Publishing Costs

---

SUMMARY: OA IRs provide free supplemental copies of published, refereed journal articles. The way to access them is via a harvester/indexer. Direct searching of the IR is more relevant for (1) institution-internal record-keeping, (2) performance assessment, (3) CV-generation, (4) grant application and fulfillment, and perhaps also some window-shopping by prospective (5) faculty, (6) researchers or (7) students. The main purpose of depositing refereed journal articles is (8) so they are accessible to all would-be users, not just to those whose institutions happen to have a subscription to the journal in which they were published. That way (9) the usage and impact of the institutional research output is maximized (and so is (10) overall research progress).
   What content an IR accepts is an entirely different matter from what content an IR mandates. Harvard is mandating OA target content, which is the refereed, accepted final drafts of peer-reviewed journal articles. Harvard is not publishing its journals articles. The journals are publishing them, and providing the peer review and copy-editing. Harvard is merely providing supplementary access to the peer-reviewed final drafts for those would-be users who cannot afford access to the publishers published (and copy-edited) version.
   OA is needed for researcher (peer to peer) access. The lay public benefits indirectly from the enhanced research productivity, progress, impact and applications generated by OA, not from direct public access to esoteric, technical reports. But whatever is one's primary rationale -- public access or peer access -- the other comes with the OA territory.
   Are would-be users whose institutions cannot afford subscription access to the publisher's copy-edited version better off with a refereed final draft, not copy-edited, or are they better off without it? (The answer is obvious.)
   If and when Green OA self-archiving (of refereed, non-copy-edited final drafts of journal articles) and Green OA self-archiving mandates should ever make journal subscriptions unsustainable, author-institutions can pay for peer review by the article out of their windfall subscription cancellation savings on the Gold OA cost-recovery model. If they find it worth paying for too, the copy-editing service can be bundled with the peer review service.

---

[See also: No Such Thing As "Provostial Publishing": I]

---

On Fri, 30 May 2008, Sandy Thatcher wrote in liblicense:

ST: "[O]ne is not likely to start with an IR to find the most important work in a discipline, unless one happens to follow the work of a particular scholar, in which case one would likely go to the scholar's own web site first, not the IR."

OA IRs provide free supplemental copies of published, refereed journal articles. The best and most likely way to find and access them is via a harvester/indexer that links to the item, not by directly searching the IR itself. (Direct searching of the IR is more relevant for (1) institution-internal record-keeping, (2) performance assessment, (3) CV-generation, (4) grant application/fulfillment, and perhaps also some window-shopping by prospective (5) faculty, (6) researchers or (7) students.)

The main purpose of depositing refereed journal articles is (8) to maximize their accessibility, so they are accessible to all would-be users, not just to those whose institutions happen to have a subscription to the journal in which they were published. That way (9) the usage and impact of the institutional research output is maximized (and so is (10) overall research progress).

ST: "But I do continue to question what the institution gains from its IR."

It seems to me that (1) to (10) above is quite a list of institutional gains from their IRs.

ST: "Does Harvard really need, or will it gain, any more "prestige" by having its faculty's work deposited there?"

Prestige is only a small part of it. All universities want and need (1) - (10) (and not all universities are Harvard for prestige either!).

ST: "It seems equally likely that it will lose some respect if too many scholars post articles that are first drafts or occasional pieces that would never appear in any peer-reviewed forum. It could easily become a grab bag of miscellany that will not reflect well on Harvard's presumed reputation for quality."

What content an IR accepts is an entirely different matter from what content an IR mandates. Harvard is mandating OA target content, which is the refereed, accepted final drafts of peer-reviewed journal articles.

It does not seem to have done the prestige of high energy physics any harm to have been (for over a decade and a half) self-archiving their pre-refereeing preprints too, even before their refereed, final drafts. But that is an individual and disciplinary choice, not one that the Harvard mandate is making for its faculty, and not the primary purpose of an OA IR.

ST: "Harvard authors, on the whole, are no better writers than scholars elsewhere, I would suggest, and their unedited prose will not do any good for the institution."

That may or may not be a good argument against depositing unrefereed preprints, but it has nothing to do with OA, OA mandates, or the primary purpose of OA IRs.

ST: "And, as for the general public, what members of that public are really going to bother spending their time pouring over esoteric scholarship when they can go to Wikipedia to get the information they need? This seems to me as false an assumption as the expectation that somehow members of the public are going to benefit greatly from reading the technical articles posted on PubMed Central under the new NIH program."

I think you are quite right about that, Sandy. I myself have been somewhat uncomfortable all along with the heavy emphasis that some of the advocacy for OA has placed on the putative need of the general tax-paying public for access to peer-reviewed research journal articles. The need is there for health-related research, and perhaps a few other areas, but I have always felt it weakens rather than strengthens the case for OA to argue that its primary purpose and urgency is for the sake of lay public access. It is not. OA is needed for researcher (peer to peer) access. The lay public benefits indirectly from the enhanced research productivity, progress, impact and applications generated by OA, not from direct public access to esoteric, technical reports.

However, I have to admit that my worries may have been misplaced, because the Alliance for Taxpayer Access (which I strongly support) managed to get through to a lot more people (lay public, academics and congressmen) with the message of taxpayer access than they would have done if they had stressed only peer-to-peer access. So they perhaps overgeneralized the special case of health-related research, suggesting that it was a matter of similar urgent public interest in all public-supported research. In reality, it is a matter of similarly urgent interest for the public -- but an interest in maximizing research progress and applications, by making research openly accessible to its real intended users, namely, other researchers. That is the way the tax-paying public maximizes its benefits from the research it supports -- not by reading it for themselves!

This public access argument, however, applies more to the public funding of research grants (such as those of NIH) than to university research access policy. In the university case, it is clear that the objectives of OA mandates are mostly peer-to-peer (or, if you like, university-to-university access) rather than a burning need for public access.

It is to be noted, though, that whatever is one's primary rationale -- public access or peer access -- the other comes with the OA territory. So the outcome is exactly the same.

Harnad, S. (2007) Ethics of Open Access to Biomedical Research: Just a Special Case of Ethics of Open Access to Research. Philosophy, Ethics and Humanities in Medicine, 2 (31).

ST: "I imagine that very few members of the public are going to be able to understand the vast majority of these articles, let alone derive any useful lessons for life from them. There seems to be a general fantasy that the whole world is somehow waiting breathlessly for access to all this highly specialized knowledge. I speak as director of a press that has a hard time selling books that we think to be of "general interest," compared with our monographs. The audience just isn't there, folks! And institutions that believe their reputations are going to soar because of what their faculty post on their IRs are just kidding themselves."

I agree with most of this, as you see.

(But be careful, Sandy! Some might be tempted to say you have a hard time selling books, but you might have less of a hard time giving them away free (OA)! As you know, I do not make this argument for books in general, because I know that there are true and unavoidable production costs to cover there -- whereas with Green OA IRs, journal publication reduces to just the costs of implementing peer review alone. Moreover, book authors are more inclined to seek royalties than to pay production costs and give their books away free. But for some esoteric monographs, an online-only OA edition, with costs covered by the author's institution, may very well turn out to be the optimal model, much as it will probably end up for journal article publication.)

Harnad, S., Varian, H. & Parks, R. (2000) Academic publishing in the online era: What Will Be For-Fee And What Will Be For-Free? Culture Machine 2 (Online Journal)

Ceterum Censeo: The notion of "provostial publishing" is utter nonsense:

No Such Thing As "Provostial Publishing": I.

ST: "On what basis do you [claim journal copy-editing is minimal]? Have you surveyed journals to find out how much copyediting they do? Are you basing this on your own personal experience with copyediting done by the journals to which you have submitted your own work primarily?"

It is based on the CUP journal I edited for 25 years, plus many other journals for which I have refereed and in which I have published. It is also a judgment shared by many not only in the research community but also in the publishing community: Journal article copy-editing today is down to which-hunting and reference-querying, for the most part. Software can do the latter and we can do without the former, I think.

ST: "I, of course, cannot claim sufficiently wide knowledge to make sweeping generalizations about the degree and level of copyediting done for journals compared with books at all publishing houses. But as director of a press that publishes 11 journals in the humanities, and a past employee of another press that published three (including one in mathematics), I can attest that the copyediting done for these journals is at the same level as done for books, which in university presses is pretty high. I suspect that other university presses operate in this respect the same way we do--which would mean that at least 1,000 scholarly journals get far more than 'minimal copy-editing'."

That may well be. But there are only two underlying issues here, so it's best not forget them:

(1) Are would-be users whose institutions cannot afford subscription access to the publisher's copy-edited version better off with a refereed final draft, not copy-edited, or are they better off without it? (The answer is obvious, I think.)

(2) If and when Green OA self-archiving (of refereed, non-copy-edited final drafts of journal articles) and Green OA self-archiving mandates should ever make journal subscriptions unsustainable, author-institutions can pay for peer review by the article out of their windfall subscription cancellation savings on the Gold OA cost-recovery model. If they find it worth paying for too, the copy-editing service can be bundled with the peer review service.

Nothing hangs on either of these things, and all other questions (such as how much copy-editing is really being done on journal articles today) are irrelevant to the Green OA and Green OA mandate issue.

ST: "I can also attest, from my own years of experience as a copyeditor, that the job does not just involve polishing prose and improving grammar. Not uncommonly, copyeditors will find and correct egregious factual and other errors, thus sparing the authors from considerable embarrassment. Without their "value added" services, much will get published in Green OA form that will NOT serve either the authors' peers or the general public well."

Right now, that value-added is being paid for by subscriptions. If subscriptions ever vanish, and the value is still desired, it can be paid for along with peer review on the Gold OA cost-recovery model.

ST: "Hence, I conclude, Harvard and others that follow its example and are content to publish less than the final archival version will be opening themselves to the exposure of all the flaws of scholarly writing that now get hidden from public view by the repair work done by copyeditors. Caveat lector!"

Harvard is not publishing its journals articles. The journals are publishing them, and providing the peer review and copy-editing. Harvard is merely providing supplementary access to the peer-reviewed final drafts for those would-be users who cannot afford access to the publishers published (and copy-edited) version. To complete the circle again, please refer to (1) above...

Stevan Harnad
American Scientist Open Access Forum

---

SUMMARY: Talat Chaudhri and I agree that Green OA via self-archiving is feasible and desirable, and that publication will eventually consist of peer review alone. The only points of disagreement are about how to get there from here. I advocate Green OA mandates, whereas Talat advocates direct transition to peer-review-only, administered by university consortia. What Talat does not explain, however, is how we are to get the 25,000 journal titles that are currently being published by their current publishers to migrate to (or be replaced by) such consortia. Nor does Talat explain how the consortia's true peer-review expenses would be paid for, even if the 25K journals titles did miraculously migrate to such consortia of their own accord (although the answer even then is obvious: via Gold OA author-institution fees, paid out of their subscription cancellation savings).

---

Talat Chaudhri, Repository Advisor, Aberystwyth University, wrote (in the American Scientist Open Access Forum):

TC: "Gold OA [1] isn't popular and, I suspect, [2] never will be."

Talat Chaudhri is right about the first point [1], and the reason is partly the current price of Gold OA and partly the fact that Gold OA is not yet necessary, because Green OA (self-archiving) can provide OA.

But whether the price of Gold OA once it amounts to nothing more than the cost of peer review will be "popular" [2] -- if and when it becomes a necessity (i.e., if and when universal Green OA makes subscriptions unsustainable) -- is not a matter of either popularity or suspicion: As long as peer review is necessary, paying the true costs of implementing it will be necessary, if one wants to publish (peer-reviewed research) at all.

The good news is that the cost per paper of peer review alone then will be far less than the cost per paper of (subscription) publishing now, and the subscription cancellations will release many times the amount of money needed to pay for peer review alone.

For the perplexed reader: Talat and I are not disagreeing on most of these points. We both agree that Green OA via self-archiving is feasible and desirable, and that publication will eventually consist of peer review alone.

The only points of disagreement are about how to get there from here.

I advocate Green OA mandates, whereas Talat advocates direct transition to peer-review-only, administered by university consortia.

What Talat does not explain, however, is how we are to get the 25,000 journal titles that are currently being published by their current publishers to migrate to (or be replaced by) such consortia.

Nor does Talat explain how the consortia's true peer-review expenses would be paid for, even if the 25K journals titles did, mirabile dictu, migrate to such consortia of their own accord (although the answer even then is obvious: via Gold OA author-institution fees, paid out of their subscription cancellation savings).

But apart from not wanting to call this sort of payment "Gold OA" (even though that's exactly what it is!), Talat also does not like Green OA mandates.

The trouble is that Talat has no better way -- nor even an equally good alternative way -- to get the 2.5 million articles published annually in the 2.5K journals to "migrate" to their authors' Green OA IRs -- any more than he has a way of getting the university peer review consortia created, or of getting the journal titles to migrate over to (or be replaced by) them.

So let's focus on the substantive points of agreement: (1) Universal Green OA and (2) publishing costs reduced to just peer review alone. Talat can leave the problem of generating that Green OA to the Green OA mandates, and he can call the funding of the peer review something other than "Gold OA" if he likes -- it all comes to the same thing anyway...

TC: "On "downsizing" to Gold OA, I'm afraid that I agree with the original point in the [Letter to Nature] to which N. Miradon posted a link recently. The developing world doesn't want it."

Reply: Downsizing is for publishers (not researchers) to do, under Green OA cancellation pressure. The only thing the developing world need do is to provide Green OA to its own article output by self-archiving the accepted, refereed final drafts (postprints) in their Institutional Repositories (IRs) (which is exactly the same thing that the developed world needs to do).

TC: "Neither, I submit, does anybody in the developed world want to pay for it."

They needn't. They need only mandate and provide Green OA. The rest will take care of itself. Institutions are already paying for publication (via subscriptions). When they no longer have to pay for subscriptions by the incoming journal, institutions' savings will be more than enough to pay for peer review by the outgoing article instead.

TC: "In terms of diverting currently subscription funds progressively to OA, any librarian such as myself will tell you that getting management agreement for what looks to them like a hypothetical new publishing model is going to be complex and very possibly unworkable, leaving only the few universities that have created funds for the purpose. None to my knowledge has agreed to allocate money on a yearly basis, as the costs are currently unknown."

But I have not said anything whatsoever about libraries needing to progressively divert subscription funds to OA.

I said universities and funders should mandate and provide OA (as 44 universities and funders, including Southampton and Harvard, ERC and NIH have done) and that IF and WHEN that should ever make subscriptions unsustainable (i.e., they are all cancelled), THEN a small portion of their windfall institutional savings can and will be redirected to pay for peer review.

No one is asking libraries to divert anything anywhere now, instantaneously or progressively. (If and when the time of universal, unsustainable cancellations comes, Necessity will be the Mother of Invention. No need to speculate or counterspeculate about it in our imaginations now, pre-emptively; let's just concentrate on mandating and providing universal Green OA.)

TC: "Why will Gold OA not catch on? Because it is unjust! Only those academics whose institutions can afford to pay will be able to publish, unlike the present situation where anybody can."

You are talking about Gold OA now, Talat, at current asking prices, and I agree.

So focus instead on mandating and providing Green OA for now, and worry about the question of converting to Gold OA if and when it becomes an actual matter of necessity, not just a hypothetical matter of possibility. (By that time the asking price will be so low, and the cancellation savings so high, that the decision will become a no-brainer.)

TC: "As I am presently a librarian, not an academic, I would be very likely unable to publish in my field of research on the basis of these centrally allocated funds, like retired academics and those in the developing world. Nobody will want this model, quite simply. They don't want it now!"

To repeat, you are thinking of Gold OA today, at today's asking prices, while all the money that can potentially pay for it is still tied up in paying the subscriptions. This is unilluminating and irrelevant: Forget about Gold OA if you wish. Publish wherever you like, self-archive your postprint, and let nature take care of the rest,

(Once Green OA is universal and only peer review needs to be paid for, the cost will be low enough so these needless hypothetical worries will look risible. And provisions for the minority of researchers who are retired, institutionally affiliated or otherwise unable to pay the low costs of peer review will be made. We don't need to retain the present access-denial juggernaut in order to take care of that small minority of special cases.)

TC: "[T]here was no "plausible path" for print to electronic publishing, yet it happened. If people as well placed as yourself were advocating... [university peer-review consortia] I am sure it might have a strong chance of catching on."

(1) If I were anywhere near as "well placed" as you imagine I am, dear Talat, we would have had 100% Green OA a decade and a half ago.

(2) Electronic publishing did not face the regenerating heads of the 34-headed monster responsible for the "Zeno's Paralysis" that besets Green OA -- a syndrome to which you, Talat, are alas not immune either:

#1. Preservation
#7. Peer review
#9. Downsizing
#10. Copyright
#17. Publishers
#18. Libraries
#30. Redirecting Cancellation Savings
#32. Poisoned Apple

The only effective medication, apparently, is Green OA mandates, and, luckily, relief is on the way:

Harnad, S. (2006) Opening Access by Overcoming Zeno's Paralysis, in Jacobs, N., Eds. Open Access: Key Strategic, Technical and Economic Aspects. Chandos.

TC: "...'relieving' journals of costs also "relieves" them of profits, which they won't want. It's myopic, to use your word, to suggest that this won't cause problems fairly soon."

Fairly soon? Self-archiving at 100% levels in high energy physics has not yet begun to be felt in cancellations in 17 years (both APS and IOP have confirmed this publicly: see the publications of Alma Swan). I am optimistic about mandates, but not so optimistic that I think that 100% Green OA will be with us "fairly soon." So how can a library cancel a journal while only an unknown percentage of articles from an unknown number of mandating universities are being self-archived?

I can only repeat: It would do a lot more good if we self-archived (and supported self-archiving mandates) more and speculated about the future of publishing (or libraries) less...

TC: "[Publishers currently endorse OA self-archiving] under licence which they remain free to withdraw, if that should be in their interests. Don't fool yourself that they couldn't if need be. At present it doesn't serve publishers to do so, so they don't. This is no basis on which to plan."

I cannot fathom, Talat, why you would prefer to keep speculating about whether and when publishers might withdraw their Green light to self-archive instead of pressing on with self-archiving and self-archiving mandates while the going is Green. This is one of the 34 familiar symptoms of Zeno's Paralysis, and it's been with us for years now:

#32. Poisoned Apple
I worry about self-archiving even if the journal gives me the green light to do so, because if I do, the light may change to red."

The answer to your question is that as Green OA grows, the risk to publishers is less that of losing subscribers, but that of losing authors. And losing authors would certainly accelerate cancellations a lot faster than the anarchic growth of Green OA self-archiving will. (Losing Harvard authors today would be bad enough, but it would only get worse, if the publishers' response to OA mandates were to try to revert to Gray instead of Green.)

The only thing you need to "plan" today is how to facilitate the provision of Green OA.

TC: "I happen to believe that nobody wants Gold OA in the future, as they don't appear to want it now."

Most researchers don't want Gold OA now because the money to pay for it is tied up in subscriptions and the asking price is way too high. What they want now is OA, and Green OA mandates will see to it that they get (and give) it.

If and when the subscription funds are released, the price drops, and there is no other way to publish, researchers will want Gold OA.

But why keep speculating about if and when? Green OA is within reach, and all it needs is more and more Green OA mandates.

TC: "Arts departments have not co-operated with the Green OA revolution, as has recently been brought home to me here by our English Department."

No? It seems to me that the Harvard Faculty of Arts and Science voted unanimously for Harvard's Green OA mandate. All the other universities that have mandates have English Departments too.

TC: "This is because we haven't understood their needs and continue to talk only about the most recent cutting edge science departments. Arts subjects are much more concerned with what you dismiss as "legacy" literature, preservation, book publishing, without which OA means little to them. We have sought no answers for any of these areas and so have no solutions for these academics."

OA's primary targets are journal articles; Green OA mandates only mandate the self-archiving of journal articles. The Arts and Humanities disciplines are more book-intensive than journal-intensive compared to the Physical, Biological and Social Sciences. But inasmuch as they publish in journals at all, no discipline is indifferent to the usage and impact of its journal articles. So:

(1) Preservation is important (but no more relevant to OA than non-OA).

(2) Book-publishing, inasmuch as it is not an author give-away literature, the way refereed journal articles are, is (so far) irrelevant.

(3) The legacy literature will have its turn once the current and forward-going literature is OA. (World hunger, the cure of all diseases and the righting of all wrongs and injustice will have to wait too, I am afraid: We're just talking about righting one wrong here: needless and obsolete access-denial for refereed research journal articles in the online age.)

TC: "...repository managers, their libraries and therefore their institutions... may not be so eager to follow your predictions as you hope, given that you have such a poor view of their "legacy" holdings and given the comments I have made on the failure to address the needs of all disciplines. I'm not sure we even have a solution for sciences and social sciences."

OA IRs are not preservation archives, they are access-provision archives. And the access is to their own institutional research output, not to the licensed subscription content their libraries have bought in from other institutions. (There is some fundamental underlying confusion here, or a conflation of two agendas, only one of which is OA. And none of this has anything to do with discipline differences.To my knowledge, no significant discipline differences have been reported across all the disciplines tested, either for the size of the OA impact advantage, or for the willingness to comply with OA mandates.)

TC: "as I have agreed, we are stuck with the necessity for mandates asap. [But] publishers and universities alike need to find different funding models now, ahead of time, before conflict arises with the publishers, as it must inevitably do if we simply eyeball them from the trenches waving our mandates."

I'm glad we agree on the necessity of mandates, Talat (although only one of us regrets that we are "stuck" with them).

But we will have to agree to disagree on the advance need to find different funding models. Or rather, I would say we have found already a different funding model (Gold: author-institutions pays for publication output instead of user-institution paying for publication input), but its time has not yet come: Universally mandated OA is needed first, to pave the way.

TC: "The relationship between library (i.e. those who acquire both resource and locus of deposit) and researcher is key to the solution, as any good subject librarian will tell you. I fear that you don't understand how libraries form a key part of the way in which the institutions who they serve, and who you mention as players in this, actually change policy in the interests of the researchers. This is the main point of contact in the institution."

Talat, is the role of libraries and librarians in the transition to free online access to research journal articles really that apparent? Did good subject librarians know where we're all heading a-priori, even before the online era?

OA is largely a matter for the research community: They are the providers as well as the users. But unlike with books, which need to be bought in and collected by their libraries, it is not at all obvious that OA requires library mediation.

It is fine if the library is made the manager of the IR, but then let the task be taken on as the radically new task it really is, rather than forced into the Procrustean Bed of what the library's traditional expertise and functions have been. OA is new territory, requiring new, Open mind-sets, not "what any good subject librarian already knows"...

TC: "I would instead hope to hear direct answers to the points raised, as well as a reasoned argument against "consortia" journals rather than merely waving them aside as a foolish repository manager's fancy."

You've had direct answers, Talat: Self-archiving mandates have already been tested and demonstrated to generate Green OA, they are feasible, they are growing, and they scale (to all universities and all funders).

What has been tested and demonstrated with (1) generating peer-review consortia and (2) getting journal titles and authors to migrate to them?

It does not help to repeat an N of 1 (the Board of Celtic Studies). There are over 3000 Gold OA journals, and you yourself doubt they will scale...

TC: "Sadly it appears this point has been side-stepped deliberately, as I confess that I anticipated. Core Green OA forecasts, however speculative, are to be supported. Others are to be rejected as mere speculations, a double standard."

No double standard. I have left no substantive point unanswered.

TC: "from the academics' expressed point of view... [self-archiving] is a new obligation that impinges on, as they see it, what they do with their copyright. Hence it looks like coercion. Taking heed of this reaction is the only way to get true co-operation."

Arthur's Sale's studies (and the continuing evidence since) keep confirming that authors willingly comply with self-archiving mandates. The latest two mandates (from Harvard) have been unanimously voted in by the academics themselves.

TC: "I'm not in a position to [university peer-review consortia]. But I suggest that someone whose advocacy on the subject will be heard, such as yourself, might be in a position to popularise the idea speedily, if you wished to put your efforts into it."

I did put my efforts behind alternative publishing models, a decade and a half ago, when I thought the problem was in the publishing model. Crashing failure made me realize that the problem was not in the publishing model but in academics' heads (Zeno's Paralysis). And the tried and tested cure is Green OA mandates, and they are happening. So why should we go back to old, far-fetched, and discarded hypotheses?

TC: "Simply, what we don't have is an answer to how peer review, copy editing and so forth will actually be provided after the Green OA revolution."

What we need now is not an answer to that question! What we need now is universal Green OA!

TC: "If there is no way forward, the revolution cannot happen."

There is a way forward: Green OA self-archiving mandates, by universities and funders, and they are happening.

TC: "I support Green OA but I do not believe at all that it will, or should, lead to Gold OA."

Fine. Let it just lead to Green OA!

TC: "There is no natural progression in this whatsoever, as nobody wants Gold OA anyway."

Fine. It is OA that research needs, not necessarily Gold OA.

TC: "If you destroy the publishers, as you suggest, who will then do the peer review?"

Where did I ever suggest "destroying the publishers"? (The talk of impending destruction, catastrophe and doom has come from speculators (and mostly by those with vested interests in the status quo, such as publishers, but sometimes also, for different reasons, librarians).

And peer review will continue to be done (for free) by the peers who review, regardless of who is paying to implement the process, or how.

TC: "All this talk about costs is a whitewash: they are relatively insignificant anyway compared to the research process. Universities could easily shoulder them, especially given savings from subscriptions, which are exorbitant."

You sound like you are saying the same thing I am now. So why are we talking about costs, and who will pay them, and when, when the urgent issue is Green OA, and getting it mandated so that we have it, at long last, now?

TC: "Find a solution to the future source of peer review (that isn't merely Gold OA) and you solve the whole thing. This peer review problem is all that is holding back Green OA. Forget Gold OA, it simply isn't part of the solution."

No, it is Zeno's Paralysis that is holding back Green OA, and one of the symptoms of Zeno's Paralysis is fretting needlessly about who is going to pay for peer review if and when it is no longer being paid by subscriptions.

But the OA problem is not "who is going to pay for peer review if and when it is no longer being paid by subscriptions." It is the research access and impact that continues to be lost daily, as we sit counterfactually speculating, day in and day out, about the future of publishing instead of the present of research.

And the solution is at researcher' fingertips: They just need to do the keystrokes that will get their articles into their Green OA IRs.

And the cure for what is holding back researchers' fingers is Green OA mandates.

You have created an IR, Talat, but we know that's not enough. You now have to help fill it, and your scepticism about university mandates and preference for conjectures about university peer-review consortia certainly is not helping to fill it.

Stevan Harnad
American Scientist Open Access Forum

"Disseminating research via the web is appealing, but it lacks journals' peer-review quality filter," says Philip Altbach in: Hidden cost of open access Times Higher Education Supplement 5 June 2008

Professor Altbach's essay in the Times Higher Education Supplement is based on a breath-takingly fundamental misunderstanding of both Open Access (OA) and OA mandates like Harvard's: The content that is the target of the OA movement is peer-reviewed journal articles, not unrefereed manuscripts. Professor Altbach seems somehow to have confused OA with Wikipedia.

It is the author's peer-reviewed final drafts of their just-published journal articles that Harvard and 43 other institutions and research funders worldwide have required to be deposited in their institutional repositories. This is a natural online-era extension of institutions' publish or perish policy, adopted in order to maximise the usage and impact of their peer-reviewed research output.

The journal's (and author's) name and track record continue to be the indicators of quality, as they always were. The peers (researchers themselves) continue to review journal submissions (for free) as they always did.

The only thing that changes with OA is that all would-be users webwide -- rather than only those whose institutions can afford to subscribe to the journal in which it was published -- can access, use, apply, build upon and cite each published, peer-reviewed research finding, thereby maximising its "impact factor." (This also makes usage and citation metrics Open Access, putting impact analysis into the hands of the research community itself rather than just for-profit companies.)

And if and when mandated OA should ever make subscriptions unsustainable as the means of covering the costs of peer review, journals will simply charge institutions directly for the peer-reviewing of their research output, by the article, instead of charging them indirectly for access to the research output of other institutions, by the journal, as most do now. The institutional windfall subscription savings will be more than enough to pay the peer review costs several times over.

What is needed is more careful thought and understanding of what OA actually is, what it is for, and how it works, rather than uninformed non sequiturs such as those in the essay in question.

Stevan Harnad
American Scientist Open Access Forum

---

SUMMARY: There are many valid arguments for Open Access (OA), but the claim that researchers or their universities are currently "double-paying" to "buy back" access to their own research output -- by (1) paying for the conduct of the research, giving it to journals, and then (2) paying for subscriptions to access it -- is invalid. Researchers already have their own results, and so does their own university. What their university library subscriptions are paying for is to buy-in (not buy-back) the research output of other universities, exactly as with books written by authors from other universities. (Some research, universities and university libraries are funded by public funds, some not. Unless OA is to be limited to only that subset, that is not a sufficient rationale for OA either.)
   There is a valid "double-payment" argument, though, when it is not the payer (the researcher, funder, or university) paying twice, but the payee -- the publisher -- getting paid twice (not necessarily by the same payer): This is the "Trojan Horse" of paying a "hybrid Gold OA" publisher (i.e., not a pure Gold OA publisher but a subscription-based publisher who offers the option of making individual articles OA in exchange for a fee) at a time when the potential funds for doing so are still tied up in the university subscriptions that are already paying that publisher's costs in full.
   Publishers need only be paid once. If mandated Green OA should ever make cost-recovery from subscription payments unsustainable (because it makes university subscription demand disappear), then the resultant university subscription windfall savings can be redirected to pay for the peer review on the Gold OA model, with all access-provision and archiving off-loaded onto the distributed network of university OA Repositories.

---

There are many valid arguments for Open Access (OA): It maximizes research usage and impact, it enhances research and university visibility as well as researcher and university income, it accelerates national and international R&D productivity and progress, it helps developing countries, it helps inform the general public, it is useful for education and training, it might eventually reduce university library subscription costs.

But the widespread claim that researchers or their universities are currently "double-paying" to "buy back" access to their own research output -- by (1) paying for the conduct of the research, giving it to journals, and then (2) paying for subscriptions to access it -- is both invalid and unnecessary.

If you ask those who are making this argument exactly what they mean, you will get two different answers, neither of them coherent or defensible:

(1) "I am salaried, and government-funded, to do my research. I give the report of my research findings to my publisher for free. Then I or my university must buy a subscription to the journal in order to access my results."

This is incorrect. You already have your own research results, and so does your university. What your university library subscriptions are paying for is to buy in the research output of other universities. And the ones to whom your publisher is selling your research are other universities, not your own.

This is no more "buy-back" or "double-paying" than it is to pay for the books written by authors from other universities. Or any other output that the university both itself generates and also buys in from other universities.

Nor does the fact that books sometimes generate royalty revenue change this picture: Most books generate negligible royalty revenue or none at all.

(It is relevant however -- not as a buy-back/double-pay argument, but as a conflict of interest -- that researchers, unlike other authors, ask for no royalty income from their publisher, and publish their articles purely for the sake of their research impact, not their royalty income. Hence this royalty-free author give-away writing is not a "work for hire," and publishers are unjustified in trying to prevent or embargo its authors' attempts to maximize access to it by self-archiving it free for all online.)

(2) "My research is funded by public funds; so is my university, and so is my university's library. Hence it is somehow buy-back/double-pay for them to be paying for subscription access to it too."

This no longer sounds like buy-back/double-pay but something much more complicated, with some grains of truth. But those grains of truth are only being buried or distorted in calling this buy-back/double-pay. Not only is a lot of published research not publicly funded, not only are neither university researcher salaries nor institutional library budgets all or wholly derived from public funds, but again, much the same argument could be made for books and other university outputs (even commercial ones) if the underlying problem were indeed the buy-back/double-payment one.

So unless you think that OA should only be accorded to research if -- and in proportion to the degree to which -- public funders or universities are literally drawing twice from the same pot, you may be narrowing OA to a smaller and more difficult-to-determine subset of OA's real target, which is all of peer-reviewed research. (And you may also be making an argument against either the public funding of R&D or the selling of its output: That may be arguable, but it is a far, far bigger argument than OA!)

There is, however, one "double-payment" argument that is valid, but it is not the one the buy-back/double-payment complainants have in mind: rather the opposite. For in this valid double-payment argument it is not the payer (the researcher, his funder, or his university) who is paying twice, but the payee -- the publisher -- who is getting paid twice (but not necessarily by the same payer)

I am referring to the "Trojan Horse" of paying an extra fee to a "hybrid Gold OA" publisher (i.e., not a pure Gold OA publisher but a subscription-based publisher who offers the option of making individual articles OA in exchange for a fee) at a time when the potential funds for paying for that fee are still tied up in the university subscriptions that are already paying that publisher's costs. Hence universities or funders who opt for optional-Gold in a hybrid OA journal today must find extra funds to pay that optional-Gold OA fee (e.g., by redirecting money from research funds), even while all publishing costs are still being fully covered by subscriptions.

But this is not the kind of "double-payment" that the buy-back/double-pay complainants are agitating against: On the contrary, this is the kind that they are (prematurely, and unwittingly) agitating for.

The resolution of the "double-payment" puzzle is very simple: Publishers need only be paid once. Today they are mostly being paid (once, via subscriptions), for providing both (1) the peer-review service and (2) the paper and online editions. As long as the subscription market covers the costs, the only thing researchers need do for immediate OA is to self-archive the final, peer-reviewed drafts of their own published articles in their own university OA Repositories (Green OA); and the only thing their universities and funders need do is to mandate that they do so.

If and when Green OA should ever make cost-recovery from subscription payments unsustainable (because university subscription demand disappears), then the resultant university subscription windfall savings can be redirected to pay for the peer review (on the Gold OA cost-recovery model). And both the paper and online editions, for which there is (ex hypothesi) no longer any demand at that time, can be terminated (along with their expenses), off-loading all access-provision and archiving onto the distributed network of university OA Repositories that has already assumed the access-provision role de facto, in virtue of providing Green OA.

That's a tad more complicated than "buy-back/double-pay," but it is coherent, reflective of the facts and factors (actual and hypothetical) involved, and leads to the same outcome: 100% OA (first Green, and perhaps eventually Gold), but without any fuzziness, double-talk, or unsupportable arguments.

Stevan Harnad
American Scientist Open Access Forum