Cites & Insights: Crawford at Large
ISSN 1534-0937
Libraries · Policy · Technology · Media

Selection from Cites & Insights 7, Number 8: July 2007

Library Access to Scholarship

First a few more bits of “opposition literature,” albeit less extreme than some discussed in C&I 7:4, April 2007. Marc H. Brodsky, executive director of the American Institute of Physics, offers “Fair and useful copyright: A primer” in the Professional Scholarly Publishing Bulletin for Fall 2006. Brodsky reminisces about filling out copyright-transfer forms as an author: “I probably thought it was for my benefit more than anyone else’s. While that turns out to be true, it is also an oversimplification…” I’ll agree “give us your copyright: it’s for your own good” is an oversimplification!

He offers to fill in “a few basic concepts all authors should understand because the future of their societies may depend on their decisions.” That’s not quite the same as protecting the author or improving scholarship as such: Now the society is the focus. Brodsky goes through some law and history, then returns to the society focus. Transfer of copyright from author to publisher is “a very positive ingredient for a scientific journal. It gives the society important freedoms of action available no other way.” [Emphasis added.] If authors retain copyright and license some nonexclusive rights to the publisher, that “would undermine the subscription value” of a journal—by making the scholarship more widely available, although that’s not the emphasis Brodsky wants to provide.

Brodsky asserts four “forces” driving OA—none of which seem to have anything to do with broader access to improve scholarship. No, in Brodsky’s world, OA is driven by ideologues “who feel ‘information should be free’”; by funding agencies who want to mandate “free access…without paying for the reviewing and editing costs incurred by publishers”; by libraries “whose budgets cannot keep up with the growth of research and the materials that they and their patrons want”; and by “technologies, which lower some of the barriers to entry for publishing and make it easier to post copies of almost everything.”

Brodsky clarifies that last point: he asserts that electronic publishing is more expensive than print publishing—print-specific costs total less than 15% of total production costs and “extra production costs for electronic-specific production, such as tagging and linking, more than eat up that 15%.” I find that hard to believe, although publishing-related costs can be as malleable as movie studio “costs.”

Did you notice the little trick in one of the other OA “forces” (none of which seem to have anything to do with improved access!): Library budgets can’t keep up with “research growth,” not “subscription price increases.” It gets worse: Brodsky imputes motives to OA advocates. “There are many who would like to see publishers of costly journals fail, and attacking copyright has become one element of a strategy towards that end.” Naturally, Brodsky tells authors that OA would not result in “wider promotion, dissemination and acceptance of their results” because, after all, it’s publisher branding, marketing and distribution that really count.

The piece ends with an AIP “Position on Open Access & Public Access.” It’s a remarkably “fearful” statement. Two (of four argumentative) bullets in full:

Ø    “AIP is fearful of and against government mandates that provides [sic] rules in favor of one business model over another.

Ø    “AIP is against funding agencies mandating free access to articles after they have undergone costly peer review or editing by publishers.”

The first two sentences of the last bullet: “AIP is also fearful about what government agencies might do with articles they receive under any deposit system. AIP is fearful of mission creep with government agencies using the deposited material beyond the goal of open access…” That’s three “fearful”s in a half-page statement!

The Spring 2007 issue of the same bulletin includes an editorial by Brian D. Crawford (whose comments also appear in C&I 7:4). I’ll refer you to Open access news for March 26, 2007 for the full statement and Peter Suber’s extensive commentary. Crawford tosses around the term “myth-slingers,” seems to attack Nature’s coverage of PSP’s PR decisions (without citing inaccuracies) and reduces arguments that government-funded research should be openly available to this: “The truth is that all this debate boils down to is some people wanting something for nothing—or claiming not to need to pay the tailor for making the suit, because they provided the starting fabric.” As expertly dissected by Suber, it appears Crawford is trying to take the advice of the PR man: “If the other side is on the defensive, it doesn’t matter if they can discredit your arguments.” Crawford says, “The hypocrisy is breathtaking.” I agree, but would suggest he’s looking in the mirror when he says that.

In May 2007, AAP/PSP, IASTM and ALPSP jointly released a position paper, “Author and Publisher Rights for Academic Use: An Appropriate Balance.” It’s an odd position paper, particularly if you believe “balance” means something other than 100% fidelity to one extreme. Of course there’s the claim that publishers need to hold copyright so they can “enforce copyright claims with respect to plagiarism and related ethical issues” (I’d love to see an accounting of such enforcement). We are told publishers “are in the business of making content available to the widest possible audience, provided they can do so in financially-viable fashion,” certainly not pricing as high as the market will bear in the interest of highest possible profit; exclusive rights are “critical to administering the scientific record.” Much is made of potential “waste” from deposit systems.

Key to this whole statement is that it focuses on researchers’ own reuse of their material, not broad access to that research. Here’s the balance statement:

* Academic research authors and their institutions should be able to use and post the content that such authors and institutions themselves provide…for internal institutional non-commercial research and education purposes; and

* Publishers should be able to determine when and how the official publication record occurs, and to derive the revenue benefit from the publication and open posting of the official record (the final published article), and its further distribution and access in recognition of the value of the services they provide.

There’s the balance. Authors and their institutions can use their submissions internally; publishers control everything else. It’s about as unbalanced a statement of “balance” as I’ve seen (and, of course, Peter Suber points out some of the problems on May 9, 2007).

Since this discussion of opposition follows from the earlier discussion of extremes, it’s worth noting a followup on “the other side.” I was unhappy with Dorothea Salo’s failure to denounce a tasteless analogy used by a major OA proponent, and said so. Salo responded with “Caught, and an apology, and thanks” at Caveat lector on March 23, 2007. She offers reasons why she was wrong—failing to read more deeply about the incident beyond Tom Scott’s reporting, as a result failing to spot how deeply wrong the analogy was—and, by failing to nail Richard Smith for descending to such tactics, hurting the “smart, articulate arguments for my side of the OA debate.” She also explains how it could happen, and I hear just what she’s saying. I want to thank Dorothea Salo for the public apology, for the clear explanation, for her consistently thoughtful and honest approach to OA—and, to be sure, for indirectly helping convince me to write On Being Wrong in C&I 7:6 (June 2007), a brief essay I’m particularly proud of.


For one purist approach to open access, it’s not about money at all—it’s only about fully open access to article-length scholarly literature. The most purist version says it’s not really OA unless people are free to do datamining on the text and republish text for profit (with attribution). So, for example, if Cites & Insights was scholarly (OK, stop laughing), it wouldn’t qualify as OA even though there’s no charge to the end user and redistribution is explicitly allowed under the Creative Commons license—because the CC license is BY-NC and “noncommercial” is too restrictive for one OA definition. And one camp of OA says library budgets aren’t an issue at all; we can have full OA while letting the big sci-tech publishers drain every dollar they can from library coffers.

That’s not my approach to OA. These pieces are called Library Access to Scholarship, not Open Access; I’m primarily interested in OA as a way to improve library abilities to provide access—not only to article-length scholarship but also to monographs and literature outside the “scholarly” circle. So I’m very interested in the money and a little impatient with the camp that says “oh, we don’t want to disturb current journal pricing; we just want full access.”

The problem with money and open access is that it’s so complex—and, in the case of actual publishing and distribution costs, so opaque. We regularly hear numbers cited for the cost of handling a scholarly article—but those numbers are pretty clearly based on dividing total revenue by number of articles, which is at best a misleading way to state costs. Total revenue for a journal includes profit (or the nonprofit equivalent), corporate or societal overhead and potentially a whole bunch of other things that have nothing to do with the actual costs of refereeing, editing, markup and dissemination (whether print or electronic).

Bill Hooker posted “Open question on open access” at Open reading frame on November 15, 2006 ( He quotes “Mark D” on this issue:

The problem is, I haven't seen any hard data that documents the cost of peer review, redaction, and publishing. Everyone throws numbers around as if they were confetti. We are all, supposedly (publishers and librarians) in the scientific/technical community, yet so very few people take a scientific approach to this issue.

The first step on the road to open access should be a review of the processes and costs associated with scientific publication. Sounds like a good paper for the library association journal. Any librarians out there that want to tackle this paper?

And as for the publishers, if they really do wish a dialogue, then why don't they reveal their redaction costs? Any takers out there in the publishing world?

Hooker follows that quote with this—which disagrees with Marc Brodsky’s claim near the top of this article:

Online publication dramatically lowers costs relative to printed journals, but it is not free. Copyediting is still required, peer review must be coordinated even though the actual reviewing is done by authors for no charge, and the digital objects (articles, data, etc) must be created, archived and maintained in an accessible format. There are surely other important costs, too, that do not occur to me right now. All of this costs money, but the Big Question of OA is: how much money?

Hooker cites a bunch of figures—but some of those figures, from commercial publishers, almost certainly represent desired revenue, not actual costs. As it stands, the question remains open: Just what does it cost to run an OA scholarly journal? There is no single answer, to be sure. The range offered is so wide as to be more infuriating than helpful—from PLoS’ $2,000 to $2,500 charges (why would second-tier PLoS journals be cheaper to process manuscripts for than first-tier ones?) to BioMed Central’s $1,000 to $1,800—and down to Hindawi’s $500 or so (Hindawi charges $60 to $120 per page). Hindawi claims to be profitable; PLoS is supposed to be nonprofit. In any case, informed comment requires more than raw numbers. For example, more than half of OA journals don’t charge “author-side” fees at all: What do their budgets look like? Informed comment requires analysis of where the money goes and whether the costs make sense. To do such analysis requires open access of a different sort—to the balance sheets in some depth.

Hooker concludes:

It would be very informative to see inside the finances of a variety of OA publishers. Knowing what publishers charge, as above, does not tell us what it actually costs to run the journals. Beyond saying “we are showing profit,” Hindawi does not seem to be forthcoming on that issue. I take it as read that for-profit ventures charge what the market will bear, but when the market in question is largely scientists and their allies (librarians, clinicians, &c.), it seems logical that the market should look for data on which to base decisions about just what it will bear. Commercial entities rarely have open-access balance sheets, but perhaps OA publishers could take the lead there as well?

Peter Suber comments in a November 17, 2006 Open access news post, noting that estimates of some publishers “test our credulity” (e.g., £30,000 or nearly $60,000 per article for Science) but also that different estimates count different aspects of the publishing process—and different publishers have different levels of overhead and efficiency. Suber also notes that OA publishing has (or may have) fewer expenses than non-OA publishing (e.g., no subscription management or marketing). He cites a 2002 study giving $400 as an average cost of peer review per published article; this is mostly the cost of facilitation and should be coming down as clerical tasks are automated.

Paying for green OA

A reminder: “green OA” is OA archiving—as opposed to “gold OA,” actual OA journal publishing. In the lead article in SPARC open access newsletter 108 (April 2, 2007), “Paying for green open access,” Peter Suber notes that “some publishers want to charge for OA archiving and at least one foundation is willing to pay for it.” He notes that this could slow green OA, “either by the direct imposition of new and needless costs or by confusing policy-makers about the economics of green OA.”

First the American Chemical Society (ACS) re-announced its hybrid journal program, AuthorChoice, and reminded us that authors who wish to self-archive must pay the AuthorChoice fee. Then Elsevier and the Howard Hughes Medical Institute (HHMI) agreed that when an HHMI-funded author publishes in an Elsevier journal, HHMI will pay Elsevier a fee to deposit the peer-reviewed postprint in PubMed Central six months after publication.

ACS lets authors deposit articles in independent repositories—but does not let them retain copyright, will not promise to reduce subscription prices in proportion to AuthorChoice uptake—and, the killer here, will charge the AuthorChoice fee “even to authors who want to self-archive.” Notably, ACS was never a “green publisher”—and about a week before the ACS announcement, Wiley announced a hybrid (that is, “author-pays” OA options within toll journals) program with the same effect. Wiley charges $3,000 for OA archiving, depositing the published version upon publication. ACS charges the same (seemingly high) fee but with discounts for ACS members and subscribing institutions.

At both publishers, these fees pay for gold OA, and I should make clear that I have no objection to charging for gold OA. On the contrary; if we are to have it, we must pay for it (through author-side publication fees, institutional subsidies, or some other way). However, I do object to charging for gold OA when authors only want green OA. It's like offering a car with a free bicycle to people who only want to buy a bicycle.

Green OA need not be the published version; it can be a preprint or the peer-reviewed but not copy-edited version. When ACS official Adam Chesler was asked explicitly whether the same fee would apply for self-archiving of the peer-reviewed manuscript (rather than the published version), he said yes. “Chesler's answer makes the ACS policy even worse than it seemed at first. It's bad enough to force authors to pay for gold OA in order to get green OA; at least they really get gold OA too, wanted or not. But under this new wrinkle in the policy, even self-archiving authors who don't get gold OA must pay for it.”

What about Howard Hughes Medical Institute (HHMI) and Elsevier? Part of Suber’s commentary:

The Howard Hughes Medical Institute (HHMI) is similar in many ways to the Wellcome Trust (WT), although it has kept a lower profile in the OA debates. The WT is the largest private funder of medical research in the UK, and the HHMI is, or was, the largest private funder of medical research in the US. (I haven't seen recent figures but HHMI might have been overtaken by the Gates Foundation.) HHMI agreed long ago to pay publication fees at fee-based OA journals, and may have been the first funder anywhere to do so. With PLoS, it convened the 2003 meeting that produced the Bethesda Statement on Open Access Publishing. Now it's about to adopt an OA archiving mandate for HHMI-funded research.

The WT OA mandate requires deposit in PubMed Central (or its UK equivalent, UK PMC) and the forthcoming HHMI mandate will do the same. However, Elsevier allows OA archiving only through the author's personal web site or institutional repository. That's why HHMI and Elsevier first sat down to talk.

If neither side revised its policies, then HHMI-funded authors would have to shun Elsevier journals and Elsevier journals would have to shun HHMI-funded authors. Both organizations would gain from a compromise. But unfortunately that fact alone didn't determine which side had to budge. In this case, it was HHMI and it caved. It's a shame because it had considerably more bargaining power than Elsevier…

HHMI will pay Elsevier $1,000 for each article published in a Cell Press journal and $1,500 for each article in any other Elsevier journal.

Suber discusses possible consequences if HHMI had not caved in. He also notes that Wellcome Trust signed a similar deal—but Wellcome Trust pays more ($3,000 and $5,000) and gets more—immediate OA to the published version instead of embargoed OA to an unedited version. For that matter, Cell Press seems to be expanding its embargo period as part of the deal. Here’s how Suber sees the HHMI-Elsevier deal: “I have to conclude that HHMI was ripped off. Or if that’s too negative, Elsevier got a fantastic deal.”

Part of Suber’s conclusion:

I'm not saying that the distinction between green and gold OA is immutable. Of course the two types can be blended. One blend—the best for researchers—would charge no fee to the depositor and provide immediate access to the published edition. Another blend—the worst for researchers—would charge a fee, impose an embargo, and limit access to the final version of the author's peer-reviewed manuscript.

The problem with the HHMI-Elsevier deal is not that it blurs the distinction between green and gold but that it needlessly adopts the worst blend for researchers. The problem with the ACS deal is that it artificially clamps green and gold together and forces anyone wanting green to pay for gold to get it.

Learned Publishing, April 2007

This issue includes a guest editorial and article related to costs, both of which can be read free from the website (in PDF form; look for “Learned Publishing” and go from there). The guest editorial, by Rick Anderson, is entitled “Open access—clear benefits, hidden costs.” It’s a tricky editorial, coming from a librarian but effectively undermining what one might assume to be library interests. It’s also a very strange editorial, as Peter Suber notes: How often is a guest editorial accompanied by a press release and a call for “supporting signatures”? Indeed, the editorial ends with a box stating the following:

ALPSP strongly supports the sentiments in the above Editorial; in particular we agree that, while open access to the scientific research literature may offer benefits to society, the true costs of a change of business model must be investigated.

We would like to encourage other organizations and individuals to show their support for this statement by adding their names to the list of signatories… The Washington DC Principles for Free Access to Science Coalition…is one of the first signatories.

The DC Principles Coalition is, I’m afraid, notorious for its misleading statements and adamant opposition to anything that might actually improve access to science (see C&I 5:1 and several more recent essays); ALPSP is also hardly known as an objective observer in this area.

Some of what Anderson has to say:

There is no question that OA offers potentially significant benefits to society. All other things being equal, free public access to scientific information is clearly a good thing. But all other things are never equal, and to know whether and to what degree any particular OA solution is really a good thing requires a calculation not simply of its benefits, but of its net benefits once costs are taken into account…

In the case of an OA journal, costs are most commonly borne by authors....

In fact, mandates that result in widespread and effective OA will inevitably drive at least some publishers out of business, whether or not such an effect is intended by those who promote OA.

…A solution that provides universal access without supporting publishers may be perfectly acceptable. [This stance assumes] that publishers add no value to the scholarly information chain, and can therefore be harmed with impunity and without concern for negative consequences to the scholarly community in general…

    In fact, most STM publishers are not profit-seeking corporations from outside the scholarly community, but rather learned societies and other non-profit entities, many of which rely on income from journal subscriptions to support their conferences, member services, and scholarly endeavours—as well as the peer-review and publishing activities that will remain important in a self-archiving environment. In other words, a publishing system that undermines the ability of publishers to make money in the marketplace thereby may also undermine scholars and scientists in their ability to do their work....

    In summary: OA offers real benefits to society. However, the net value of those benefits cannot be determined unless its costs are computed as well. The purpose of this statement is not to call on participants in the scholarly information chain to fight against OA, but only to move forward while taking full account of costs as well as benefits, and to work towards solutions that offer a net benefit to society....

There’s more (including raising the specter that diverting 0.5% of NIH’s budget to support author-side charges means a significant reduction in medical research). Readers of previous LAS sections or the better OA literature will have spotted problems right off the bat. Peter Suber took apart this editorial far better than I can. Some of his comments:

No serious OA proponent has ever said that it makes costs disappear. OA does shift costs, and some shifts are better than others. But OA does more than shift costs; it also reduces them…

In the case of an OA journal, costs are most commonly borne by authors....This is untrue and I'm surprised to see it asserted in an ALPSP journal with the unusually strong ALPSP endorsement represented by the call for signatures. For it was an ALPSP-sponsored study that showed that only a minority of OA journals charge author-side publication fees.

I’ll add here that there is absolutely no reason to believe that most author-side charges would actually be paid by authors…and I think Rick Anderson knows that.

The argument that OA archiving might not harm publishers has…been based on the evidence from physics, the field with the highest levels and longest history of OA archiving. Not only have the American Physical Society (APS) and the Institute of Physics Publishing Ltd (IOPP) seen no cancellations to date arising from OA archiving, they both host mirrors of arXiv, the premier OA archive for the field. (Now for my standard demurrer: while there's no evidence yet that high-volume OA archiving will kill subscriptions, it might really have this effect in some fields and, if it did, it would still be justified.)

…Speaking for myself, I've never denied that journals add value. To me the question is not whether a journal adds value but how to pay for the most essential kinds of added value without creating access barriers for readers.

Suber agrees that a “move forward” ought to take into account the full costs—and he and other OA advocates have been willing to make that case. “However, I doubt this will make the debate any easier to resolve than it has been up to now.”

Suber doesn’t take Anderson to task for the segment beginning “In fact, most.” I will and have already: While society publishers are neither the primary “villains” in the current publishing scheme nor the primary targets for OA, it is nonetheless unreasonable to claim that libraries should be propping up conferences, member services and other non-publishing aspects of societies (other than library societies). If subsidies are needed, they should come from the departments related to the society, not hidden in subscription prices. And, of course, tossing in peer review in this case is an indirect hint that OA might reduce or eliminate peer review, although Anderson doesn’t include this common canard as a direct statement.

What of the article—“The cost of journal publishing: a literature review and commentary” by Donald W. King? It’s long (22 pages), detailed and inconclusive. King distinguishes between price and cost; unfortunately, he also tends to use scare quotes around some OA claims (such as that subscription prices pose a “barrier” to access, a claim that seems self-evident enough to remove those quote marks). King knows his numbers and his research methodology (in this case primarily a literature review), and I recommend the article for those wishing to study these issues further—but I don’t have a lot to say about it here. I think it’s clear that you can’t use a single cost (e.g., the price of a “journal publication system”) as the basis for asserting proper OA costs per article. I think it’s also clear that some publishers will legitimately be far more efficient than others, and that at some point some levels of inefficiency may be insupportable.

Conclusion and Apology (of Sorts)

That’s less than half the source material in my LAS folder. There’s some fascinating stuff on institutional repositories and how they are (or aren’t) used. Peter Suber has done some long pieces on last year’s progress and this year’s probabilities that richly deserve excerpts and comments. FRPAA seems likely to be resurrected—and that leads to preemptive stuff like a Wired News article that appears balanced but seems to provide much more space to quotes from opponents than proponents.

Maybe I’ll get back to some of those other access elements in a later issue. Maybe they’ll become so aged that I’ll let them go. In either case, now is not the time. For those of you who want to follow OA in more depth, you already know the mantra. Read Peter Suber’s blog and newsletter. Read DigitalKoans. Read Caveat lector for real-world comments on repositories and the potential for library publishing. Read other OA blogs and resources.

This section is incomplete. It always will be. I’m just being more explicit than usual.

Cites & Insights: Crawford at Large, Volume 7, Number 8, Whole Issue 92, ISSN 1534-0937, a journal of libraries, policy, technology and media, is written and produced by Walt Crawford, a senior analyst at OCLC.

Cites & Insights is sponsored by YBP Library Services,

Opinions herein may not represent those of OCLC or YBP Library Services.

Comments should be sent to Comments specifically intended for publication should go to Cites & Insights: Crawford at Large is copyright © 2007 by Walt Crawford: Some rights reserved.

All original material in this work is licensed under the Creative Commons Attribution-NonCommercial License. To view a copy of this license, visit or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.