A field guide to misunderstandings about open access
SPARC Open Access Newsletter, issue #132
April 2, 2009
by Peter Suber
The woods are full of misunderstandings about OA. They thrive in almost every habitat, and the population soars whenever a major institution adopts an OA policy. Contact between new developments and new observers who haven't followed the annual migrations always results in a colorful boomlet of young misunderstandings.
Some of these misunderstandings are mistaken for one another, especially in the flurry of activity, because of their similar markings and habitat. Some are mistaken for understanding by novices unfamiliar with the medley of variant plumage, adaptive camouflage, and deceptive vocalizations. This field guide should help you identify 25 of the most common visitors to your neck of the woods.
Leave your binoculars at home. All of these can be seen with the naked eye. With no more than this guide, and some patient observation, every trip to a conference, and even an occasional faculty meeting, can be an enjoyable and educational outing.
(1) "All OA is gold OA."
* "OA mandates require submission to OA journals." (Requiring OA means requiring OA through journals, right?)
* "OA mandates limit our freedom to submit work to the journals of our choice." (OA mandates require submission to OA journals, right?)
* "OA mandates are simply not feasible. There aren't enough OA journals to absorb the volume." (OA mandates must be gold OA mandates, right?)
* "I wish I could provide OA to my own work but there are no good OA journals in my field." (There's no other way to provide OA to my own work, right?)
* "To make large advances in the overall levels OA we need large concessions from publishers." (All OA is provided by publishers, right?)
Gold OA is OA through journals, regardless of the journal's business model. Green OA is OA through repositories. The mistake here is simply to overlook the existence of green OA. This misunderstanding is common enough to have a name, or in fact two. Stevan Harnad calls it "gold fever" and I once called it "JAM" (for Journal-Archive Mixup).
There's little excuse not to know about green OA. At the end of last year, the Registry of Open Access Repositories and Directory of Open Access Repositories each listed more than 1,200 repositories, about 30% more than the year before. According to Scientific Commons, more than 7.5 million items were deposited in these repositories in 2008 alone, or more than 20,600+ items every day, about 45% more than the year before.
If you're looking in another direction, you might not know about this growth. But the crowing and cacophony alone should draw attention to the existence of repositories and their differences from journals.
There's even less excuse to mistake a policy requiring deposit in OA repositories for a policy requiring submission to OA journals. One only has to read the policy. For example, the NIH and MIT policies do not require submission to OA journals. Moreover, no OA mandates anywhere require submission to OA journals.
(2) "Low levels of spontaneous self-archiving reflect opposition to OA."
Self-archiving is the act of providing green OA to one's own work by depositing a copy in an OA repository. About 15% of new research articles are spontaneously self-archived by their authors, and OA policies can nudge the number up toward 100%.
But why is the spontaneous rate only about 15%? You could read the studies about what researchers know and believe about OA. Or you could wing it.
* "The rise of OA mandates proves that researchers oppose OA and must be forced." (All researchers understand their OA options, right?)
* "University OA mandates coerce faculty." (Faculty never vote for OA mandates, right?)
* "OA mandates violate academic freedom." (No mandates, such as teaching classes, or publish or perish, are compatible with academic freedom, right?)
In the early days of OA, most university OA policies were adopted by administrators rather than faculty. Alma Swan showed in two separate studies that an overwhelming majority of faculty would "willingly" comply with an OA mandate from their funder or employer. In the second study she also showed that 71% of authors who had never self-archived had never heard of it.
Recently, however, and for more than a year, most university OA policies have been self-imposed by faculty, usually by unanimous votes. Unanimous faculty votes adopted strong OA mandates at the Harvard Faculty of Arts and Sciences, Harvard Law School, Stanford School of Education, Oregon State University Library Faculty, and MIT. The policies at Macquarie University and Boston University were adopted unanimously by the faculty serving on their University Councils. The OA mandate at Stirling was adopted by faculty vote (I don't have the tally) and the OA mandate at the Harvard Kennedy School of Government was adopted by a 98% supermajority.
If someone published a speculation five years ago claiming that we'd soon see this kind of faculty support, the ivory-billed woodpecker of scholarly communication, it would have been dismissed as wishful thinking. But now that this kind of faculty support has been repeatedly attested, how should we regard the continuing dismissals?
(3) "OA archiving will kill journal subscriptions."
* "OA archiving is already killing journal subscriptions." (We can point to examples, right? At least somebody can, right?)
* "OA archiving must be killing subscriptions, even if we don't have evidence yet." (There's no reason to renew a journal when inferior versions of an unknown number of its articles will become OA after a delay, right?)
* "As OA archiving reaches high levels, it will eventually kill journal subscriptions." (Even though physicists have self-archived for nearly 18 years, at levels approaching 100%, and physics journals report no cancellations attributable to OA archiving, it's only a matter of time, right?)
* "OA archiving will kill peer review." (OA archiving will kill subscription journals, right? Only subscription journals can perform peer review, right? Unpaid referees will not evaluate a paper until they are satisfied that the journal charges subscriptions, right?)
* "You can't sell something when essentially the same thing is available free of charge." (Nobody would buy bottled water, right?)
While this prediction is clamped to a misunderstanding of the evidence, it may turn out to be true. There's no evidence yet that high-volume OA archiving will kill subscriptions, and there's substantial evidence from physics that it won't. But for all we know now, the disciplines might really differ in this respect. For a fuller discussion of the evidence and nuances, see my article from 2007.
(4) "OA is about bypassing peer review."
* "Funders shouldn't mandate OA. Peer review is an essential filter on research." (OA mandates require authors to bypass peer review, right?)
* "Universities mandating OA? They should be the last institutions to give up on peer review."
* "I can't submit my work to OA journals. I need to publish in peer-reviewed journals for promotion and tenure."
* "Research articles are not like blogs or home pages, which you can just slap online without any peer review or quality control."
* "I don't understand how the OA movement can attract serious researchers. Wikipedia isn't a serious research tool."
It has never been true that OA is about bypassing peer review. The OA movement focuses on OA for peer-reviewed literature. The goal is to remove access barriers, not quality filters.
An important part of the OA movement is devoted to preprint exchanges, which deal with unrefereed manuscripts. But these manuscripts are en route to peer review, not circumventing peer review. One important purpose of preprint exchanges is to give authors feedback to improve their manuscripts before they are published in peer-reviewed journals. (Another is to put an early time-stamp on their findings or conclusions.)
But this misunderstanding does not arise from the phenomenon of OA preprint exchanges. It arises for most people even before they understand the difference between OA preprints and OA postprints. It seems to be fed by a simple, careless presumption. Most of the stuff online, especially the stuff that's free online, was never subjected to peer review. It's easy to put stuff online without subjecting it to peer review. The blooming, buzzing internet, its spectacular growth, and its veins of pure crap, are due to the ease of making stuff free of charge and free of old-fashioned editorial hurdles. The project to make research free online must be more of the same, right? (By the same logic, free online research must be pornography.)
Some of the misunderstanding (as in #1) arises from not reading the OA policies one wishes to criticize. The NIH policy is typical in limiting the OA requirement to peer-reviewed manuscripts.
(5) "OK, then, but OA journals skimp on peer review."
* "I never heard of that --admittedly new-- journal. How good could it be?" (Journals, even new journals, always have reputations which match their quality, right?)
* "Our journals are not OA. On the contrary, they are high in quality and use rigorous peer review." (You can't attract good authors, editors, and referees by promising wide access, right?)
* "Quality is our top priority, which is why we can't consider OA." (When you make work widely available, the quality deteriorates, right?)
* "You get what you pay for." (If OA journals have no subscriptions, they must have no revenue, right? Authors, editors, and referees are expensive, right? See #8.)
OA journals can use the same peer review procedures, the same standards, and even the same people as toll access (TA) journals. This isn't hypothetical, and actually happens whenever established TA journals convert to OA. The key variables in journal quality are the quality of authors, the quality of editors, and the quality of referees, all of which are independent of the journal's price or medium.
Some critics think they make the case by pointing to weak OA journals, as if no one could point to weak TA journals. To make the case, one would have to show that the average OA journal is less rigorous with peer review than the average TA journal, or that something intrinsic to OA, more than anything intrinsic to TA, conflicts with rigorous peer review. Publisher-critics once tried the line that author-side publication fees bias the editorial process. But we've heard this line much less often since (1) most of the publisher-critics have adopted hybrid OA models which charge author-side publication fees, (2) Kaufman and Wills showed that most OA journals didn't charge author-side fees at all, and (3) Kaufman and Wills also showed that more TA journals charge author-side publication fees than OA journals, both by numbers and by percentages.
Some of this misunderstanding is simply the result of confusing quality and prestige. A new journal can be excellent from birth, but cannot be prestigious from birth. Because most OA journals are new, most of them are still climbing the slope, trying to gain prestige in proportion to their quality. Some are high in quality and some low; and some are achieving reputations faster than others of equal quality. But authors --and promotion and tenure committees-- are accustomed to using prestige as a surrogate for quality. The inference is: low on my radar, therefore low in quality.
For the misunderstanding that there must be a trade-off between OA and prestige, see #20. For the misunderstanding that OA is best suited to second-rate work, see #24.
(6) "OK, then, but at least *green* OA is about bypassing peer review."
* "Green OA mandates will lead to junk science." (They pass over journals for repositories in order to bypass peer review, right?)
OA repositories don't provide their own peer review, but they can and do host articles peer-reviewed elsewhere. Those articles can be, and are, peer-reviewed at the best journals anywhere. OA policies which request or require green OA never ask scholars to bypass peer review. Across the board, at funding agencies and universities, these policies call for OA to *peer-reviewed* manuscripts. (See #4.)
(7) "OA is about punishing greedy or obstructive publishers."
You can't throw a brick out a university window without hitting a researcher, librarian, or administrator frustrated and furious with a set of TA journal publishers. For many of them, the problems for which OA is the solution are defined by these frustrating and infuriating experiences. But it doesn't follow that OA must function as punishment, for anyone. To pursue it as a punishment is to mistake the goal.
As I put it in my OA Overview: "The purpose of the campaign for OA is the constructive one of providing OA to a larger and larger body of literature, not the destructive one of putting non-OA journals or publishers out of business. The consequences may or may not overlap (this is contingent) but the purposes do not overlap."
The rise of personal computers in the 1980's may have hurt the typewriter industry, but it doesn't follow that the purpose was to hurt the typewriter industry. But when you've suffered at the hands of Royal and Olivetti, it's easy to become distracted and take your eyes off the prize.
This misunderstanding has a surprisingly diverse habitat. You'll find it among some caffeinated academics who are avid for OA. But you'll also find it among besieged TA publishers who would rather believe that OA is an ideological attack on what they are doing than acknowledge that it's a serious and sophisticated alternative or supplement to what they are doing. The lesson for both is that OA would still be an urgently good idea if TA journal prices were low and licensing terms reasonable. For more along these lines, see my reflections on OA as solving problems and OA as seizing opportunities.
(8) "OA journals couldn't possibly pay their bills."
* "We can't convert our journal to OA because we need the revenue." (OA journals can't have revenue, right?)
* "We can't convert our journal to OA because we need the profit." (OA journals can't make profits, right?)
* "There's no free lunch." (Publishing costs money and requires revenue. I bet you never heard this one before, right?)
* "It's not sustainable to give something away forever." (Broadcast television and radio are not sustainable, right?)
The misunderstanding here is to assume that OA journals have no revenue or subsidies, or that their revenue and subsidies could never suffice. In the early days, a related misunderstanding was that OA journals were na´ve about publishing and thought it could be made costless. (Because the only way to run a journal with no revenue is to have no expenses, right?)
Journals that charge publication fees have revenue from the fees. Journals that charge no fees use a variety of business models, often relying on subsidies rather than revenue. The long-term viability of new OA journals (like new TA journals) can only be assessed in the long term, but we can already point to a group of OA journals and publishers which have gone beyond revenue to profits or surpluses: BioMed Central, Hindawi, Medknow, the Optical Society of America, and PLoS ONE.
OA is like broadcast TV or radio: the content is free of charge for the end user, the providers can be more than solvent, and there is more than one way to generate the revenue or subsidies needed to pay the bills. Of course there are also some differences: With research articles, the productions costs are much lower and the creative talent gives away its work.
This misunderstanding sometimes forms a lifelong bond with #4 or #5, in which case their combined song is slightly different from either song alone. "OA journals don't really perform peer review, or don't perform real peer review. Facilitating real peer review costs money." (OA journals can't find the money, right?)
(9) "OA is a business model."
* "We can't adopt the OA business model because we need revenue to cover our costs."
* "We adopted the OA business model last year and are doing fine."
* "The OA business model is untested (or tested) and inadequate (or adequate)."
This vocalization is often mistaken for #8, "OA journals couldn't possibly pay their bills." To tell it apart, listen for the definite article: "the OA business model...." If you're patient and lucky, on occasion you will hear it opposing #8 and asserting that "the OA business model" is a proven success.
OA is a kind of access, not a kind of business model. It's compatible with a large number of business models, and creative journals are still discovering new ones and new variations on older themes. Here's a growing collection of them, on a wiki where you can add new examples and detail.
Many people who say that OA is a business model are assuming that there is only one business model for OA journals and that all charge publication fees. Both assumptions are false. (More on the second in #10.)
But the fundamental mistake I'm trying to identify here is the claim that OA is a kind of business model rather than a kind of access. We don't get to OA by tweaking business models, even if we must tweak business models to get to OA. We get to OA by removing access barriers. That rules out the classic business model of charging for access. But once we do that, we have OA and a large space of alternative business models to explore.
(10) "All OA journals charge publication fees."
* "I can't publish in an OA journal because I can't afford the fees." (All OA journals charge fees, right? No fee-based OA journals waive their fees, right? No funders or universities pay the fees on behalf of authors, right?)
* "OA is supposed to be some kind of revolution. But in the past readers had to pay, or libraries had to pay on their behalf. Now authors have to pay, or libraries have to pay on their behalf." (All OA is through journals, and all OA journals charge fees, right? See #1 and #13.)
* "OA journals are bad for authors in developing countries. They can't afford the fees."
* "OA journals are bad for authors in fields where little research is funded. They can't afford the fees."
It's false that all OA journals charge publication fees, and it's even false that most do. Most do not. Cara Kaufman and Alma Wills first discovered this fact in a 2006 study for ALPSP, AAAS, and Highwire Press. Their finding that about 53% of OA journals charged no fees has been extended and confirmed by later studies. In November 2007, Caroline Sutton and I found that 83% of *society* OA journals charged no fees, and in December 2007, Bill Hooker found that 67% of all the full (non-hybrid) OA journals in the DOAJ charged no fees.
Sometimes this misunderstanding mates with the erroneous wild guess that funding agencies are unwilling to pay these fees. The result: "If all peer-reviewed journals converted to OA, universities would pay more in publication fees than they pay now in subscriptions." (All OA journals charge fees, and all fees will be paid by universities, right?)
(11) "OK, then. But all *high-quality* OA journals charge publication fees."
* "Yes, some OA journals don't charge fees, but the good ones do." (I've heard of some good ones that charge fees, and the rest must be like that, right? If I haven't heard of a good no-fee OA journal, there couldn't be any, right?)
I've seen no studies comparing fee-based and no-fee OA journals for quality. Nor have I seen studies comparing them for citation impact, which is many people's surrogate for quality. The misunderstanders are substituting guesswork or prejudice for evidence. In the absence of more systematic studies, we can point to first-rate no-fee OA journals from organizations such as the American Mathematical Society, Beilstein Institute, Duke Law School, the Max Planck Society, MIT Press, the University of Michigan, and Taylor and Francis.
Today no-fee OA journals are suffering much the same stigma that all OA journals suffered a few years ago: the groundless assumption that they must be fly-by-night. If they're unheard of, then they must be unheard of. Ten years ago, misunderstanders could say, "I haven't heard of any good online-only journals, so there couldn't be any." Five years ago they could say, "I haven't heard of any good OA journals, so there couldn't be any." Now they can say, "I haven't heard of any good no-fee OA journals, so there couldn't be any." Field experience tells us that not every bird appears in every wood, and that even when they are abundant in the wood where you are standing, you have to look or listen.
As in #5, some of this misunderstanding is simply the result of confusing quality and prestige. It's unlikely that many scholars will know all the high-quality journals in their field, but inevitable that many will know all the high-prestige journals in their field. The reason is that prestige is defined by what many people know or believe, and quality is not. Quality can be widely overlooked and unknown, but prestige cannot. To assume that prestige is always a good surrogate for quality, or that prestige and quality never diverge, even for new journals, is to be tricked by another misunderstanding.
By the way, of the profitable OA publishers mentioned in #8, one of them --Medknow-- charges no publication fees at any of its 81 OA journals. No-fee journals are capable financial viability as well as high quality. You might not have guessed it, but that's a reason not to guess.
(12) "Publication fees at OA journals must be paid by authors out of pocket."
* "This is a good journal but I can't afford the fee." (The journal would never waive it, right? My funder or employer would never pay it, right?)
* "Do they really expect authors to pay these fees?" (Why else would they call them "author fees"?)
* "My promotion and tenure committee would never give weight to a journal charging author-side fees." (Publishing based on author fees is vanity publishing, right?)
There are three reasons why authors who publish in fee-based OA journals may not have to pay the fees out of their own pockets: the journal may offer a discount or waiver in cases of economic hardship; their universities may host funds to help pay the fees; and their funders may allow grantees to pay the fees out of grant funds (see #10). Unfortunately I haven't seen systematic tallies of the journals offering waivers, universities with funds, or funders willing to pay fees. So we can't estimate how likely it is that you'll have to pay publication fee out of pocket if you want to publish in a fee-based OA journal. Just know that there are options worth investigating when you face a fee that you can't afford.
Nor should we consider anything to be vanity publishing if it includes serious peer review. Publication fees don't change that. As Jan Velterop (former publisher of BioMed Central) likes to point out, paying to take a driving test doesn't mean that you'll pass.
One of the most interesting results of the Kaufman-Wills report (mentioned in ##5, 10) is that TA journals charge author-side publication fees significantly more often than OA journals, by numbers as well as percentages. K & W found that while only 47% of surveyed OA journals charged author-side fees, more than 75% of surveyed TA journals did so. At TA journals these fees take the form of page and color charges and do not even have the beneficial side-effect of paying for OA. If author-side fees signal vanity publishing, or exclude indigent authors, or corrupt peer review and editorial judgment, then the effect is more widespread among TA journals than OA journals.
The terms "author pays" and "author fees" feed the present misunderstanding and are entirely unnecessary. Some writers, trying to be careful, only use these terms when trying to describe fee-based OA journals, but make the mistake of assuming --or at least saying-- that the fees are to be paid by authors. Other writers are less careful and use "author pays" to describe all OA journals (falling for #10, "All OA journals charge publication fees"), and some even use it for all OA (falling for both #10 and #1, "All OA is gold OA"). I've been trying to shoo this flock from the forest since 2006, but without success: "As more institutions become willing to pay the processing fees charged by the subset of OA journals that charge fees....the terms 'author fees' and 'author pays' will be even more deceptive than they are today. Let's kill them once and for all. They're false when applied to the majority of OA journals that charge no fees. They're misleading when applied to journals whose fees are frequently waived or paid by sponsors on the author's behalf. And they're harmful for raising groundless or exaggerated fears among authors." Their effect today is to impede clear communication and camouflage costly misunderstandings.
(13) "Publication fees at OA journals are just subscriptions in disguise."
* "OA journals pretend to free everyone from subscriptions, but what they're doing is really equivalent." (Not charging for access is equivalent to charging to access, right?)
* "It doesn't matter whether journals are paid by readers or authors, except of course to the readers or authors." (As long as journals cover their costs, it doesn't matter how they do it, right?)
This misunderstanding overlooks the fact that publication fees at fee-based OA journals *buy OA*, which is the whole point. Subscriptions don't buy OA. Both sorts of payments might cover the costs of publication, and both might come from library budgets. Publication fees buy access for everyone with an internet connection, including those who aren't paying fees, while subscriptions pay only for private use and consumption. There are some similarities, but to overlook or minimize the differences leads to misunderstanding.
Another part of the misunderstanding is to assume that journals serve people equally well whether the money comes from the reader side or the author side. If it's too early to make the call for journals, consider the case of mailed letters. Until the 19th century, you could send a letter without charge but had to pay to pick up your incoming mail from the post office. Addressees often had to leave their mail unread, which depressed the utility and volume of mail. When Rowland Hill introduced postage stamps in 1837, precisely to shift the costs to senders, the volume of mail skyrocketed, as did the rate at which attempted mail communications were actually completed. Shifting the costs from the receiving end to the sending end made all the difference.
(14) "OA deprives authors of royalties."
* "It's hard enough to make a living as a novelist without radicals trying to steal your work and block your income." (The OA movement advocates OA for novels, right? All publications pay royalties, right?)
* "OA destroys the incentive to write good work." (Scholars are paid for journal articles, right? You have to twist a scholar's arm to get him/her to publish a journal article without royalties, right?)
The misunderstanding here is to transplant the idea of OA from the domain of royalty-free literature like journal articles, where it originated, to the domains of royalty-producing literature like novels. The OA movement focuses on journal literature for the good reason that journals don't pay authors for their work and generally don't pay editors and referees who facilitate peer review either.
There are growing efforts to provide OA to monographs and textbooks, and even to novels, all of which can pay royalties when sales are good enough. But even these are efforts to win author and publisher consent, not to create vigilante OA without consent. They depend on persuading authors that the benefits of OA (in greater audience and impact) exceed the value of royalties, or persuading both authors and publishers that OA editions actually increase the net sales of non-OA editions.
This may be the single most common misunderstanding of OA outside the academic habitat. (And why not? Outside academe, most publications pay authors for their work. Scholarly journals are the exception whose peculiarity makes them the focus of the OA movement.) But even academics who know little about OA know that scholarly journals don't pay authors for their articles, and that scholarly authors don't let that slow them down. Unfortunately, OA often depends on decisions by non-academics, such as legislators, and popular understanding of OA can be shaped by non-academics, such as journalists.
A related misunderstanding: "Strong copyright is a necessary incentive for author productivity." This may be true for publications on which authors earn royalties, and therefore true outside the domain of scholarly journals. But scholars are eager to publish in journals, even knowing they will not be paid. They are even eager to sign away their copyrights, the supposed incentive for their creativity. The incentive to publish in journals, even without pay and even without ownership rights, is so strong that it drives academics to shirk their other responsibilities and twist their extra-professional lives out of shape. I haven't seen any studies comparing the incentives which drive academic authors with the incentives copyright provides to royalty-earning authors, but I wouldn't be surprised if the copyright incentives paled by comparison.
(15) "Preprint archiving violates copyright."
For this purpose a preprint is any version prior to peer review, for example, the version you submit to a journal.
* "I can't self-archive my preprint. It might violate my publisher's copyright." (I know I don't have a publisher yet, but all copyrights belong to publishers, right?)
Authors who want to self-archive their preprints have reason check the policies of their intended publishers. But the reason has nothing to do with copyright. Some publishers follow what is called the Ingelfinger rule, and do not accept manuscripts which have already circulated as preprints, or which have previously been published or sometimes even publicized. Except in medicine, the rule appears to be on the decline. What matters here is that it doesn't derive from copyright, but from the more fundamental right to refuse to publish any work for any reason. Since the acceptance of your work may really be at stake, by all means check your publisher's policy. But don't assume that the Ingelfinger rule is more widespread than it is, and don't assume that a publisher's right to refuse work is based on copyright.
Many scholars don't realize that they own the copyright in their own work as soon as they create the work, without any need to register it, and that they hold the copyright until and unless they transfer it to someone else, like a publisher. Nor do they realize that publishers only own the rights that authors voluntarily transfer to them. (The neighboring copyright forest has even more misunderstandings than the forest of OA.) It follows that, until authors sign a copyright transfer agreement, they don't need anyone's permission to circulate their work. The preprint archiving decision is all their own.
It also follows that authors should learn to be good custodians of their rights, transfer no more than a publisher really needs, and retain the right to authorize OA.
(16) "Postprint archiving violates copyright."
For this purpose a postprint is any version approved by peer review.
* "I support OA, but I'm a professional and can't afford to violate copyright law." (OA is like music file sharing, right?)
* "I can't self-archive my postprint. That would violate my publisher's copyright." (Publishers don't allow postprint archiving, right? Authors can't retain the right to authorize postprint archiving, right?)
* "OA mandates must make exceptions for dissenting publishers." (Publishers always oppose OA archiving, right? Publishers always acquire the rights to block OA if they want, right?)
* "I just read about a university adopting an OA mandate. How did it ever get publishers to agree to that?" (At least for postprints, the OA decision is always the publisher's decision, right?)
* "OA competes with the publications we're trying sell. This diminishes protection for our copyrights." (We should be able to block OA even when authors retain the right to authorize OA, right? Any threat to our bottom line must be a violation of our rights, right?)
* "When authors sign copyright transfer agreements, publishers acquire the right to control the distribution of the work." (These contracts never allow postprint archiving, right? Authors never divide the bundle of copyright and keep part and transfer only the rest to publishers, right?)
One mistake here is to assume that TA publishers don't or can't consent to OA. But they do and can. Today between 51 and 70+% of surveyed TA publishers give advanced permission for postprint archiving, depending on which figures you accept. (Note that most publishers who allow postpring archiving allow archiving of the final version of the author's peer-reviewed manuscript, not the published edition.)
Another mistake is to assume that OA must be authorized by publishers. It must be authorized by the copyright holders, and authors are the copyright holders before they transfer rights to publishers. The strongest sort of OA policy takes advantage of this fact. At the NIH, for example, grantees must retain the right to authorize OA through PubMed Central (for a certain version after a certain delay). If a given publisher won't accept those terms, the grantee must look for another publisher. The Harvard policy works the same way except that when a publisher won't accept the terms, the author may request a waiver. Apart from the possibility of waivers, these policies ensure permission for OA without seeking it from publishers. It's good that a majority of journals already permits postprint archiving, but it's even better that NIH-style policies can effectively boost the percentage to 100% of the journals in which authors eventually publish their work.
When some publishers object that the NIH policy will harm their business (see #3), and conclude, "There ought to be a law," they're acknowledging that the policy does *not* violate current law and that the only way to block the policy is to amend current law. This is the logic behind the Conyers bill in a nutshell. Nevertheless, publisher rhetoric in support of the bill asserts that the NIH policy does violate copyright law. This is either an incoherent blend of truth (NIH policy is currently lawful) and popular misunderstanding (postprint archiving violates copyright law), or it's a cynical blend, using the prevalence of the misunderstanding to inflate support for an objection they know is false.
Similarly, you might think that publishers who say that the NIH policy will harm their business would not voluntarily accommodate it. When NIH-funded authors submit their work to those publishers, and explain that they are bound by their funding contract to comply with the NIH's OA policy and cannot transfer full copyright to a publisher, you might think that these publishers would exercise their fundamental right to refuse to publish. But that's another misunderstanding. No publishers refuse to publish NIH-funded research. Some, however, will voluntarily accept a submission from an NIH-funded author, knowing the consequences for OA, and then say in public that the NIH "comandeered" "their property" or forced them to "surrender" "their copyrighted articles". This is a deception, camouflaging experienced businesses who sign contracts with their eyes open as victims of theft and expropriation.
Perhaps we can dispel much misunderstanding with a simple admission: Of course OA could be implemented badly so that it infringes copyrights. So could publishing. What matters is that we can also implement it well so that it doesn't infringe copyrights. Institutions care about the difference, and OA policies are evolving to find clearer and easier ways to satisfy this important constraint. One source of the current misunderstanding is that people are guessing at how institutions implement OA. They hit first on ways that violate copyright and pay no attention to how policies have actually evolved to foster lawful OA. One result of this policy evolution is that there at least three clear ways for OA mandates to avoid infringement: a weak way, which requires OA except when the publisher does not allow it (the loophole method), a strong way, which requires OA without a loophole and require authors to retain the right to authorize that OA (the rights retention method), and a middle way, which starts with the rights-retention method and then allows author opt-outs rather than publisher opt-outs (the Harvard way).
(17) "Self-archiving takes too much time."
Les Carr and Stevan Harnad have shown that in the wild self-archiving takes about 10 minutes per paper. This is isn't the time required by practiced archivangelists racing the clock, but the time shown on the user logs of a much-used repository. For scholars who publish a paper a month, and who have an average amount of help from co-authors, librarians, students, or assistants, self-archiving takes less than 40 minutes per year.
Not surprisingly, it takes even less time for scholars who have done it once or twice. Alma Swan and Sheridan Brown have shown that while scholars often worry that self-archiving will take too much time, only 20% had any sort of trouble the first time, and only 9% thereafter.
One habitat for this misunderstanding is the unspoken anxiety of many working scholars. The thought is, "I'm already overworked. I don't know how much time self-archiving takes. But it's one...more...thing. I don't have time for one more thing." This thought is often true. But if this thought is not itself a misunderstanding, neither is it about self-archiving. The right response is not correction, but perspective. Authors spend more time than self-archiving would require on other chores designed to make their work better known: preparing a list of their publications for their department chair or dean, mailing or emailing individual copies to individual colleagues, keeping their online c.v. or bibliography up to date, and so on. Self-archiving offers more impact in less time. The misunderstanding here is not that self-archiving is one more thing, but that it's an inefficient thing, and an unimportant thing when one has to give priority to making one's work better known.
(18) "Self-archiving hides work more than it exposes work."
* "Why should I put my work in a repository if nobody will know it's there?" (To find something online, you always need to know where it's located, right?)
* "Who would ever come to our institutional repository and run a search on yada yadonics, my narrow topic? Get real." (To find something online, you always need to search locally, right?)
* "OA repositories are content ghettoes where content is difficult for users to find." (Online collections have impermeable walls, right?)
* "Nobody searches for work by the institutional affiliation of the author rather than by field or topic." (Institutional repositories are not as visible to search engines as disciplinary repositories, right?)
The misunderstanding here is that the only way to find work in a repository is to visit the repository and run a local search. This overlooks the interoperability of OAI-compliant repositories, which supports cross-archive searching by academic search engines like OAIster. It also overlooks the crawling of repositories, which supports cross-archiving searching by mainstream search engines like Google, Yahoo, and Microsoft.
It's like saying that the only way to find information on the front page of the New York Times or my personal home page is to visit the Times or my home page and run a search. People who believe this are the ones who need to be warned not to post embarrassing photos of themselves in chat rooms.
Because repositories are indexed in academic and non-academic search engines, users needn't know that a particular repository exists, where it is located, or what it contains.
There are two sorts of exception: Sometimes OA is deliberately turned off for a certain item, and sometimes crawlers overlook a repository or some of its contents. The latter is a mistake that befalls disciplinary repositories as often as institutional repositories, and both the repository and non-indexing search engine will want to fix it.
A related misunderstanding actually comes to the opposite conclusion. Scholars who believe that repository contents are only discovered by local searching and browsing sometimes worry about the juxtaposition of work on different topics. "If I self-archive to my institutional repository, then my work on protein polymers will belong to the same miscellaneous collection as work on postmodern poetry. Nobody will take it seriously!" It will be embarrassingly visible, as opposed to embarrassingly invisible. The mistake is to think that works on protein polymers and postmodern poetry will come up in the same search or browsing list just because they are on deposit in the same repository. That's like thinking that books in the same library must be on the same shelf.
(19) "OA invites plagiarism."
* "Putting my work out there will just invite ripoffs." (Nobody plagiarizes from TA books or journals, right?)
* "There's nothing to stop someone from plagiarizing from an OA source." (Nobody can detect plagiarism from an OA source, right?)
The real thought here seems to be that OA sources make cutting and pasting easier than TA sources do. Hence OA sources "invite" plagiarism from the plagiarism-inclined. If so, then the objection is really to digital scholarship at large, not just the OA subset of digital scholarship. Most TA journals have digital editions today, and the trend is to drop the print editions. There is far more digital TA journal literature than OA journal literature.
Many people have been correcting this misunderstanding for many years, but it still comes back. Here's one of my attempts from 2006:
http://www.earlham.edu/~peters/fos/newsletter/10-02-06.htm#qualityOA deters plagiarism. In the early days [!], some authors worried that OA would increase the incentive to plagiarize their work. But this worry made no sense and has not been borne out. On the contrary. OA might make plagiarism easier to commit, for people trolling for text to cut and paste. But for the same reason, OA makes plagiarism more hazardous to commit....[P]lagiarism from OA sources is the easiest kind to detect. Not all plagiarists are smart, of course, but the smart ones are steering clear of OA sources....Because OA will only reduce plagiarism by smart plagiarists, the effect may be small. And today the effect is small in any case because so little of the literature is OA. But just as we can expect good things from a pest-resistant strain of wheat, even when we've just introduced it in one field, we can expect good things from this plagiarism-resistant strain of research literature.
Some of the misunderstanding here may arise from confusing plagiarism and copyright infringement. But not only is OA lawful (see #15 and #16), plagiarism and infringement are separate, if overlapping, offenses. As I noted in SOAN for June 2007: "Someone can commit plagiarism without infringing copyright (by copying a fair-use excerpt and claiming it as one's own) and infringe copyright without committing plagiarism (by copying a larger excerpt but with attribution). One can also commit both together (by copying a large excerpt and claiming it as one's own)...."
(20) "Authors must choose between prestigious publication and OA."
* "If I had to choose, I'd prefer prestige to OA." (And unfortunately, I do have to choose, right?)
* "OA is good for research, but all the incentives in the system make scholars choose prestige instead." (You can't have a prestigious publication and OA at the same time, right?)
* "I like to support OA when I can, but I couldn't pass up the chance to publish in a very prestigious journal." (The journal is TA and that precludes OA, right?)
There are two reasons why OA is compatible with prestige: a gold reason and a green one. First, a growing number of OA journals have already earned high levels of prestige, and others are earning it. If there are no prestigious OA journals in your field, just wait. And do your part to move things along as an editor, referee, reader, and when you can, as an author, by submitting your best work to suitable OA journals. In the meantime consider the second reason. Most TA journals allow OA archiving; and when authors retain the right to self-archive, on their own or under a funder or university policy, then all journals willing to publish their work also allow self-archiving (see #16).
Scholars who consider the first reason but not the second, or who don't find a prestigious OA journal in their field and conclude that OA is out of the question for them, are falling victim to misunderstanding #1, "All OA is gold OA."
The current misunderstanding may be the most harmful one in the woods. When scholars know about OA and don't choose it, they are generally not opposed to it; many support it strongly. They are simply giving higher priority to prestige, as if they couldn't have both at once. But because OA is compatible with prestige, we rarely have to choose.
When *do* you have to choose? When a prestigious journal doesn't already permit postprint archiving and when it rejects your individualized request for permission. Note, though, that when NIH-funded researchers encounter that kind of publisher resistance, they must choose another publisher. They cannot get a waiver from the NIH. Knowing this, publishers who don't normally permit postprint archiving make exceptions for NIH-funded authors. The lesson: today most TA journals already allow postprint archiving, and when a critical mass of researchers and research institutions demand it, all will.
Because OA is compatible with prestige, we don't have to fight the lure of prestige, pretend it doesn't exist, or nobly rise above it. We rarely have to choose between them, but to have both at once we will often have to choose to self-archive.
(21) "OA helps readers but not authors."
* "Authors actually lose under OA." (Unlike OA journals, TA journals pay authors for their articles, right?)
* "OA offers readers free access, but has nothing to offer authors." (Authors don't gain from a larger audience and impact, right?)
This misunderstanding is common outside the academic habitat, where few people realize that scholarly journals don't pay authors for their articles (see #14). But it can also be found in some academic niches.
OA articles are accessible to everyone with an internet connection, a vastly larger audience than any scholarly journal can claim. Not all internet users will care to read your research, of course. But making your work universally accessible to the connected guarantees that it will be accessible to the subset which does care. If there's an exception for the digital divide, there's a larger exception for the non-digital or print divide.
Moreover, there's abundant evidence that OA articles are cited more often than non-OA articles, even more than non-OA articles from the same issues of the same journals. Many different studies have tackled this phenomenon, taking on different bodies of literature, using different methods, controlling different variables. They disagree on whether the OA impact advantage is large or small, and whether OA causes the increase in citations or is merely correlated with it. But they agree that OA articles are cited more often than non-OA articles.
The 350 year old custom under which scholars write journal articles without payment only works because scholarly authors get something else out of the deal. What they get is circulation of their ideas, or better, circulation with a time-stamp. They may hope to earn royalties from their books, but they write journal articles for impact, not for money. Journals began displacing books in the 17th century as the showcase for cutting-edge research because they could promise greater and quicker impact. The data now show OA journal literature has the same advantage over non-OA journal literature.
I commented on the OA-impact studies in 2006, when they were only half as numerous: "These studies bring a welcome note of self-interest to the case for OA. Providing OA to your own work is not an act of charity that only benefits others, or a sacrifice justified only by the greater good. It's not a sacrifice at all. It increases your visibility, retrievability, audience, usage, and citations. It's about career-building. For publishing scholars, it would be a bargain even if it were costly, difficult, and time-consuming."
(22) "All OA is gratis OA."
Gratis OA removes price barriers but not permission barriers. It makes content free of charge but not free of copyright or licensing restrictions. It gives users no more reuse rights than they already have through fair use or the local equivalent. Libre OA removes price barriers and at least some permission barriers. It loosens copyright and licensing restrictions and permits at least some uses beyond fair use.
* "Our journal is free of charge. What more could anyone want?" (Fair use covers all user needs, right? Fair use is the same the world over, right?)
* "Our journal is free of charge. Why does it also need an open license?" (Users only need to read, right? Users never need to know what they may and may not do with an article, right? All users are human, right?)
There is some excuse for the opposite view, that all OA is libre OA. The Budapest, Bethesda, and Berlin definitions of OA all describe forms of libre OA. However, there are good reasons to recognize gratis OA as a kind of OA --reasons I described in an article last year.
The current misunderstanding accepts that gratis OA is a kind of OA, but goes one step too far and assumes that gratis OA is the only kind of OA. The misunderstanding is that there is no libre OA, that libre OA adds nothing to gratis OA, or that what libre OA adds isn't necessary or desirable.
One habitat for this misunderstanding is the kind of OA journal which stops at gratis when it could easily move on to libre. In many of these journals, I don't detect opposition to libre OA so much as incognizance of libre OA. I don't mean journals which prefer CC-BY-NC to CC-BY, for example, since both are libre licenses. I mean OA journals without any libre licenses and without any apparent reason except that they never thought about it.
In general, OA repositories have good reasons to stick to gratis OA but OA journals don't. Repositories can't generate the needed permissions on their own, but journals can.
The misunderstanding is also sometimes found among free software veterans who understand the gratis/libre distinction and value libre. If the first examples of OA research or OA policies they see are gratis and not libre, it's easy to assume that OA is all about gratis.
(23) "OA is about providing access to lay readers."
* "The primary purpose of the NIH policy is to provide access to lay readers." (Its chosen term, "public access", refers to the lay public, right?)
* "OA to cutting edge research is unnecessary. Most lay readers don't care to read it and wouldn't understand it." (Professional researchers already have access to everything, right? See #25.)
OA provides access to everyone, professional and lay readers alike. The ratio of professional to lay readers of peer-reviewed research undoubtedly varies across the disciplines. But for the purpose of OA policies at universities and funding agencies, it doesn't matter what the ratio is in any discipline. Many professionals work at universities that can't afford access or outside universities. OA helps them. Many lay readers do care to read peer-reviewed research, can't afford subscriptions, and don't have access through public libraries. OA helps them. That's all we need to know.
If you don't believe that lay readers ever care to read peer-reviewed research, ever understand it, or ever benefit from it, and if you only have time to read one eye-opening testimonial, read Sharon Terry's.
Most institutions that actually draft policies are thinking of providing access to professional researchers first and lay readers second. But because the priorities don't matter much when everyone gets access, the distinction is seldom articulated. When the NIH spelled out the rationale for its policy, however, it put the beneficiaries in this order: "By creating an archive of peer-reviewed, NIH-funded research publications, NIH is helping health care providers, educators, and scientists to more readily exchange research results and the public to have greater access to health-related research publications."
OA directly benefits researchers, but it indirectly benefits everyone by benefiting researchers. Publisher criticism of OA policies on the ground that lay readers don't care is like criticism of the interstate highway system on the ground that most people don't drive more than 25 miles from home. Even if you rarely drive out of town yourself, you want highways for the people who deliver mail, apples, and fuel. In 2004, a Harris poll asked Americans what they thought about the idea of open access to publicly-funded medical research. One question asked whether "people" should have access to the research. Only 4% said no. You might imagine that those 4% were lay readers who really didn't care. (But even then, wouldn't you imagine a larger number?) Another question asked whether "doctors" should have access to the research. 3% said no. Those were probably not lay readers who didn't care; even lay readers who don't care to have access for themselves want it for their doctors. Those must have been publishers.
(24) "OA makes sense for second-rate work, but not for first-rate work."
* "This is really good. Don't think about OA. Send it to Nature." (Nature doesn't allow self-archiving, right?)
* "I don't need to self-archive this piece. It was published in Nature." (Nature already reaches everyone who could apply or build on my work, right?)
* "First-rate work doesn't need the alleged boost it would get from OA." (First-rate work doesn't need to reach everyone who could apply or build on it, right?)
This misunderstanding is the result of extensive cross-breeding among other misunderstandings. The idea behind it is this: the best work generally winds up in the best journals, where it has the best chance of being seen. At least it should be steered toward the best journals, where it will have the best chance of being seen. When we add the suggestion that this path doesn't allow OA, or that OA can't improve upon it, then an idea that was largely true becomes completely false. It assumes that the best journals are never OA (#5, #20); that only journals can deliver OA (#1); that the best journals never allow OA archiving (#16); and that OA archiving can't increase the visibility and impact of work published in the best journals (#21).
(25) "The current system is not broken."
* "If it's not broken, don't fix it." (It's not broken, right?)
* "All who need access to scholarly literature already have it." (All libraries subscribe to everything, right?)
* "I don't have access to all the literature I need, but I think I'm an exception." (My research interests are highly specialized, which is unusual, right?)
This misunderstanding is found in just one habitat: publisher rhetoric. It overlooks a mountain of evidence for dysfunction: journal prices rising faster than inflation since the 1970's; journal prices rising faster than library budgets for roughly the same period; journal cancellations at most libraries in most years, and often mass cancellations; bundling deals making it expensive for libraries to cancel second-rate or little-used journals; scholars giving their work to publishers without charge and then having to pay to access or copy it; publishers demanding exclusive rights to publicly-funded research; taxpayers paying for most scientific research three times, through research grants, researcher salaries, and journal subscriptions; publicly subsidizing research and researchers but allowing the results to be channeled through publishers who believe, correctly or incorrectly, that they must erect access barriers in order to recover their costs; allowing the accessible percentage of research to shrink as the volume of published research continues to grow; and cultivating artificial scarcity when the internet allows us to end information scarcity altogether. Even the evidence that OA articles are cited more often than TA articles (#21) is evidence that TA articles are not reaching all who need to see them.
Publishers who don't understand that the system is broken need to talk more with the libraries who subscribe to their journals, and to those who don't. In 2004 the University of California concluded that "[t]he economics of [subscription-based] scholarly journal publishing are incontrovertibly unsustainable." In 2006 the RIN concluded that "around 50% of all [UK] researchers, regardless of discipline, experience problems" accessing the journal literature they need. In 2008 the ARL concluded that "when it comes to libraries, journals are the most important item for faculty, and the source of their greatest dissatisfaction." Last month Virginia Tech announced that it must cancel $900,000 worth of journal subscriptions.
I once had the idea to limit the list to the 10 most common misunderstandings. Then I could redo the list every year to track which misunderstandings were on rise, which were declining, which new ones had sprung up, and which ones had retired. But I was too busy to make it an annual chore, and while I was busy doing the work which preempted it, I began to realize that we face essentially the same misunderstandings year after year.
I also thought that ranking the misunderstandings by frequency would not tell me all that I wanted to know. Some misunderstandings are not very harmful but depressingly common, and some are not very common but distressingly harmful
A useful ranking system would combine these two factors. For example, we could score misunderstandings for prevalence and harmfulness, each on a 10 point scale. Their combined score multiplies the two factors together. If you like, call this the Open Access Misunderstanding Index, or OAMI, pronounced "Oh my!"
prevalence (max 10) x harmfulness (max 10) = OAMI (max 100)
OA is making steady progress on every front, especially the launch of new journals, new repositories, and new policies. All this progress is both a cause and effect of improved understanding. So why do we face the same misunderstandings year after year?
That's a good question, and we can use the OAMI to shed some light on it. OA progress tends to reduce prevalence faster than harmfulness. For example, five years ago "OA bypasses peer review" (#4 above) had an OAMI of about 9 x 10 = 90. Today the same misunderstanding is about half that, 5 x 9 = 45. It's still near the max for harmfulness, but the prevalence coefficient is down considerably --a cause and effect of OA progress. We all still see this misunderstanding every year, but we see it less often. When it reappears, it does slightly less harm than in the past because many more people know it's a misunderstanding.
Unfortunately, OA progress tends to create endangered species but not extinctions. Or perhaps it's that these misunderstandings are such simple creatures that they can re-evolve independently of their past incarnations, and extinction is not forever.
In this field guide I decided to focus on variant plumage and false assumptions rather than OAMIs, which would have been subjective estimates at best. In the end, it's more important to help watchers identify the many species of bluff, blooper, blather, and bunkum in this richly diverse ecosystem than to rate them. If we learn to identify them, then we can put them on the endangered species list. And the more who can identify these misunderstandings, no matter their plumage, and no matter their prevalence, the less harm they will cause.
Read this issue online
SOAN is published and sponsored by the Scholarly Publishing and Academic Resources Coalition (SPARC).
Additional support is provided by Data Conversion Laboratory (DCL), experts in converting research documents to XML.
This is the SPARC Open Access Newsletter (ISSN 1546-7821), written by Peter Suber and published by SPARC. The views I express in this newsletter are my own and do not necessarily reflect those of SPARC or other sponsors.
To unsubscribe, send any message (from the subscribed address) to <SPARC-OANewsfirstname.lastname@example.org>.
Please feel free to forward any issue of the newsletter to interested colleagues. If you are reading a forwarded copy, you can subscribe by sending any message to <SPARC-OANewsemail@example.com>.
SPARC home page for the Open Access Newsletter and Open Access Forum
Newsletter, archived back issues
Forum, archived postings
Open Access Overview
Open Access News blog
SOAN is licensed under a Creative Commons Attribution 3.0 United States License.
Return to the Newsletter archive