on research conduct. Has anyone else done work on this?

In terms of reader behavior, you might be interested in looking at where they are adding article metrics based online use so we have more than peer review and journal impact factors to help filter our reading. (Although “altmetrics” means “alternative metrics,” not article metrics).

As for author-pays problems, perhaps you are aware that a whole passel of predatory publishers and fly-by-night tricksters have arisen to prey on authors. See Beall’s list

Yes thanks Judith, I am aware of the metrics movement (and things like Kudos look interesting) as well as the predatory author-pays pack (and the recent “sting” which seemed to begin separating the sheep from goats). I’m also aware of a tendency to continue “bottom-feeder” journals in all sorts of publishing lists which fulfil the need to publish without worrying too much if anyone is reading the output.

This is not an argument necessarily between OA vs subscription but a more fundamental author vs reader issue in which the publisher plays a part. When print was expensive publishers tended to be a throttle on authorship; now digital is cheap there is a tendency to create an information swamp that our discoverability and metrics efforts can hardly contain. The move to author-pays means the important review journal article, along with other valuable navigation tools, are becoming endangered. Low costs mean journal death, even when merited by dire performance, becomes rare. Perhaps I am overstating this and I certainly have no easy solution, but would welcome thoughts. It would be depressing to find publishers were making it harder for people to add to the world’s knowledge, not easier.

Indeed, we have discussed these two issues at length here. Searching on “altmetrics” lists 25 articles using the term and on “Beall” lists 8, some quite recent in both cases. And this search function only finds Kitchen articles using these terms, not comments, where both have turned up frequently.

On the other hand, if these publishers are actually publishing the articles it is not clear how this is predation on the authors. Beall complains about them being publishers of last resort but that seems like a valuable service from the author’s point of view. Thus the term “predatory” has always bothered me.

I agree “predatory” is unfair in those cases, but I know why people feel preyed upon because I have a .edu email address, so I get invitations to publish nearly every day. I put them on my blocked senders list then get still more from other email addresses. Perhaps those are all fly-by-night folks rather than “last resort” ones. I have to admit I’ve never checked.

Comments are closed.