The historic mandate from the NIH, voted into law by Congress in January 2008, is perhaps the primary catalyst for the focus on OA in the literature.1 The NIH Public Access Policy requires researchers funded by the NIH to submit an electronic version of their final peer-reviewed manuscript to PubMed Central (PMC) no later than twelve months after publication. Suber discusses the details and significance of the NIH mandate in “Open Access in 2008.”² In this article, he provides a comprehensive outline of significant events and initiatives related to OA that occurred in 2008. Suber states that the “NIH is the world's largest funder of non-classified research, public or private, and its research grants result in 80,000 peer-reviewed articles per year.”³ Suber's “Open Access in 2009” article (published in January 2010 but included in this review because of its 2009 focus) provides additional OA updates occurring in 2009.⁴ English and Joseph also provide a useful overview of the mandate.⁵ These authors describe the law, discuss the history of congressional action leading to the mandate (including discussion of the failure of the previous voluntary policy), and explore library community involvement. Okerson provides further discussion of issues that have emerged since the mandate, including the impact on copyright ownership and on the contractual terms between authors and publishers, which version of the final manuscript should be submitted, and the effects on the information supply chain of researchers, librarians, and publishers.⁶

English and Joseph describe opposition to the NIH mandate by the Partnership for Research Integrity in Science and Medicine (PRISM), an organization sponsored by the American Association of Publishers (AAP).⁷ PRISM supporters argue that the NIH policy violates “copyright and international intellectual property agreements, increase[s] opportunity for government censorship, and threat[en]s … the system of peer review and … science itself.”⁸ Van Orsdel and Born describe an immediate outcry from many publishers in response to this initiative, noting that directors of both the MIT and Columbia Presses resigned from the American Association of Publishers/Professional Scholarly Publishing (AAP/PSP) executive council, and that by the beginning of 2009 at least nine publishers had disavowed PRISM.⁹ Still, with support from the AAP, the Fair Copyright in Research Act was introduced to Congress in September 2008 to overturn the NIH mandate.¹⁰ The bill died without a vote in 2008 but was reintroduced in February 2009. Suber comments in his 2009 update that the bill still had not garnered much support.¹¹ In contrast, he emphasized the continued support for the NIH mandate by Congress with the passing of another bill making the mandate permanent. In addition, public comments called for by Obama in 2009 on the extension of OA policies (and available to Suber as of the writing of his 2009 update) seemed to be “in favor of a government-wide green OA mandate with a short embargo period.”¹²

Partly as a lesson in the value of publisher support for this kind of legislation and partly to provide historical context for the NIH mandate, Miller compares the NIH mandate to the U.S. Office of Education (USOE) policy introduced in the 1960s to place works created by the USOE into the public domain.¹³ This policy was eventually reversed in 1970 after much publisher opposition because the bill provided no copyright protections. Publishers complained of the unfairness in denying copyright, claiming that the works would become unmarketable. This is not the case with the NIH mandate. Miller explains why:

The fact that the NIH policy explicitly leaves the copyright with the author of the work, in contrast to the USOE policy, which placed the work in the public domain, is potentially one of the most important differences between the two policies and may eventually lead to a very different outcome for the current attempt to broaden access to a slice of government-funded research.¹⁴

Miller's discussion provides an interesting history lesson concerning issues in copyright and the value of publisher support.

Meanwhile, the NIH mandate has paved the way for many other mandates and policies. Suber states that the NIH mandate has “made similar policies easier for other federal agencies.”¹⁵ His article also provides several examples of publisher support for the mandate:

• Several publishers will deposit articles into PMC for authors.
• The Nature Publishing Group only requires a six-month embargo, instead of the usual twelve-month embargo, before deposit and is working on developing a system to deposit articles in university IRs as well.
• Springer will deposit all articles from their journal, Genomic Medicine, into PMC, not just those funded by the NIH.

This support hopefully is indicative of increasing momentum for the OA movement. Okerson observes that initial evaluations indicate that deposits into PMC have increased dramatically since the NIH mandate became law.¹⁶ As authors begin to comply with this mandate, they will need to be aware of publishers’ parameters for compliance. Grillot compares agreements from twelve publishers to investigate the varying levels of compliance with the NIH Public Access Policy.¹⁷ In his analysis, Grillot focuses on the “terms affecting an author's ability to comply with the current NIH Public Access Policy. These publication agreements differ from each other in three primary ways: the terms of deposit, the length of the embargo period, and the rights retained by the author—both generally and during the embargo period.”¹⁸ A few of the publishers reviewed by Grillot include ACS, Elsevier, Taylor and Francis, and Nature. This study of publisher response to the mandate should prove to be helpful as authors consider where to submit their publications.

With the attention of governments and prestigious universities around the world directed at OA, serialists increasingly need to have a general understanding of the arguments and models behind this movement. For those readers not intimate with the OA trend and searching for an introduction to the topic, two works consolidate many of the primary concerns surrounding this initiative from unique points of view. One source is Velterop's chapter on OA in the UKSG's E-Resources Management Handbook, an OA publication on e-resources management that expands as authors are commissioned to write on pressing topics for the information industry.²⁷ (UKSG originally stood for the United Kingdom Serials Group, but the organization has grown beyond the United Kingdom and now uses UKSG as its name.) Velterop's discussion of OA provides thoughtful yet readable explanations and criticisms of the various arguments for OA. He makes several points: OA articles are cited more often than their non-open-access counterparts and have more scientific impact; the tax-paying public should have free access to the science literature funded by tax dollars; OA increases the efficiency of scientific discovery; and OA is technologically possible, thanks to the Internet. This last argument is the most compelling to Velterop. He explains that “the strongest argument for OA is simply that it is possible, now that the Internet has reached a level of stable maturity and reliability.”²⁸

Since Velterop writes as a member of the publishing industry, his comments about the role of the publisher amidst the changes in scholarly publishing are insightful. He discusses the value that publishing provides regarding “scientific recognition” and notes fundamental changes to the information environment because of the Internet allowing quick and easy dissemination of information.²⁹ Because the economic value of publishing had historically centered on the dissemination of information, publishers will now have to market the service of publishing to authors to remain financially viable. Velterop posits that “publishers who recognize the impact of the Internet on the environment in which they work, and who are willing and able to adapt to the new environment, have a very good prospect of surviving and thriving.”³⁰

The second work of note provides a more cautious view of the value of OA. In his essay, “Open Access 2.0,” Esposito provides a realistic assessment of the most effective use of OA models in scholarly communication, observing that they are effective in “providing services to authors whose work is so highly specialized as to make it impossible to command attention of a wide (at least in academic terms) readership” and supporting communication across a small number of researchers working in the same area.³¹ For Esposito, OA as a replacement for traditional publishing is simply not economically feasible without stripping away all of the administrative costs associated with peer-reviewed services. He notes that “one of the reasons that many open-access ventures have had a hard time financially is that they have been built on the mistaken assumption that they are replacing traditional publishing and thus have to recreate all of the services that traditional publishers now provide.”³² Esposito further explains that “a high cost structure demands sizeable revenue streams to offset it.”³³ He argues that, given the critical mass of information available on the Internet, the service of publishing to validate and qualify research efforts is more important than ever. For Esposito, traditional publishing and OA should fulfill uniquely different roles to facilitate scholarly communication.

The chapter by Velterop and the essay by Esposito taken together provide a useful snapshot of the arguments surrounding OA. A few other works fill the gaps in between. As noted earlier, Suber's “Open Access in 2008” and “Open Access in 2009” provide exhaustive lists of facts concerning OA policies, growth numbers IRs and journals as well as OA initiatives involving nonserial materials such as data and books.³⁴ Lawrence's “Staying Up to Date” column focusing on OA provides a short bibliography of additional sources of background and news information concerning OA.³⁵ Both of these works are useful sources for learning more information about the OA initiative.

Other works in the literature focus on the argument that OA articles have a greater citation impact than non-open-access articles. Harnad and associates explore this concept in their discussion of the research and access impact problem, i.e., for other researchers to cite works and therefore give them greater impact, they have to be able to find them.³⁶ The authors credit self-archiving for the increased availability of OA articles and point out that many publishers already have policies in place that allow self-archiving in IRs. Universities have been slow, however, to archive much of their body of work. Another study, by Norris, Oppenheim, and Rowland, analyzes the citation advantage of OA articles across four subject areas but could not determine the cause of the observed advantage.³⁷ These authors determined that the citation advantage did exist but varied across disciplines. A study by Frandsen found inconsistent citing behavior by authors in the fields of biology, mathematics, and pharmacy.³⁸ The goal of Frandsen's study was to analyze the integration of OA journals into the scholarly communication system. The author determined that, in some fields, authors publishing in OA journals have different citing behaviors than authors publishing in non-open-access journals. One possible inference from this study is that OA authors and non-open-access authors do not always include citations from both OA and non-open-access journals in an integrated fashion.

Another argument surrounding the OA debate is the value of public access to scholarly research. Velterop bluntly states “that most scientific research is inaccessible (intellectually) for the lay person, even if he or she had access to it (physically).”³⁹ Even if public access is important for many OA advocates, Velterop argues that public access to research may not be desired by much of the public with the exception of medical information. In Kuchma's discussion of the value of OA for developing countries, she cites several public access benefits described by the U.S.-based Alliance for Taxpayer Access (ATA).⁴⁰ This organization supports public access as “an opportunity for publishing and other industries” as a means to ensure that “stakeholders have access to research they need, taxpayer equity,” and support for innovation and research.⁴¹ Two other articles by Harnad and Ferguson argue that the value of OA is not public access to scientific research but peer-to-peer access.⁴² Harnad states,

It is researchers worldwide who are today being denied access to the research findings they need in order to advance their research … for the benefit of the general public for whom it is hoped that some of the findings will eventually by applied in the form of technological advances and medicines for illnesses, as well as intellectual and cultural food for thought.⁴³

Harnad's thoughts stem from a concern about faculty's willingness to self-archive. The public access argument is not as effective as the peer-to-peer argument for convincing faculty to take the “green road,” i.e., self-archiving scholarly publications in an IR. Ferguson addresses this argument in response to one of his peer's complaints about public access to research. Ferguson's peer posits that funds reserved for research are better utilized when applied directly to research efforts rather than being funneled to support OA initiatives. In response, Ferguson redirects the researcher to the value of peer-to-peer interaction, explaining that “OA communicates more widely and will reach scholars denied access to expensive publications.”⁴⁴

Finally, concern about the economic sustainability of OA publishing models was another common theme in much of the OA literature. Both Esposito and Velterop comment about the financial sustainability of altruistic OA models supported by volunteers that re-create the traditional editorial process.⁴⁵ Esposito argues that a critical mass of publications will not allow for this kind of free service. He comments that “the cost of ‘free’ is simply too expensive unless we strip away almost all the administrative costs. This is why libraries are very poor places to establish open-access services: libraries provide outstanding high-touch service and are culturally out of synch with the need for literally impersonal technical services.”⁴⁶ In these statements, Esposito is not criticizing librarianship as a profession, he is providing a reality check that the IR model also requires capital, in both dollars and human resources. Velterop also comments on altruistic support for OA, noting the difficulty volunteers would have in publishing thousands of journals. He does see, however, possibilities in the author-pays model, but wonders if the information industry is ready for this switch.⁴⁷ Hagenhoff, Blumenstiel, and Ortelbach discuss the author-pays model in depth in “An Empirical Analysis of the Amount of Publication Fees.”⁴⁸ This article analyzes the impact of author-pays or publishing-fees models in different subject areas by calculating an estimated publishing fee based on the subscription fee of a title divided by the number of articles published per year. When considered in the context of the amount authors were willing to pay and the subscription budget for a subset of libraries, these authors conclude that the fees publishers would need to charge to maintain their current revenue would be more than what scientists are willing to pay. Furthermore, subscription budgets for universities (or at least the subset studied) are probably not robust enough to financially support an exhaustive author-pays model. Given the infancy of this model and the unlikely conversion of a critical mass of titles to an author-pays model over a short period of time, additional studies will be needed to continue to assess the sustainability of this model as it evolves over time. In the short term, however, skepticism exists about the author-pays model and concern remains about acceptance of this model by scientists.

A column by Black raises additional concerns about the financial viability of OA from journal editors for both OA and non-open-access journals (in addition to other current topics in serials).⁴⁹ The journal editors interviewed by Black in this article questioned the sustainability of models that depend on sponsorship or volunteer efforts. They noted the costs associated with publishing and commented on the misconception that OA journals are free. One editor stated that “the viable journal costs money to produce.”⁵⁰ Another editor raised concern over existing models like the Public Library of Science (PLOS) that have not yet proven to be financially sustainable, and one pointed out that the green model of traditional publishing followed by self-archiving at least allows publishers to maintain their revenue streams. Because these interviews were with editors that supported OA both in practice and in theory either by allowing self-archiving or because their journal was OA, their comments come from a credible perspective. These editors were hopeful about OA, but practical concerns remain ever-present in their thoughts. One editor directed his concerns to proponents of an exclusively OA scholarly communication system, observing, “They are looking at [the publishing] system in an unrealistic way. It does require expertise and time and lots of materials and machinery and special knowledge and equipment in order to put a journal out.”⁵¹ This editor argued that these efforts are undervalued by many OA advocates.

Given many of these valid worries, SCOAP³, another publishing model discussed in the literature, received positive support for trying to realistically tackle the issue of the publisher revenue stream. SCOAP³ is a consortium of high-energy-physics funding agencies, high-energy-physics laboratories, and leading national and international libraries and library consortia. In his article, “Open Access Publishing in High-Energy Physics,” which outlines the SCOAP³ model, Mele directly criticizes OA models, such as the author-pays model.⁵² He comments that direct payment models are “perceived negatively” by the high-energy physics (HEP) community and also notes that many concerns exist for this model among libraries because of the potential for increased costs to the library through both increased subscription rates and additional author-pays fees, at least during a transition period between the subscription model and the OA model.⁵³ Mele moves on to describe the SCOAP³ project and the intended role of libraries. Libraries and other funding organizations will essentially redirect subscription dollars to SCOAP³, which will then negotiate with publishers in a bidding process or call for tender to cover the costs of peer-review services of seven journals that focus on HEP. Mele points out that the success of SCOAP³ is dependent on all universities and HEP programs supporting the project.

SCOAP³ is the focus of another article in Serials Review, which presents three essays with varied points of view.⁵⁴ The first essay, also by Mele, outlines the SCOAP³ model, discusses the preprint culture of the HEP field instrumental in aligning scientists in this field with OA, and emphasizes the need for support for the model, especially from U.S. institutions. Mele comments further that SCOAP³ will be unable to put out a call for tender until a significant number of universities have signed on. The next essay, by Morrison, encourages libraries to take a leadership role in supporting the SCOAP³ initiative. Morrison also outlines numerous benefits of the model, including the fact that it is well conceived. Organizers of the initiative also have considered how to tackle licensing entanglement through big deals and multiyear contracts. SCOAP³ calls for publishers to unbundle these titles and reduce costs as appropriate. The final essay, by D'Agostino, reflects on the value, scalability, and transferability of the model. He cautiously notes the slow acceptance of this model by libraries and comments on the unique characteristics of the HEP discipline that may limit this model to very few disciplines. HEP is a small community with a concentrated group of journals. This discipline also has historically supported OA for several years, as exemplified by the success of the arXiv preprint repository. Two additional concerns D'Agostino discusses relate to the point of view of the funder. He comments that as libraries realize savings from HEP journals moving to OA, those libraries may be reluctant to contribute them to SCOAP³. He further notes that the model for assigning the contribution costs is “free-floating” and not stable, with the potential to change from year to year.⁵⁵ These kinds of varied costs make budgeting difficult. D'Agostino does acknowledge at the end of his essay that regardless of whether SCOAP³ fails or can be transferred to other disciplines, administrators of this initiative will be able to provide a realistic assessment of a journal's production costs.

Along with the attention given to the SCOAP³ model, several publications provide suggestions and ideas to support publishers in changing from a subscription model to an OA model. Crow's Income Models for Open Access is a resource for publishers investigating alternative ways to distribute scholarly publications through OA.⁵⁶ This guide provides an overview of current income models used to support OA that might be applicable for a broad spectrum of publisher operations—commercial, nonprofit, small or large-scale. Models discussed include demand-side types, such as versioning or use-triggered fees, as well as service supply-side types, such as author-pays, advertising, or sponsorships. The intention of Income Models for Open Access is to support publishers in discovering potential OA options. An article by Björk and Hedlund also addresses the topic of converting publishing models by outlining two possible scenarios for publishers interested in converting to OA: the instantaneous-change business model and the gradual-change business model.⁵⁷ SCOAP³ is described as a good example of the instantaneous, or flip, model, in which the publisher changes its model completely to OA. One method of achieving this effect, first discussed by Rowse, is for publishers to convert existing license agreements to allow “the academics in those institutions to submit content rather than to access content. The publisher would have successfully flipped its business model completely.”⁵⁸ Björk and Hedlund comment that this model is probably not suited for many publishers and that the HEP publishing environment is unique with its early emphasis on preprints to support scholarly sharing. However, a more gradual model for supporting OA might have potential for publishers unwilling to consider an immediate conversion from a traditional publishing model. For instance, Springer Choice and Wiley-Blackwell's Online-Open are examples of a model in which the publisher allows the author to purchase OA rights. This model “is one in which a major publisher starts to bundle the license to all its journals (‘big deal’) with an institutional payment for the selective open access of the articles from authors at the institution in question.”⁵⁹ The obvious benefit from this model is that the publisher is able to maintain their revenue stream.

As many publishers experiment with OA models, authors push to openly share research findings and librarians promote the value of self-archiving, the future of scholarly communication is taking shape. Several articles and reports have started to evaluate the effects of OA on scholarly communication. Economic Implications of Alternative Scholarly Publishing presents a study commissioned by the Joint Information Systems Committee (JISC) that attempts to quantify the savings the United Kingdom would experience in a comprehensive move to OA for scholarly publications.⁷⁷ This report claims, “For UK higher education, these cost differences would have amounted to savings of around £80 million per annum from a shift from toll access e-only to OA e-only publishing circa 2007.”⁷⁸ Another collaboration, Publishing and the Ecology of European Research (PEER), was created to study the consequences of storing scholarly research manuscripts in IRs.⁷⁹ This three-year project is still underway, but Shepherd and Wallace provide an update describing the partners and objectives of the study. These objectives include investigating the effects of archiving on journal viability and understanding whether access is increased for journals in the study. The effects of IRs on European research and on self-archiving behavior by researchers also were of interest. Researchers hope to gain an understanding of how traditional and OA models can co-exist. Findings from this study will, PEER hopes, allow stakeholders in the scholarly communication system to move forward in supporting OA through self-archiving in a constructive fashion.

Covey presents another study focused on self-archiving; this time the focus is self-archiving practice.⁸⁰ Through an analysis of cited or referenced journal articles on faculty websites, the author's findings from the study indicate that self-archiving practices are inconsistent across disciplines and that faculty often do not self-archive when allowed by publisher policy. In addition, faculty appeared to pay little attention to publisher policy when or if they chose to self-archive. This study points to a lack of awareness or concern about copyright issues for journal content. This kind of awareness is important for researchers if publishers are to support OA initiatives collaboratively. These findings also reiterate the continued need to market IRs to faculty. The final evaluative study cited here circles back to the beginning of the OA discussion—the public access mandate. Brazzeal and Carr evaluate the potential effects of public access mandates on forestry research publications.⁸¹ The authors evaluated the number of articles with federal authorship or funding from 2006 and determined that a significant number of articles supporting forestry would need to be made available through public access. Although the ramifications of large-scale archiving on the publishing industry are not yet known, this study suggests that some research fields will not have long to wait before seeing a change in scholarly communication culture for their discipline.

The tone from the collective body of articles focusing on OA in the 2008 and 2009 serials literature is cautious yet optimistic. The idea of OA to scholarly content suggests a possible end to the serials pricing crisis defined over the last few decades by dramatic price inflation and decreasing library budgets to support serials collections. However, practical concerns about the viability of OA models also dominate the literature. Many articles published in 2008 and 2009 focus on the NIH mandate as a potential catalyst to increase faculty awareness and support for the OA movement. The increasing number of access policies adopted by universities also indicates increased acknowledgement by faculty of the value of OA. Arguments promoting OA across the literature included better public access to research, greater citation rates, and improved scientific discovery through enhanced peer-to-peer communication.

Even as the Internet provides a feasible platform for discovery of content, the economic feasibility of OA models, both gold (OA journals) and green (self-archiving in IRs) remains a valid concern. The slow adoption of self-archiving by faculty is another problem noted by OA advocates. Strategies, such as financial incentives, and value-added services, such as copyright advising and rights checking have been deployed to increase acceptance; however, mandates from funding organizations and universities may provide the strongest motivation for faculty to self-archive in IRs. Publishers, authors, and librarians alike continue to explore their respective roles in supporting the OA movement. The literature reveals several areas that need further investigation into the effects of OA on the scholarly communication system. These include potential savings, traditional journal viability, and effects of OA on researcher behavior. Only time will tell if the OA movement will gain enough traction to revolutionize the publishing industry; but without a doubt, librarians will continue to demonstrate avid interest in the future of OA.