Nine questions for hybrid journal programsIn the last month alone, four publishers have launched hybrid OA journal programs. That's serious momentum. Almost as many publishers launched hybrid programs in the last month as in the previous year, and more launched in the previous year than in all earlier years combined.
SPARC Open Access Newsletter, issue #101
September 2, 2006
by Peter Suber
* Here are the four programs from August in the order in which they were announced:
--Announcement, August 3, 2006
--My blog comments
--BMJ Publishing Group (no web site yet for BMJ Unlocked)
Wiley's Funded Access
--Announcement August 8, 2006
--My blog comments
--John W. Wiley & Sons (no web site yet for Funded Access)
Cambridge University Press's Cambridge Open
--Announcement August 11, 2006
--My blog comments
--Early warning of CUP's intentions from May 2006
--Cambridge University Press (no web site yet for Cambridge Open)
The American Physical Society's Free To Read
--Announcement August 16, 2006
--My blog comments
--American Physical Society (no web site yet on Free To Read)
By a "hybrid journal" I mean one that publishes some free-access research articles and some toll-access research articles, when the decision between the two kinds of access is the author's rather than the editor's. Authors who choose the free option must usually pay a fee (or find a sponsor to pay a fee) to cover the journal's expenses. In return the publisher provides immediate free online access to the article at its own web site. Authors who don't choose the free option don't pay a processing fee, although they might still pay page and color charges. Nor do they get immediate free online access, although they might get delayed free online access if the journal provides free access to its sufficiently old back issues.
(I've made this definition a little thicker than necessary in order to avoid the term "open access". Some publishers carefully and properly avoid the term "open access". But now that I've been precise, I will sometimes, for convenience, refer to these as "hybrid OA journals" and to the new option as an "OA option".)
The momentum for hybrid journals is understandable. The option is nearly risk-free for publishers. If the uptake is low, they still have subscription revenue to pay the bills. If the uptake is high and subscribers start to cancel, they have fee revenue to pay the bills. If they promise to reduce their subscription price in proportion to author uptake (and some do), then fee revenue should pay the difference. If they don't promise to reduce subscription prices (and some don't), then they have what a friend calls a "double charge" business model. Hence, in one form or another, the model should spread fast and far.
How good would that be for OA? Before answering, consider:
* Nine key questions to ask about any hybrid journal program
(1) Does the journal let participating authors retain copyright?
If not, then authors (or their sponsors) are only paying to remove price barriers, leaving permission barriers in place. The journal is forcing users to put up with the delay and expense of seeking permission whenever they want to exceed fair use, for example, to quote long excerpts, to print full-text copies, to email copies to students or colleagues, or put copies on CDs for bandwidth-poor parts of the world, to migrate copies to new formats or media to keep the text readable, to archive copies for preservation, to deposit a copy in an OA repository independent of the publisher (more on this in Question 3 below), to include the work in a database or mashup, to copy the text for indexing, text-mining, or other kinds of processing, to make an audio recording of the text, or to translate it into another language.
None of the August Four lets authors retain copyright. The same is true of most of the other publishers with hybrid programs. The stand-out hybrid publisher is Springer, which originally (July 2004) asked authors to transfer copyright, but after Jan Velterop arrived as its Open Access Director (August 2005) changed its mind (October 2005) and let authors retain copyright.
(2) Does the journal use an OA-friendly license, like those from Creative Commons? Does it let authors do so?
If the journal removes important permission barriers or lets authors retain key rights, but doesn't formalize this permission in a human- and machine-readable license, then in effect it's hiding information about which uses are permitted and which are not, forcing users to choose between the delay of seeking permission and the risk of proceeding without it. In this situation, conscientious users will either seek permission or err on the side of non-use --two kinds of harm that OA was designed to prevent. When the publisher has already given permission but hasn't made the permission easy to discover, then the harm is caused by poor communication.
Again, none of the August Four uses an OA-friendly license and Springer is the stand-out from previously announced hybrid programs. At the same time that Springer decided to let authors retain copyright, it adopted an equivalent of the Creative Commons Attribution-NoCommercial license.
(3) Does the journal automatically deposit participating articles in an OA repository independent of the publisher? Does it allow the author to do so?
If not, then authors and readers have no guarantee that the article will remain available, and remain OA, in case the journal folds up, is bought out, or simply changes its access policy. PLoS and BMC, for example, automatically deposit all their OA articles in PubMed Central.
There's another issue here beyond insurance against policy-change at the publisher. The Wellcome Trust requires that articles resulting from Wellcome-funded research be OA through PubMed Central or UK PubMed Central. OA through from the publisher's site or even an institutional repository is not good enough. Unless hybrid publishers allow deposit in PMC or UKPMC, they are essentially excluding Wellcome-funded authors.
BMJ is the only one of the August Four that routinely deposits all participating articles in an independent OA repository --in this case, PubMed Central. It does so immediately upon publication. Wiley will deposit participating articles in a funder's repository when the funder requires it, but not otherwise. Authors may not apparently deposit in other repositories even if they pay Wiley's fee. APS lets authors post participating articles to a personal web site or deposit them in an institutional repository; but authors need ASP permission to deposit them anywhere else. Cambridge is silent on the subject.
(4) Does the journal waive fees in cases of economic hardship?
If not, the journal's OA option may lie beyond the reach of most authors --those without the means to pay on their own and those without a funder or employer to pay on their behalf. This is not itself a barrier to publication, since hybrid journals will still publish in the conventional way for authors who don't choose the OA option. But if it's not a barrier to publication, it's a barrier to the extra benefits of OA publication, at least as provided by that journal. Moreover, more non-OA journals than OA journals charge author-side fees in the form of page and color charges (more under Question 8, below). So even the conventional or non-OA publishing route at a given hybrid journal may erect barriers to indigent authors.
Fee waivers are very common at full OA journals and very rare at hybrid journals. None of the August Four offers fee waivers and I don't know of any other hybrid program to do so either.
(5) Does the journal promise to reduce the subscription price in proportion to author uptake?
If not, then it's simply introducing a way to be paid twice for the same articles. Neither authors nor subscribers should tolerate this; at least one of those parties is entitled to some relief.
Cambridge is the only one of the August Four to promise to reduce its subscription prices in proportion to author uptake. Wiley and BMJ are silent on the question, which means they are not making the promise. The APS position is more complicated. APS will use fees to "augment [its] current subscription income" (the double-charge model), at least "at first". If uptake is high and libraries cancel subscriptions, then it will use fees to "offset such losses". If the fees generate revenue beyond the amount needed for such offsets, then APS will lower subscription prices but only for selected institutions.
(6) If authors have a prior obligation to their funding agency to provide OA to their peer-reviewed manuscript, does the journal let them comply without choosing the new OA option and paying the associated fee?
If not, the journal is trying to interfere with a funder-grantee contract to which it is not a party. Authors shouldn't have to pay their publisher in order to live up to a contract with their funder. Publishers shouldn't ask them to do so; and if they do, neither authors nor funders should tolerate it.
APS, BMJ, and Cambridge are silent on this subject. Wiley requires authors to pay if they want to comply with an OA mandate from their funder. To twist the knife, Wiley says that it developed its hybrid program in order to "support" authors with obligations to funders.
There may be a loophole. The Wiley policy seems to apply only to authors who want to deposit the published edition of their article. But most funder policies only apply to the final version of the author's peer-reviewed manuscript, not to the published edition. Wiley doesn't say whether it would charge fees for authors who want to comply with funder mandates in that form.
(7) If the journal previously allowed author self-archiving without an embargo, does it still allow it for authors who do not choose the new OA option?
If not, then the hybrid program constitutes a retreat on the OA archiving front even if it's an advance on the OA journal front. Authors who would have chosen self-archiving must now choose between accepting an embargo and paying a fee for what used to be free. If they want immediate OA, then the archiving option is closed and only the fee-based journal option remains.
When a green publisher introduces a hybrid journal program, it creates new pressures on itself to retreat from green by putting an embargo on self-archiving. But authors should exert a countervailing pressure to prevent journals from retreating on green in order to advance toward gold. Authors should be allowed to self-archive their peer-reviewed postprints without an embargo or fee, regardless of their choice on an optional new journal program. Publishers with policies to the contrary are asserting control over an edition that has not undergone copy editing, final manuscript preparation or mark-up, and that is usually not subject to the copyright transfer agreement.
APS and Cambridge are the only two of the August Four to allow non-participating authors to self-archive their peer-reviewed postprints without an embargo or fee. (Cambridge also allows self-archiving the published PDF but only after a 12 month embargo.) BMJ lets non-participating authors deposit their articles anywhere they like, but only after a six-month embargo; this is a retreat from its previous green policy.
Wiley says nothing about how its hybrid program, which tightly restricts where OA copies can end up, affects its self-archiving policy. It was green before it launched its hybrid program and we should assume it still is.
Also see Bill Hubbard's summary of the self-archiving policies of the major hybrid journal publishers.
(8) For participating authors, do the OA publication fees cover page and color charges or are the latter laid on top of the former?
If authors (or their sponsors) must pay both kinds of fee, this needn't be a reason to look elsewhere. It's just a reason to keep their eyes open and understand the full price.
From one point of view, it's perfectly fair for a hybrid journal to charge authors both kinds of fee. The journal is providing more than one service and there's nothing wrong with charging separate fees for separate services. But from another point of view, it's a sleight of hand that could easily mislead authors. The only purpose for OA publication fees is to cover the journal's costs of publication. Journals that charge author-side processing fees and then add page and color fees on top are either charging twice for the same service or keeping their processing fees deceptively low.
(This objection doesn't apply to journals that can keep their fees low because they have other sources of revenue, such as advertising or subsidies. The reason is simply that they never charge the author more than the one low fee.)
BMJ, Cambridge, and Wiley are silent on this subject. APS levies page and color charges on top of the OA processing fee. However, even after paying all these fees, APS authors don't get OA to the published edition, but only to a version stripped of links to citing and cited articles.
(9) Is the fee high or low?
There's no right answer to this question, but there are at least two criteria. For publishers, the test is whether the fee is high enough to cover its costs. For authors, the test is whether they can afford it, after taking into account any money they can raise for this purpose from their funder or employer. Of course, a given fee may pass the first test and fail the second.
When publishers justify high fees by saying that they reflect their actual costs, we may believe them. But that doesn't change the fact that other publishers seem to have lower costs and that the fees may still be too high for most authors to afford without a subsidy from a funder or employer.
That's just one reason for a note of caution: we cannot conclude that a low level of author uptake indicates a low level of author interest in OA.
At Cambridge, the fee is a flat £1500/$2700. The Wiley fee is a flat $3,000. APS has a two-tier fee, $975 for articles in Physical Review A-E and $1300 for Letters in Physical Review Letters. BMJ also has a two-tier fee for different BMJ journals, though the two fees are higher, £1,200/$2,220/€1,775 and £1,700/$3,145/€2,515.
Buyer beware: if the fee is too high for you or your sponsor, you don't have to pay it. You can submit to an OA or hybrid journal with a lower fee or you can submit to a conventional non-OA journal (one of the 70% that already consent to postprint archiving) and self-archive. Most hybrid journals make it hard to elect a third option that might otherwise be attractive: choose the non-OA path at the hybrid journal and self-archive.
* A few other issues
Those are the nine major questions. But here's a minor tenth:
Did the publisher previously criticize the very idea of charging author-side fees for OA dissemination, arguing that it corrupted peer review? (Elsevier and the Royal Society did.) If so, how does the publisher escape its own criticism? Has it issued a retraction?
It's one thing for a publisher new to this model to institute a good editorial firewall to prevent its editors from thinking about fee revenue when evaluating papers. But it's another to stand by false and harmful criticism of other journals that had good firewalls all along.
On the one hand, I want publishers who formerly opposed OA journals to change their minds and support them. I want to praise that as progress, not criticize it as inconsistency. But on the other, when a publisher tries to gain by condemning a good practice it has since adopted itself, then I want to point out the double standard.
Hybrid journals raise a host of other issues not covered by the questions above. For example, APS applies its program retroactively, and will make any of its previous articles OA if someone, anyone, pays to ransom it. Cambridge includes participating OA articles in the print editions of its journals. APS is explicit that its hybrid program "represents a path by which APS could gradually transition to full Open Access."
* Is anyone buying?
Of all the hybrid publishers, Oxford has done the most to share data about author uptake.
Oxford's report after the first quarter of operation of Oxford Open, November 3, 2005
Oxford's June 2006 workshop on its data and the workshop report
Oxford's report after the first year of operation, August 30, 2006
Here's a key section from its report last week on one year of operation:
In the first year of launch, almost 400 papers have been published under the optional open access model across 36 of the 49 participating titles. The majority of uptake of optional open access has, as predicted, been in the life sciences, with approximately 10% of authors selecting the open access option across 16 participating journals in this area, compared with approximately 5% in medicine and public health, and 3% in the humanities and social sciences. Three life sciences titles in the areas of molecular and computational biology have seen over 20% uptake. The highest of these was for Bioinformatics, which has published over 50 open access papers in 2006. 2007 online subscription prices have been adjusted for these journals to reflect this uptake.
Oxford had a better experience with Nucleic Acids Research, which converted to a hybrid journal in January 2004 after 32 years as a conventional subscription journal. The revenue and author uptake were good enough to persuade Oxford to convert it to a full OA journal in January 2005.
Nucleic Acids Research (NAR)
NAR's August 2003 announcement of its January 2004 conversion to a hybrid journal
NAR's June 2004 announcement of its January 2005 conversion to a full OA journal
During its first year as a full OA journal, NAR's increased its already impressive prestige and impact. From an Oxford press release, March 23, 2006: "With three articles in the top 40 [of Thomson Scientific’s recently published Top 40 "Red-Hot Research Papers" for 2005], Nucleic Acids Research, (NAR) was ranked as the "hottest" single-discipline journal in the world and the fifth "hottest" journal overall."
Until we see similar data from other hybrid publishers we won't know how much new OA literature they're producing.
* Strengths and weaknesses of the hybrid model
Hybrid journals are good for OA roughly in proportion to author uptake. To that extent they enlarge the body of OA literature. But the overall balance of costs and benefits depends on how a given program answers the nine questions. If the author uptake is low, if the self-archiving policy has been saddled with a new embargo, and if authors must start paying their publishers for the right to comply with their own prior funding contracts, then the balance could easily tilt the other way and a hybrid program could do more harm than good.
One strength of a hybrid program is that it helps the publisher learn the economics of OA publishing, or at least the kind supported by author-side publication fees. Journals can experiment to set the fee high enough to pay the bills and low enough to encourage author uptake, and they can do this without a wholesale conversion to OA. They can try different variations on the theme to discover which ones best attract authors.
Unfortunately, hybrid journals are regrettably limited on the last point. Because they can always fall back on subscriptions, they don't need to make the OA option work. As laboratories for the economics of OA publishing, they omit the strongest incentives to make the offer appealing to authors. That's one reason why we see so few hybrid journals, as opposed to full OA journals, let authors retain copyright, use OA-friendly licenses, deposit articles in independent OA repositories, offer fee waivers, cooperate (rather than meddle) with funding contracts, permit self-archiving without embargo or fee, dispense with page and color charges beyond the OA publication fee, or do all they can to hold their fees down.
In SOAN for June 2006, I argued that the chief strength and the chief weakness of hybrid OA journals were the same: because only some authors in a given issue will select the OA option, libraries cannot justify cancelling their subscriptions. This is a strength for publishers because it protects them from risk and encourages them to try the experiment. It's a weakness for libraries and universities because it postpones the day when they will save money from OA journals.
All OA initiatives help researchers, but some do and some don't help libraries. Hybrid journals are a perfect example of the last type. We have to remember that this is deliberate. If these programs helped libraries save money by cancelling subscriptions, then publishers wouldn't be rushing to experiment. To the extent that hybrid journal programs help OA, we can celebrate. But we should remember that researchers are only one of the groups with a stake in the progress of OA.
Ironically, hybrid OA journal programs could overcome most of their defects if they did more to help libraries. The best long-term source of funding for OA journals is the money now spent on subscriptions. The more money freed up by cancellations, the more readily and reliably we can pay for the OA alternative. If this money were available today, then journals could be full OA instead of hybrid OA. They would charge no subscriptions, helping libraries as well as researchers and mooting the question whether they would reduce prices in proportion to author uptake. They would feel no pressure to retreat from a green self-archiving policy, since they would have no financial interest in enforcing embargoes. They would let authors retain copyright (as full OA journals typically do), since they would no longer have an interest in limiting copying and redistribution. They would let authors deposit in independent OA repositories, with some (like PLoS and BMC) even making these deposits on their own. They would not charge authors for the right to live up to a prior funding contract, since full OA journals couldn't threaten to shunt authors to the journal's non-OA track. Every university-employed author would have a sponsor, freeing authors from worry about high fees and freeing publishers from worry about uncertain revenues. And finally, if most authors had sponsors, then fewer would need fee waivers, making it easier for all journals to offer them, eliminating the barrier to entry for those not already covered by funders or employers.
My hope, therefore, is that publishers will use the current hybrid models only as a stepping stone to full OA journals, not as a destination.
* Postscript. Here are the major hybrid programs announced before the August Four (chronological order by launch). I'm sorry to have to omit individual journals, for some were important in the evolution of the hybrid model.
Springer's Open Choice (launched July 2004)
--My SOAN article about it
Blackwell's Online Open (launched February 24, 2005)
--My SOAN article about it
Oxford University Press' Oxford Open (launched July 1, 2005)
--My SOAN article about it
Elsevier's Sponsored Article journals (launched May 24, 2006)
--My SOAN article about it
The Royal Society's EXiS Open Choice (launched June 21, 2006)
--My SOAN article about it
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, see the instructions for subscribing at either of the next two sites below.
SPARC home page for the Open Access Newsletter and Open Access Forum
Peter Suber's page of related information, including the newsletter editorial position
Newsletter, archived back issues
Forum, archived postings
Conferences Related to the Open Access Movement
Timeline of the Open Access Movement
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