All That Glisters: Investigating Collective Funding Mechanisms for Gold Open Access in Humanities Disciplines

BACKGROUND This article sets out the economic problems faced by the humanities disciplines in the transition to gold open access and outlines the bases for investigations of collective funding models. Beginning with a series of four problems, it then details the key players in this field and their various approaches to collective “procurement” mechanisms. DESCRIPTION OF PROJECT The Open Library of Humanities seeks to instigate a collective funding model for an open access megajournal and multijournal system that should enable for a phased transition to a gold open access model that does not require author-facing article processing charges. Libraries who participate then have a governance stake in the platform. NEXT STEPS The project is currently working towards sustainability and launch. Authors’ pledged papers are being called in and libraries are signing up to the model. © 2014 Eve. This open access article is distributed under a Creative Commons Attribution 4.0 Unported License, which allows unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PRACTICE jlsc-pub.org | Journal of Librarianship and Scholarly Communication Received: 01/12/2014 Accepted: 04/11/2014


INTRODUCTION
At some point mid-2013, a tipping point was reached for open access. The UK government implemented strong national mandates; the EU's "Horizon 2020" major funding cycle did likewise; and there were steps forward in the US and Australia, among other places. As positive as this might sound, the humanities still trail behind the sciences in open publishing, and there has been extremely vocal opposition to implementations of open access.
While some of this antagonism can be attributed to an elitist approach, and other parts can be seen as a scramble for revenue protection by publishers and learned societies, a third group is convinced of the need for open access but nonetheless raises important questions of funding for such efforts. After all, the humanities often operate on an entirely different basis to their scientific counterparts, exemplified in the fact that most work is unfunded and rests upon institutional support. Indeed, in the humanities disciplines, there would be substantial benefits in formulating a model that could enable gold open access in a sustainable fashion but one that presents no author-facing charges.
To this end, the systems of "Article Processing Charges" (APCs) proposed in the scientific disciplines pose a different challenge for the humanities subjects. This article sets out the economic problems faced by the humanities disciplines in the transition to gold open access and outlines the bases for investigations of collective, or collaborative, funding models. Beginning with a literature review that presents four historical contexts, I then detail the key players in this field and their various approaches to collective "procurement" mechanisms. Finally, I narrate the approach that we are taking with the Open Library of Humanities to investigate such measures.

LITERATURE REVIEW: BACKGROUND AND CHALLENGES FOR GOLD OPEN ACCESS IN THE HUMANITIES
The economic challenges of gold open access for the humanities can be set in many contexts but they are particularly well situated against four historical phenomena: a crisis of library budgets; the rise of the open access movement; the emerging dominance of the APC model; and the cultural backlash against the inequality that this could engender in publication practices.
In terms of a crisis of library budgeting, it is now a widely known fact that academic library subscription costs have outstripped inflation by 300% since 1986 (Brembs, 2012;Eve, 2012; University of Illinois Library at Urbana-Champaign, 2009) and, while the humanities' expenditure accounts for a smaller portion of this than the natural sciences in absolute terms, this is reflected proportionately in the humanities (Bosch & Henderson, 2013). The result of this is that, as their libraries are unable to afford subscriptions, academic researchers at many institutions come up against paywalls that hinder their ability to carry out research, evidenced by the Open Access Button project (McArthur et al., 2013). Likewise, those without access to library subscriptions, such as independent researchers, find themselves locked out of a pay-to-read system if they cannot afford the fees. Similarly, the isolation of research in subscription environments is making it harder to justify the value of the humanities to the public at a time when universities are increasingly facing this demand, as it can appear, from the outside, as though those in the humanities subjects are writing for an incredibly small audience of peers in closed silos while excluding those outside of university environments.
This problem in library budgets is also set against the background of the Open Access movement, the goals of which are to lower permission and price barriers to academic research. These ambitions can be achieved through two different mechanisms, dubbed the "gold" and "green" routes respectively (Suber, 2012, p. 53). The green route involves authors depositing their outputs in institutional repositories (often after a publisher-imposed embargo period) once they have published the work in a journal. While this is desirable for reasons of access, this mode often maintains many aspects of a broken status quo, including the above problems in the world of library budgets, restricted re-use rights that prohibit text mining, delayed access and problems citing material from a form that is not the final publisher's version. In the gold route, by contrast, the material is made available openly at the source through an inversion of the conventional economic model. In this re-imagined scenario, publishing becomes a service in which payment is given to a publisher for the production and hosting of a scholarly object (article/ book, etc.) that is then distributed for free, rather than in the conventional model where publishers sell copies of the same object multiple times. Note well, as the punning title of this piece is supposed to re-enforce, that "gold" open access does not mean an "author-pays" business model (or, indeed, any business model). It rather refers to the dissemination of free-to-read research through journals or books, openly available at their original source in the final publisher version, instead of through institutional or subject repositories (Suber, 2012, p. 53).
These aspects of a library budget crisis, the Open Access movement and the ability to widely disseminate research on a non-rivalrous basis over the internet, has led to the rise of national-level, institutional and funding-council mandates for open access in the UK (HEFCE, RCUK), the EU (Horizon 2020) and Australia (ARC), as well as at the federal and institutional level in the US. Sometimes, as in the case of RCUK, these have stated a preference for the gold route. In many ways, this makes sense: at present, especially in the humanities, normative citation practices make it difficult (and frowned upon) to cite research deposited in an institutional repository (green) as this is often not the final publisher version.
In many of the natural sciences, the OA movement has found substantial success, particularly with the PLOS (gold) and arXiv (green pre-print) projects, the former of which is now among the largest scientific publishers of gold open access material. There are also moves in the social sciences with SAGE's launch of their megajournal SAGE Open. Each of these enterprises enjoys a different degree of success and reputation within their respective fields. PLOS ONE, which launched in 2006, is now the world's largest journal with a reported 75,382 articles as of mid-October 2013 (Binfield, 2013). Even to those sceptical of PLOS ONE's review criteria, which emphasise technical soundness but do not include originality or importance, this represents a substantial indicator of its acceptance by the scientific community. As of 2010, the disciplines with the largest number of articles in PLOS ONE were Genetics and Genomics, Cell Biology and Infectious Diseases while there was less interest from those working on Women's Health and Opthalmology, although this may be because these sub-disciplines are smaller in their scope and definition (PLOS, 2010). Interestingly, also, in John Bohannon's flawed "sting" on open access journal review policies recently, in which there was no sample and an assumption that open access journals were inferior, PLOS ONE was almost the only venue to flag up the ethical problems in the study, demonstrating rigour in their review process, a key feature of any journal's reputation (Bohannon, 2013). Likewise, arXiv has a large number of papers available (894,443 on the 28th November 2013) and it is viewed, within its disciplinary scope, as a valuable resource. However, since arXiv is not a journal and has no review criteria (although certainly a peer reputation system), but is a pre-print repository, it is not "trusted" to carry content of a reviewed quality in the same way as journals with gatekeeping policies or modes of post-review and weighting. As of the 28th November 2013, no category of SAGE Open, which launched in 2011, had more than 100 articles of the 371 total calculated by Binfield (Binfield, 2013;SAGE Open, 2013). The most popular areas for the journal were Education, Communication and Sociology. Likewise, as of November 2013 there are only 19 articles published in SAGE Open's "Humanities" section, perhaps here indicating the problems of a social science publisher attempting to break into a sphere in which it is traditionally less involved.
PLOS and SAGE operate their gold journals on a model called "Article Processing Charges" (APCs). Under this model, authors, their institutions or their research funders must pay a charge. For PLOS' journals this ranges from $1350 to $2900 per article but is waiverable in the case of the author not having the available funds. In the case of SAGE Open, the publisher currently charges $99 after a launch price of $695 with no waiver option. Traditional publishers are also now more frequently offering an open access option, so-called "hybrid" open access publishing in which OA content sits alongside subscription material. For Taylor and Francis, at the time of writing, the price of publishing an article in one of these venues is $2,950.
These rates of APC can, evidently, work in many areas of scientific practice where a large portion of research work is externally funded but, in many cases, humanities research is internally funded by the institution and fees at these rates are not available. This is exacerbated when dealing with books, a field that has been prised open by the strong mandate of the Wellcome Trust. Commercial publishers such as Palgrave Macmillan have proposed an APC (or, rather, a BPC: Book Processing Charge) of $17,500 (£11,000 GBP) per book, which is simply unaffordable for scholars in many unfunded humanities disciplines (Palgrave Macmillan, 2013). Non-profit, scholar-run entities such as Ubiquity Press put the figure for books closer to $3200 (£2000 GBP).
This problematic supply-side payment shift has meant that, despite the substantial advantages open access would present in terms of research, much resistance to OA in the humanities has centred around an "authorpays" model for gold open access (for just one example, see Sabaratnam & Kirby, 2012). Academics are justifiably concerned that the system becomes one in which those who can pay are published and that their institutions will divert funds only to their most favoured researchers. Furthermore, there have been problematic conflations of the APC model with "predatory" publishing, in which the fee payment acts in lieu of true quality control mechanisms. In this case, however, there is a distinct lack of transparency from many conventional publishers as to the actual costs of their operation and we are forced to take publishers' figures at face value. To address this, in the next section of this article I will undertake a baseline costing exercise in which I propose figures for an independent publisher operating on a non-voluntary (i.e. salaried) basis. This will then allow a progression to examine alternative models that could emerge.

BARE COSTS TO OPERATE A SUSTAINABLE, INDEPENDENT, NON-VOLUNTARY OPEN ACCESS PUBLISHER
Excluding profit-motives, there are two primary costs involved in the operation of open access publishing: JL SC a base technological production cost (which can include technological labour costs) and the cost of labour to coordinate the publishing business (managing/editorial director). This split, although somewhat artificial as technological costs are really labour costs, is nonetheless useful for the purposes of analysis.
In terms of technological costs, platform development and maintenance is a planning-and labour-intensive operation. There are, however, several free software projects that go a long way to meeting the needs of a new publisher. The core problem, though, is that the field is currently highly fragmented. Platforms such as PKP's Open Journal Systems and PLOS's Ambra each operate well for their specific purpose, but neither is particularly modular. This means that, if a publisher desires to change publication practice, such as a shift to post-review or peer-to-peer review as advocated by Kathleen Fitzpatrick (Fitzpatrick, 2011), it will involve major modifications to the underlying technological platform. One of the aims of the Open Access Toolset Alliance ('Open Access Toolset Alliance', 2013) is to facilitate coordination and mitigate these problems of monolithic platforms. In the meantime, however, the sensible approach to technological production costs is to work with open source solutions but also to pool labour into communal providers. One such operation, used as a case study here, is the London-based Ubiquity Press.
Ubiquity Press is a technological platform provider originally established by academics from University College London whose goal is to support open access initiatives, ranging from journals to emerging digital university presses. By centralising aspects of technology (primarily open source) they aim to yield the maximum return on economies of scale. Through such a setup, Ubiquity Press can, through this system and in a sustainable manner that allows for future enhancements, provide a sustained and maintained technical platform at a cheaper rate than most could in-house.
Ubiquity Press put their base technological production cost at ~$400 (£250) per article published, and this gives us a good estimate for an article cost at this point. The technology and platform, as handled by Ubiquity Press, operates on a transparent costing philosophy about the uses to which it puts its charges. This base "APC" (which does not have to be author facing) is composed of, from Ubiquity Press' side: • £95 indirect costs (journal support, platform development and maintenance, open access advocacy, business costs); • £85 of editorial and production costs (editorial assistance, typesetting and production); • £40 of waiver premiums (to subsidise those who cannot afford to pay); • £20 of digital preservation and DOI costs (CLOCKSS and CrossRef ); and • £10 of financial administration.
For this fee, Ubiquity Press: • Provides a managing editor to work with a journal/ press; • Provides the website for the journal; • Provides the online submission and editorial management system; • Provides typesetting and hosting of all articles; • Will modify the journal hosting system to accommodate the requirements of the journal/ press; • Assists with promotion of content via calls for papers, social media, press releasing etc.; • Ensures that the journal is appropriately indexed; • Provides the journal with full article level metrics and alt-metrics indicating wider impact (tweets, facebook likes, wikipedia references etc.); • Provides full backup and long-term preservation of content; • Provides membership of COPE to help run journals according to best practices; and • Provides the facility for professional open archiving of research data and software associated with articles.
In a race to the bottom, it would surely be possible to achieve a lower price. However, Ubiquity is a good model of a sustainable, fair rate for the maintenance of a centralised technological platform based upon open source systems.
Conversely, the primary costs of labour for a publisher, on top of the technological production costs, cover: editorial coordination, business legalities, financial administration and advocacy. As with the technological production costs, these rise in parallel to the number of outputs, although there is an economy of scale with regards to the management of editorial labour. Note, however, that, as an employee reaches capacity, the economy of scale temporarily dips every time a new employee is hired to cover this shortfall; there is a stepped-or "staircased"economy of scale.
While lowering APCs to an affordable level through the type of budget operations proposed here could work, there is also another way. If there were a mechanism that preserved the exact same system whereby academics do not see any "pay to say" aspects, criticisms of OA on the financial front would fade away and open access could be appraised for its research use, rather than on the basis of institutional and economic politics. Fortunately, a variety of new models exist that could work to achieve this aim. Models such as this have a precedent in arXiv's revenue model under which, "Cornell University Library (CUL), the Simons Foundation, and a global collective of institutional members support arXiv financially" (arXiv, 2013). In arXiv's case Each member institution pledges a five-year funding commitment to support arXiv. Based on institutional usage ranking, the annual fees are set in four tiers from $1,500-$3,000. Cornell's goal is to raise $300,000 per year through membership fees generated by approximately 126 institutions. (arXiv, 2013) These models are exceptionally promising. They hold out hope of collaboration rather than competition as a principle of scholarly economics. There are, however, two primary challenges that must be overcome by models of this kind, which I will examine the context of the arXiv's and Knowledge Unlatched's approaches:

MOVES TOWARDS COLLECTIVE FUNDING
1. The "free-rider" problem 2. Finding the optimum balance point between level of contribution and number of institutions The first of these issues, the so-called "free-rider" problem, relates to the understanding, in systems of commodity exchange, that rationally self-interested actors do not wish to pay for commodities to which others gain access without paying. In other words, except in philanthropic modes, I usually would not want to pay for goods from which everybody I know would benefit but for which only I pay.
This results, for open access publishing, in a kind of prisoner's dilemma where, if all entities behave in a purely rationally self-interested way (i.e. "selfishly"), it becomes extremely difficult for non-APC models that could save library budgets to emerge. Admittedly, the enclosure of university systems within new and deeper systems of financialisation (McGettigan, 2013, p. 155) doubtless makes it harder for acquisition librarians to justify such expenditure to senior managers and the reason for this is clear: such funding systems rely on cooperation, rather than competition. Through institutional cooperation it becomes possible to build scholarly communication systems that are not possible within systems of pure market economics. arXiv recognises this problem and notes that arXiv's sustainability should be considered a shared investment in a culturally embedded resource that provides unambiguous value to a global network of science researchers. Any system of voluntary contribution is susceptible to free-riders, but arXiv is extremely cost-effective, so even modest contributions from heavy-user institutions will support continued open access for all while providing good value-formoney when compared with subscription services. (arXiv, 2013) On the flip side of the two problems, in terms of finding the optimum balance point between level of contribution and number of institutions, arXiv has chosen to focus on the top 200 institutions worldwide because, in the words of their own FAQ, "they account for about 75% of institutionally identifiable downloads" (arXiv, 2013). This has the substantial advantage of yielding a smaller number of (wealthier) institutions to target but, conversely, means that it is necessary to ask for a larger amount from each (Table 1) while also ensuring that the commodity perk that is exclusive (membership on the arXiv governance board) is primarily restricted to these already-prestigious institutions.
Knowledge Unlatched's pilot, by contrast, consists of 28 monograph titles, with an average "title fee" (the amount the publisher wants to reclaim) of $12,000, thereby totalling a need for $336,000 to be split between the participating institutions (Knowledge Unlatched, 2013c, p. 3). This yields the contribution matrix shown in Table 2.
The "cost per library" column is calculated by dividing the overall cost ($338,000) by the number of participating institutions. The "'Cost' per Book per Library" column is a somewhat artificial measure that notes that if each library were purchasing the book through this scheme, then this is the unit price. However, Knowledge Unlatched is not a purchasing scheme with a "unit price" as such because, once a title is "unlatched," it becomes available to all. That said, and for what it's worth, if comparing Knowledge Unlatched's model to traditional purchasing, the more institutions that participate, the better the value. It remains unclear how this model would scale, though, and how easy it will be to reach the title fee; this could tend towards an incredibly normative selection of open access material. Once more, though, this system is only possible through institutions working in cooperation, not through competition: "This project depends on libraries working together for the benefit of the whole community" (Knowledge Unlatched, 2013c, p. 4).

FUNDING A HUMANITIES MEGAJOURNAL THROUGH A COLLECTIVE MODEL
The Open Library of Humanities (OLH) project, of which I am a co-Director, aims to create a respected, international, prestigious, innovative, digitally preserved, open access  academic megajournal 1 and monograph platform for the humanities with branded overlay journal functionality funded by a model of distributed library subsidy, in this case a series of journals sharing an economy of scale with a communal discovery and mega-journal platform. Before detailing the investigation that we are undertaking into collective funding, it is necessary to describe the project to some degree and to also outline our system of "overlay journals" that acts as a transition mechanism.
The project takes a broad, inclusive understanding of the academic humanities, ranging from the traditional disciplinary fields of classics, religious studies & theology, modern languages and literatures through to political philosophy, critical legal studies, anthropology and newer subject areas such as critical theory & cultural studies, and film, media & TV studies.
The OLH project has two interconnected components: the OLH Base Megajournal (marked in Figure 1 with the OLH logo) and a series of overlay journals that run on top of this.

The OLH Base Megajournal
The core of the entire project is the OLH Base Megajournal. This is envisaged as a trans-disciplinary, large-scale journal that publishes scholarly articles and books on a rolling basis, rather than grouping material into volumes and issues.
Because the OLH platform is breaking into a competitive space in which peer review serves as an indicator of quality, it is vital that our quality control mechanisms work. Indeed, although some members of our steering committee advocated for modes of post-publication peer review, an equal number indicated that they thought it better to transition towards that mode and to, instead, begin review in a traditional pre-publication manner (Open Library of Humanities Steering Committee,  2013). In this light, the OLH Base will have a strong prepublication review system in place at launch.
In our initial, traditional pre-publication review mode for the OLH Base, the process of review will be that: 1. The article is assigned by an OLH Managing Editor to an appropriate disciplinary OLH Section Editor on the basis of the classification provided by the author.
2. The OLH Section Editor follows the agreed OLH review procedure for that discipline.
3. Upon completion of the process, which will be documented by OLH Section Editors, a recommendation will be returned from among: 4. In the case of (a), the Section Editor will assemble all documentation on the review process and pass it back to the OLH Managing Editor who will validate the process and confirm publication.
5. In the case of (b), the author will be requested to respond to the review feedback and to amend their article accordingly. The Section Editor will compare the revised version to the reviewer feedback and work iteratively with the author until satisfied. As with (a), this will then be validated by an OLH Managing Editor.
6. In the case of (c), the author will be requested to respond to the review feedback and to amend their article accordingly. The revised version will then be subjected to another round of review from point #2 in this list.
7. In the case of (d), the author will be informed of the process, sent the feedback but the article will be declined.
Once an article has been accepted into the OLH Base Megajournal, it will be passed on to copyediting, typesetting and proofreading, as described in the technological platform below. It will then be made publicly available, free of charge and discoverable through either search on the platform or through a disciplinary listing of articles. The article will also be digitally preserved in the CLOCKSS (Controlled Lots of Copies Keeps Stuff Safe) archive. As below with overlay journals, the article cover sheet for any publication in the Open Library of Humanities will bear precise details of the review process through which it was admitted and also the name of the Section Editor who oversaw the process.
Over time, in accordance with the progressive elements of our steering committee, once the platform has established enough credibility, we would like to move (in an opt-in fashion) towards a mode of post-publication review, where the pre-publication gatekeeping process moves away from notions of "importance" and instead towards a PLOS-ONE-esque criterion of "technical soundness," translated for the humanities as incorporating (but not limited to and purely for illustrative purposes): novelty, appropriate scholarly apparatus, appropriate range of reference and a basic standard of argument. There are many potential advantages to such an approach, not limited to a broader conception of changing notions of "importance" over time, but it is also critical to note, as documented by Kathleen Fitzpatrick, that "[i]mposing traditional review on digital publishing might help a transition to such publishing" but it should only be a transition mechanism, rather than an end goal (Fitzpatrick, 2011, p. 18).

OLH Overlay Journals
The other major component of the Open Library of Humanities project is a system of overlay journals that were favoured in committee discussions with senior academics (Open Library of Humanities Steering Committee, 2013). These are co-branded journals, each of which will bear both the distinctive marks of a named journal ("The Journal of X Studies") but also the OLH insignia, that run on top of the OLH platform. Material comes to the editors in these journals through two routes: 1. Through direct submission to that overlay journal (in exactly the same way as a conventional academic journal). This material, therefore, will appear in that overlay journal but also in the base OLH platform (across which all users can search). Review is overseen by the editors of the overlay journal according to their pre-published criteria, and the process is then vetted by OLH Section Editors and made transparently available upon acceptance and publication.
2. Through curation of material that has been prepublished elsewhere in the OLH platform.
Material that is published in an overlay journal is available through the centralised search within the Open Library of Humanities Base Megajournal and appears alongside the results there. This mechanism serves several important functions: 1. To demonstrate that value is added through the academic editorial (curation) function.
2. To ensure the widest discoverability and re-use of material.
3. To enable extant journals (learned societies and independent) to transfer onto and integrate with a broader, sustainable platform. This will help protect a number of vulnerable, poorly digitally preserved and/or unsustainable journals.
4. To allow the OLH to rapidly gain prestige on the basis of the journals that are transferring in.
5. To centralise typesetting and production systems to reduce costs.
Peer review and evaluation will be handled in the following ways: 1. Each overlay journal will retain autonomy over its review process.
2. Before OLH will accept the piece, the overlay journal must provide the record of the review process, names of reviewers, number of rounds, recommendations and any other information. This will be verified by an OLH Section Editor. The name of the overlay journal editor and the section editor will be recorded and presented on the article cover sheet.
Although, therefore, overlay journals present a unique challenge for review because material enters the platform through different routes that need their own forms of autonomy, the Committee on Publication Ethics (COPE) code of practice allows for multiple types of peer review. This means there is no ethical problem with different routes into the OLH platform adopting different review methodologies (Committee on Publication Ethics, 2011).
We also propose to actively counteract this potential problem by prominently displaying the review procedure to which a published article was subjected on the cover page of the article and on the landing page of the article itself, including the name of the editor who coordinated the review. In this way, regardless of the route through which the material entered the OLH platform, readers can be assured of a review process on the basis of the academic editor who was responsible for the review.
In order to illustrate how this works under different scenarios, it is worth laying out two of these methods diagrammatically (Figures 2 and 3, following page).
In the first scenario (Figure 2), the author has submitted an article directly to the Open Library of Humanities base platform; he or she has not submitted through an overlay journal. Review, in this instance, is then coordinated by an Open Library of Humanities Section Editor in accordance with the norms of the disciplinary specialism. While these exact specifications for each discipline are not yet formulated, they will be drawn up in dialogue with the editorial committee and section editors. They will then be formally codified and prominently displayed upon submission when an author nominates the discipline under which his or her article should be reviewed (at time of submission from OLH disciplinary taxonomy list).
Once a piece has passed review in this manner, it would be accepted for publication in the OLH base platform and would be cited as published in the Open Library of Humanities. In the case of the above diagram, the third step illustrates a second overlay journal on the platform opting to republish (or curate) the article into one of its issues.
The features of the cover sheet presentation that we propose will include: • A strong statement on review procedure: "This article has been peer reviewed through the doubleblind process of The Open Library of Humanities. The editor who coordinated the review and approved the publication was Dr. X. The details of this process are explicitly specified here: https:// www.openlibhums.org/review-policies/olh-doubleblind/." • An explanation of the co-branded appearance and re-curation: "This article appears in this issue of Journal of X Studies because its editor (Dr. X Y) has deemed it a valuable contribution to that journal, which is an Open Library of Humanities overlay Journal of Librarianship and Scholarly Communication | jlsc-pub.org JL SC journal. For more information, see: https://www. openlibhums.org/overlay-journals/." • Statements on access and digital preservation.
This mode of re-curation enabled overlay journals to use the authority of the editor to present relevant material to their readership, even once an article has been published. In all cases, though, the process of review is made transparently clear.
In a second scenario (Figure 3), the author submits an article to an overlay journal hosted on the OLH platform. Review, in this instance, is then coordinated by the editor at the overlay journal in accordance with the formalised and pre-published policy of that journal before being verified by OLH Section Editors. Once a piece has passed review in this manner, it would be accepted for publication in the journal but also be discoverable through the OLH base platform and would be cited as published in the overlay journal.
In this way, we have a transition mechanism towards an APC-free model for humanities journals, predicated upon a base shared infrastructure. There is no loss of academic freedom or autonomy; journals can remain independent in terms of their review procedures and editorial practices. We simply centralise production systems and thereby reduce outgoings, allowing us to overcome the economic problems set out at the beginning of this article. The only question that remains is how to fund such an operation.

INVESTIGATING A COLLECTIVE FUNDING MODEL FOR OPEN ACCESS IN THE HUMANITIES
From mid-2014 to mid-2015, with funding from the Andrew W. Mellon Foundation, the Open Library of Humanities project is looking to investigate and cement a business model. While we have well over one hundred articles pledged by academics, and while this also puts out our initial optimistic timeframes for a launch of the project, it would be irresponsible to begin publishing this work before we are sure that the initiative is sustainable.
We estimate that our costs to publish 250 articles per year come to $190,000, which includes $100,000 of article production costs on Ubiquity Press' model, $60,000 of staff costs and $30,000 of overheads. While these figures are rough and ready and used here for illustrative purposes, they are viable and instructive.
As comparators for a per-article cost in each of these scenarios, we will use a respected journal of literary criticism and literary theory published by a commercial publisher (Journal A) and a similarly ranked initiative from an American University Press (Journal B).
In 2012, Journal A published 46 articles, excluding book reviews. For that year, the cost of this journal was £721.91. This then equates to a cost of £15.60 (~$25) per article.
Assuming that the pricing of Journal A is consistent across institutions, this cost is replicated at every institution that subscribes. By comparison, Journal B published 26 articles in 2012, excluding book reviews, at a total cost to a single institution of £247.45. This equates to a cost of £9.51 (~$15.50) per article.
To this end, Table 3 below shows what a prospective contributor vs. cost comparison. As can be seen from the colour coding, which is based upon informal conversations with acquisition librarians in the UK, a target of 160 institutions on a banded rate should put the project at an affordable level. The price difference at that level should also be noted as $20.33 cheaper than Journal A and $10.88 cheaper than Journal B.
Once more, however, this model only works if a cooperative, rather than competitive, approach is taken by libraries to support the common good. Whether this  degree we can elicit support is something on which we hope to report back by mid-2015, at which point we hope to be ready to launch the Open Library of Humanities. We hope that this project allows us to reach the goal of gold open access without recourse to author-facing charges: after all, all that glisters is not gold. We would be exceptionally grateful for feedback and/or questions or even early expressions of library support.