Accessibility, sustainability, excellence: how to expand access to research publications/Access Mechanisms

From Wikisource
Jump to navigation Jump to search
←
Accessibility, sustainability, excellence: how to expand access to research publications
Access Mechanisms
→

​7. Access Mechanisms

7.1. We have identified three core mechanisms through which access to research publications can be increased: open access publishing, extensions to current licensing arrangements, and repositories. Each of them has a number of variations in nature and scope, and we discuss those variations, as well as the advantages and disadvantages of the three mechanisms in this section.

Open access journals

7.2. The key features of the current open access publishing landscape have been outlined earlier in Sections 3 and 4:

i. the launch of open access journals published by new entrants to the market such as PLoS and BioMedCentral
ii. the response of established publishers, with the launch of their own open access journals and, more commonly, of ‘hybrid’ journals operating on a mix of subscriptions and APCs for open access publication
iii. take-up which currently runs at between c5% and c8% of the global total of peer-reviewed articles published each year, with higher levels in science, technology and medicine, and lower levels in social sciences and humanities
iv. the relatively low levels of take-up until now of the open access option offered in most hybrid journals
v. the large open access publishers funding their journals through APCs which currently average between £1k and £2k, alongside a long tail of small publishers which publish one or two journals, many of which charge no APCs at all
vi. the recent growth of ‘repository’ journals which publish any articles which pass a peer review test of methodological rigour, regardless of the significance of the results
vii. the ability of open access journals, since they receive the bulk of their revenues before publication, to be less restrictive than subscription journals about rights of use and re-use of their contents.

7.3. For open access and hybrid journals, as for all journals, unit costs depend on a number of factors, including the rejection rate, frequency of publication, the average length of articles, and the amount of editorial material they provide in addition to research articles. All these factors therefore have an influence on the level of APCs; and journals considering a move from subscription-based to open access publishing, have to take careful account of them. The rejection rate is the most important influence in most cases, but for many journals, the amount of commissioned content that they provide—review articles, book reviews and so on which would not attract revenues in the form of APCs—will also be an important ​consideration. And in setting APCs, publishers will take account of the levels already set in the open access market as well as their current cost base and their status and reputation. One option could be to make research articles open access (funded by APCs), but to charge for access to the editorial content, reviews and so on. That is the basis on which some major journals such as the British Medical Journal already operate.

7.4. Among the large open access publishers, APCs for journals published by PLoS for 2011-12 range from $2,900 for PLos Medicine and PLoS Biology to $1,350 for the ‘repository’ journal PLoS One; and APCs for journals published by BioMedCentral range from $630 to $2,620. For the Hindawi Publishing Corporation, APCs range from $300 to $1500.[1] The high-status journals published in hybrid format by Cell Press, on the other hand, charge an APC of $5,000 for articles published on open access terms.[2]

7.5. The average level of APCs paid by the Wellcome Trust under its open access policy in the first three months of 2011 was £1,422; and the University of Nottingham paid on average £1,216 in the academic year 2010-11. How sustainable such averages would be if open access were to become more widespread among journals with high rejection rates, as well as in the humanities and social sciences, is not clear. There could be upward pressure on prices as such journals adopt an open access option; but on the other hand market competition could keep APCs low. Despite this uncertainty about the future, the evidence to date indicates that in the current market place it is possible for at least some open access journals to operate on a financially-sustainable basis.

7.6. Hence it is not surprising that a number of publishers of major journals, including learned societies such as the Institute of Physics, have already established open access journals, or moved to a hybrid model for at least some of their publications. Few have established fully open access journals as yet, however, in the humanities and social sciences; and take-up of the open access option in hybrid journals in those disciplines has been very low. Indeed, a report[3] on the journals published by a number of leading societies in the humanities and social sciences in the US found that factors including the rates of publication and of rejection of submitted manuscripts, the length of articles, and the large amounts of material—such as book reviews—that would not attract an APC, meant that a move to fully open access journals would be unsustainable: the level of APCs would be too high, and it was not clear whether funds would be available to meet them.

7.7. Recent analysis of some leading social science journals published by learned societies in the UK[4] leads to similar conclusions, especially where—as is common with many of the journals published by societies—a large proportion of ​the current subscription income comes from overseas. Hence the suggestion that open access might apply only to the research articles published, not to the reviews and other material. Another suggestion is that open access might be restricted to the UK (together with those developing countries that already enjoy access under one of the Research for Life[5] and similar schemes), in which case the level of APC would be much lower, and potentially sustainable. Such a move would not, of course, meet the objective of increasing global access to UK research outputs.

7.8. A third suggestion is that instead of charging an APC once an article has been accepted for publication, journals should levy a fee when authors submit an article. Submission fees are already quite common in certain disciplines, notably economic and finance journals and in some areas of the life sciences. A recent report[6] found that that there could be benefits to publishers in certain cases (particularly for journals with high rejection rates) to switch to such a model, not least in enabling them to set APCs much lower than they would otherwise have to be. But the risks, particularly those involved in any transition, are seen by publishers to outweigh the perceived benefits. Moreover, the advantages offered by submission fees do not provide publishers and authors—who might decide, after paying for peer review not to proceed to publication—with direct incentives to change to open access. From the perspective of authors and of funders, the financial risks of submitting an article for publication would become greater under such arrangements. We have therefore not considered submission fees as an option in our deliberations.

Policies and arrangements for payment

7.9. We have already noted (Section 4) that policies and arrangements for the payment of APCs are unsystematic and ill-understood, and that they are thus a major barrier to the adoption of open access publishing. We therefore welcome the proposals to address this issue that are emerging from discussions with the Research Councils. The precise policies and processes have still to be worked out, but it is essential that they should allow flexibility to universities, so that they can establish their own policies and procedures for the payment of APCs; if they do so, they will provide a significant stimulus to open access publishing.

7.10. It is envisaged that universities should respond to the proposed new policies and arrangements from the Research Councils with policies of their own to establish open access publication as the primary means of publishing and dissemination, with dedicated institutional funds to support it. If universities are allowed sufficient flexibility in the use of moneys from the Research Councils, Funding Councils and other sources, the new policies may be adopted both for research projects funded by the Research Councils or other external funds, and also for the research that has no dedicated source of funding, where the costs are met from the ​university’s block grant and other resources. In pursuing this path, universities will have to consider, and to consult carefully with their staff, about the precise polices and arrangements that they put in place. For while there are advantages in making researchers and others more aware of the costs of the publication process, they are likely to be nervous about the implication that universities will have significantly greater influence on the specific channels they use to publish and disseminate their work. Moreover, in managing publication funds, universities will have to work together with authors, and in line with the principle of academic freedom, in making judgements about the potential for publication in journals with different levels not only of status, but of APC: cost of publication will thus be a significant consideration for the first time on a large scale and across all disciplines.

7.11. In establishing new arrangements, it will also be important for funders, universities and publishers to work together on three key issues. First, policies and procedures should be agreed and implemented for the high proportion of articles that are produced by authors from more than one institution (often several), and often with multiple sources of funding. Nearly half (46%) of the peer-reviewed articles with a UK author published in 2010 also listed an author from overseas. No clear policy stance has yet emerged for dealing with the growing proportion of publications that are produced in this way. If open access publishing is to grow significantly, all those involved—authors, institution, funders and publishers—need to have clear guidelines on how responsibility for the payment of APCs is to be allocated, or shared, in the various circumstances that can arise with co-authorship; and on the arrangements for payment.

7.12. Second, the transaction costs involved with payments for the 120k articles published by UK authors each year must be minimised, with arrangements for aggregating payments wherever possible and appropriate. Universities, funders and publishers should work together on this, with support from subscription agents and others such as JISC Collections as appropriate. Membership and similar schemes may also have a role to play here.

7.13. Third, all players in the research communications landscape will have to work together to establish policies and arrangements for dealing with publications by researchers with no institutional affiliation, and no sources of funds from which to meet APCs. This is likely to be a particular issue in areas of the social sciences and humanities where the tradition of the independent scholar remains strong. PLoS, BioMedCentral, Hindawi and other open access publishers already have arrangements under which complete or partial waivers of APCs are provided to authors who do not have the funds to meet them.

Costs

7.14. The costs to the UK of a significant speeding-up of moves towards publishing in open access journals will depend on a number of factors. Modelling undertaken by Cambridge Economic Policy Associates (CEPA) in 2010 for the Open Road study has been revised and updated for the purposes of this report. The new modelling ​takes account of the latest estimates of the numbers of articles published by UK authors and worldwide in 2010: 123,594 and 1,935,954 respectively.[7] The starting point for the analysis presented in a series of tables in Annex E is that APCs are set at a level—an average of c £1,450—similar to that currently being paid by the Wellcome Trust. It is important to stress that the adoption of such a starting point does not amount to a recommendation; rather, it is simply a point from which the analysis of possible scenarios can begin.

7.15. The comparisons in the tables start also from the same assumption used in the ‘gold’ open access scenario in the Open Road report: that c23.3% of all articles published annually across the world are published under gold open access terms, and that all countries adopt publication of research in open access journals at the same rate. Two further limitations to the modelling should be noted.

i. The model assumes that the costs of subscriptions will fall in proportion to the increase in the number of articles published open access; it is likely, however, that during the transition to open access, universities and other organisations will maintain subscriptions even as their expenditure on APCs rises. This will occur especially if a significant proportion of open access articles are published in hybrid journals, where much of the content will remain accessible only to subscribers.
ii. The model is not dynamic; it compares costs against the starting point set in relation to funding and the numbers of articles produced in 2010, and does not seek to model changes over time (it takes no account, therefore, of the annual rise in the number of articles produced worldwide each year, currently running at between 3% and 4%).

7.16. While bearing in mind all the points outlined above, it is important to note how the modelling indicates that, at the level of APCs currently being paid by the Wellcome Trust, a significant shift to open access journals could be cost-neutral for the HE sector as a whole—although not necessarily for individual institutions—in the UK. For the modelling indicates that if open access publishing funded by APCs were to cover up to a quarter of the total of articles published each year in the UK and worldwide, the costs to the HE sector in the UK would be minimal, and that there would be cost savings in other sectors of c£5m a year, so long as the average level of APCs were to remain at c£1,450 or lower, and the rest of the world was not too far behind the UK in take-up. We consider some other scenarios below.

7.17. Savings to the HE and other sectors, of course, would be achieved in the main through reduced revenues to publishers, including learned societies. As we have noted earlier, there may be upward pressure on prices as open access becomes more widespread among prestigious journals with high rejection rates and thus higher costs. But market competition may tend to counteract such pressure, and ​since the proportion of articles published in journals with very high rejection rates is relatively low, their impact on the average level of APCs is likely to be correspondingly small.

7.18. Nevertheless, we have considered a number of variations to our model, and the impact on the costs and/or savings to the HE and to other sectors. The tables in Annex E examine four sets of possible variations to the starting point for analysis described above:

i. changing the average level of APCs by between 10% and 50%;
ii. changing the level of take-up of open access publishing both in the UK and the rest of the world from 23.3%, considering levels between 10% and 50%;
iii. allowing for lower levels of take-up of open access publishing in the rest of the world, as compared with the UK;
iv. taking account of the high-proportion (c46% in 2010) of articles published by UK authors which included also an author from overseas, and varying the proportion of UK-authored articles for which the full cost of the APC would be borne in the UK.

7.19. The tables indicate that under all but two variations from the point at which the analysis starts (thus only if the level of take-up in the rest of the world were to be as low as 40% or less of the UK rate), there would be cash savings to individuals and organisations outside the HE sector, resulting in the main from reduction in revenues for publishers. Those cash savings would be in addition to the wider benefits such organisations would receive as a result of higher levels of access to journals. But several of the variations would lead to increased costs for the HE sector.

i. If the average level of APCs were to be c£2,175, rather than £1,450 (i.e. 50% higher than the starting point for our analysis), the HE sector would face additional costs of £11m a year, on top of the £175m currently being spent on journals and providing access to them. There would still, however, be savings to other sectors.
ii. Varying the level of take-up of open access publishing in the UK and the rest of the world, so that it reaches 50% of the global total of articles published each year, would have no impact on costs to the HE sector, so long as the average level of APCs remained at c£1,450. But the cash savings to other sectors would rise significantly, to nearly £16m a year.
iii. If as a result of measures to accelerate the transition to open access publishing, the level of take-up were to be significantly higher in the UK than in the rest of the world, there is the risk that the UK, and the HE sector in particular, would bear significant costs, while reaping only some of the benefits. Articles from UK authors would be made available around the world open access; but UK universities and other organisations would still ​have to pay for access to a significant proportion of articles published by overseas authors. In broad terms, if just under a quarter of UK-authored articles were to be published open access, but only 5% of articles in the rest of the world, the HE sector in the UK would face additional costs of c£17m a year, and organisations in other sectors which produce research articles would also face additional costs, amounting to c£3.5m.
iv. If UK institutions were to have to pay the full APC for only some of the articles produced by UK authors in collaboration with researchers in other countries, the costs to the HE sector could fall significantly. It is estimated[8] that of all the articles published with a researcher from the UK listed among the authors, around 65% have someone from the UK listed as the corresponding author (which may serve as a proxy for the lead author). Reducing by 15% the proportion of all UK-authored articles for which a UK body should pay an APC (that is, for around a third of the articles where there is also an author from overseas), would reduce costs to the HE sector by nearly £4m a year as compared to the case from which our analysis starts.[9]

7.21. It will be clear from this analysis that the costs to universities of a significant acceleration in the transition to publishing in open access or hybrid journals depend critically on assumptions on four factors:

i. the average level of APCs;
ii. the extent to which the UK is ahead of the rest of the world in adopting open access publishing;
iii. the number and proportion of articles with overseas as well as UK authors for which UK institutions would be required to pay an APC; and
iv. the extent to which during the transition to open access, universities and other organisations are able to reduce their expenditure on subscriptions even as their expenditure on APCs rises (a factor which is not covered in the modelling).

7.22. Under optimistic assumptions about levels of take-up and payment of APCs overseas, where the pace of change in the UK is matched in the rest of the world, and a proportion of the costs of APCs for articles co-authored with researchers in other countries is offset by funders and institutions in those countries, the costs to the HE sector of moving to open access publishing for 50% or more of research articles would at worst be minimal. There could even be cash savings for the HE sector, again so long as the average level of APCs is £1450 or lower. And our modelling suggests that even with less optimistic assumptions, the cost savings to organisations in other sectors would be substantial. The essential risk borne by the ​HE sector would be that it would be unable to reduce its expenditure on subscriptions at the same rate as it increased its expenditure on APCs.

7.23. Under more pessimistic assumptions about take-up, with rates of adoption twice as high in the UK as in the rest of the world, the costs to the HE sector would be significant, particularly if the average level of APCs were to be high too. Our modelling suggests that if APCs were on average £2.2k, half of all UK-authored articles were published open access, but only a quarter in the rest of the world, and the UK paid the full APC for all articles with a UK author, the additional cost to the HE sector could be over £70m a year.

7.24. In a middle ground, we have modelled a scenario under which the average level of APCs is c£1.75k, the rates of adoption in the UK are (at least for a transition period) as much as twice those in the rest of the world, and the UK secures contributions from overseas towards the costs of APCs for at least half the articles published with international co-authors. Our estimate is that the additional costs to the HE sector if half of all UK-authored articles were to be published in open access or hybrid journals under this scenario would be of the order of £38m a year, allowing in addition to the figures presented in Annex E for some ‘stickiness’ in costs as universities have to maintain their expenditure on big deals and other licences even as their expenditure on APCs rises.

7.25. The cost implications for individual universities will vary, as we noted earlier, according to the extent to which they can recover the cost of APCs from the Research Councils and other external funders of research; their size and research-intensity; their mix of disciplines; and their current expenditure on the library and its contents.

7.26. The establishment at universities such as Nottingham of funds to meet APCs has led to some attempts to assess the point at which such funds might become financially sustainable for different universities.[10] But the scope for reducing expenditure on subscriptions without compromising levels of access is currently very limited; hence unless universities can recover their expenditure on APCs through the full economic costs they seek from research funders in grants for research projects,[11] their publication funds at present represent a drain on university resources.

7.27. That picture would change dramatically if the Research Councils were to establish, as they have signalled, new and flexible funding arrangements to meet APCs, especially if they were to stimulate other major research funders to act similarly. The essential point here is that the new arrangements should provide a sound basis on which universities could establish publication funds: and if all funders were to meet the full costs of APCs, the net cost to the university would be nil. But it is critically important that universities should be given sufficient scope to establish ​their own policies and funding arrangements, which will provide incentives for them to shift funds from library budgets to the payment of APCs, and to bear down on the cost of those payments.

7.28. That flexibility is particularly important in allowing universities to deal with publications arising from the large proportion of research, particularly in the humanities and social sciences, which is undertaken without any dedicated funding from external sources. In that case the university would still have to meet the costs of APCs from QR block grant and other sources available to it; and for a university where a high proportion of research is in the humanities and social sciences, the cost implications could be significant. An analysis of the impact on a research-intensive university is presented at Annex F.

Extensions to licensing

7.29. Subscriptions for licences for journals are the only route through which users can get access free at the point of use to the articles they publish that are not accessible either through a repository or through an open access or hybrid journal. Institutions from across all sectors in the UK paid in 2010 some £150m for such licences. Licensed access has increased enormously in the past decade, but as we saw in Section 4, it remains patchy across the UK, particularly outside the HE community and some parts of the large corporate and health sectors. The licensing system currently falls far short of providing ‘universal access’ to all citizens and organisations in the UK. However, since UK researchers are responsible for only 6% of the global total of such articles, and an immediate or even rapid global shift to a wholly open access environment seems unlikely, licensing will remain a key route to access at least for the short to medium term. In order to increase access, therefore, it will be important to secure some extensions to current licensing regimes.

7.30. There are three key dimensions to any such extensions to licensed access: the numbers of individuals and organisations within and across different sectors who have access to licensed content; the volumes of content—both journals and articles[12]—to which they have access; and the rights that users have once they gain access to the content. We consider each of those dimensions below.

Higher Education

7.31. No single university purchases licensed access to all the c25k journals and the 1.9m articles published worldwide each year. Staff and students in the largest and most research-intensive universitiesenjoy licensed access to a high proportion of them, especially those covering the subject areas in which they are active. For staff and students of other institutions, however, the amount of content to which they ​have access varies considerably, in accordance with the funds they have made available for the necessary licences (Section 4).

7.32. The past three years have seen a growing interest in the UK in licensing models under which access is provided not to a single university, but to a consortium. Such models are reasonably common in a number of other countries, including the US and Scandinavia.[13] But the deals negotiated nationally by JISC Collections under the NESLi2 initiative operate on an opt-in basis: individual universities decide whether or not to take up the licence at the price offered. One of the difficulties in implementing a consortium model where access is shared across all members is the allocation of costs between institutions which may differ in size, research-intensity, and subject profile.[14]

7.33. In Scotland, however, libraries for all nineteen HE institutions launched in 2009 a consortium scheme under which they have jointly purchased licences currently covering nearly two thousand journals from eight publishers. The apportionment of costs for the different licences between the Scottish HEIs is based on the historic expenditure of each institution with each publisher. Whether this apportionment model would be sustainable if joint licences were to cover a significantly larger number of publishers and journals is not yet clear. Similarly, it is widely assumed among librarians and others in the HE sector in the UK as a whole that reaching agreement on cost allocations across the much larger number of all the universities in England, for example, would prove extremely difficult.[15].

7.34. Nevertheless, the success of the SHEDL experiment has stimulated discussion about the scope for similar consortia to be established covering groups of universities in other parts of the UK; and renewed discussion about the merits of licence arrangements which provide access for the whole of the HE sector. There have been initiatives of this kind in a number of countries, including Germany, Ireland and Canada.[16] The widespread view among university librarians in the UK, however, is that the apportionment between them of the costs of such licences would best be achieved by top-slicing of their universities’ block grants from the ​Funding Councils. Such top-slicing would run counter to the policies—supported by successive Governments as well as by universities themselves—of funding universities in full and allowing each of them to manage expenditures as it sees fit. In the absence of top-slicing, other arrangements might involve universities’ agreeing—as they have done in Scotland—to put amounts equal to their current expenditure on journals into a central pot.[17]

7.35. Publishers have indicated that they could provide licences for the whole HE sector for access to all the content currently accessible only to large research-intensive universities. They estimate that such licences would cost an additional 5-10% on top of the amounts currently being paid by the sector. The additions to current prices charged by individual publishers in order to extend their licences in this way would depend, of course, on the extent of their current coverage, as well as on any modifications to current rights of use and re-use. An extension of licensed access across the whole sector, however, would bring undoubted benefits to researchers in less-well-endowed universities, although the Open Road report[18] suggests that the unit costs of the increased amounts of access (for a sector which already enjoys high levels of access) are high compared with other possible routes.

The health sector

7.36. The provision of licensed access to journals for the NHS is complex, with a range of local as well as central initiatives; and we noted in Section 4 the estimate that on average across the NHS, only about a third of relevant journals are available free at the point of use.

7.37. Many people in the NHS—doctors and other medical staff but also student doctors and nurses—also have an affiliation to a university; and a significant proportion of the content purchased for them by the NHS is available to them also via their university, though usually on a different platform, with different arrangements for access. On the other hand, the amount of content accessible through NHS licences is significantly lower than that for the HE sector.

7.38. There would be undoubted benefits from increasing and rationalising arrangements for licensed access across the health sector, with greater co-ordination between the NHS and the HE sector. There have been attempts over many years to achieve greater co-ordination, but the different procurement systems in the two sectors present a challenge for those seeking that end. Nevertheless, the two sectors in Scotland are currently planning to work together to examine the scope for collaborative purchasing as a key step towards creating a unified access system.

7.39. Estimates provided by publishers for the Open Road[19] study indicated that licensed access to relevant journals for the whole NHS could be provided at ​relatively modest additional cost of around £1m a year. The report suggests that the benefit-cost ratios would again be modest; but the benefits would nevertheless be real, especially for those NHS staff who need access to the latest research publications and who struggle at present with access to a limited amount of content on different platforms.

Other sectors

7.40. Outside the HE and NHS sectors, the provision of licensed access to significant numbers of journals is common only in large R&D-intensive companies. The survey and other evidence we have considered (Section 4), together with data on the number of ‘turnaways’ on publishers’ platforms (that is, the number of people who view the abstract of an article but then decline to purchase access to full text)[20] indicate that PPV arrangements at current prices are not an adequate substitute for licensed access free at the point of use.

7.41. During the period of transition to publishing in open access and hybrid journals, extensions to licensed access for the benefit of individuals and organisations in the public, voluntary and business sectors in the UK, would bring significant benefits in increasing the flow of knowledge and thereby in stimulating growth and innovation.

7.42. There are many attractions to the idea of a national licence to provide online access to all journals for everyone in the UK, although some risks as well, which is probably why no major nation has implemented such a scheme. We have concluded that such a licence is unlikely to be practicable, and that the costs would probably be high. The only known example of such a national licence scheme is the Iceland Consortium for electronic subscriptions (hvar.is), which provides access to over 17,000 full-text journals through every computer in the country that connects to the internet through an Icelandic internet service provider.[21] The agreement is made through an aggregator, not with primary publishers, with most of the content embargoed for a year after publication; and the model is unlikely to be scaleable for the UK.

7.43. Nor would it be straightforward to develop and implement licences that would cover large sectors of the UK economy and society: there would be considerable problems in defining different sectors and their boundaries; the risk of disputes as to which organisations fell within or outside the definitions; and the likelihood of leakage of content beyond the sectors covered by each licence. Nevertheless, in order to increase access for key groups of people and organisations who have an interest in research and its results, some extensions to current licensing arrangements would clearly be desirable, and could bring real benefits. We believe that there should be continuing discussions between publishers, representative bodies for key sectors, libraries and other organisations with relevant expertise ​(such as JISC Collections) to consider the terms and costs of broader licence agreements; and possible sources of funding.

7.44. Two specific proposals have emerged from our discussions. The first would address the needs of those small companies and other organisations that are research-intensive and have close relationships with universities, including spin-out companies. Under the current licences negotiated under the NESLi2 initiative, the staff of such companies are not eligible for desktop access to content licensed by the university with which they have a relationship. But it is unlikely that the companies can afford large licence packages themselves, or the high cost of PPV. The proposal is that on the basis of an agreed definition of small businesses engaged in research and development,[22] and in return for a relatively small extra fee, publishers might allow a university to provide access to researchers in such enterprises. The university itself could then decide how it would seek to recoup from the companies concerned the additional costs involved in the licence. Such a move would be in line with the recommendations of the Wilson Review on ways to improve collaboration between universities and business.[23]

7.45. The advantages of such an approach would be that the staff in such enterprises would benefit from efficient and low-cost access; universities would strengthen their partnership with those enterprises; both would strengthen their contribution to research, innovation and economic growth; and the publishers would secure a small increase in revenue, with the potential for greater increases as the businesses grew.

Public libraries

7.46. The second proposal is that the major subscription-based publishers should license public libraries throughout the UK—and perhaps in addition those learned society libraries that are open to the public—to provide access to peer-reviewed journals and conference proceedings at no charge, for ‘walk-in’ users on library premises. Provision through public libraries in this way would enhance the walk-in access already available via university libraries[24] and would enable anyone to have access to peer-reviewed research literature at their local public library. At a time when public libraries are under severe pressure such a move will help to strengthen their position in the communities they serve, and lead to increased usage and value. It would have an immediate effect in extending access to the great majority of journals for the benefit of everyone in the country. Hence the proposal has been warmly welcomed by representatives of the public library sector.

​ 7.47. Walk-in access would not, of course, meet the demand for access at any time and anywhere. But access free of charge to any user of a public library would provide real benefits to many people who at present face considerable barriers if they want to find authoritative information about research relevant to their interests and needs. At the very least it would be a valuable—and free—supplement to the current access options of PPV from the publisher’s platform, document delivery services such as those provided by the British Library, and other services such as DeepDyve.

7.48. It is proposed that this public library initiative should run for an initial period of two years, in order to gather and analyse data on demand and usage; and publishers hope to extend the service at the end of the two years if it has not led to any damaging loss of core revenues. The precise terms of what will be provided—whether access will be restricted to screens on library equipment, restrictions on copying to other devices, access to printing, and related matters—and issues such as discoverability and whether access will be provided to all content via a single platform, have yet to be worked out. A working group of representatives of public libraries and of publishers has been established to consider these issues, and how the proposal can be implemented to best effect.

7.49. If the initiative is to achieve its full potential impact, it will need to be accompanied by the development of clear guidance and advice for both users and the staff in public libraries on the nature and scope of journals and their contents, and on how to navigate to relevant articles. A clear marketing strategy will also need to be developed and implemented to ensure that those who are interested in gaining access to journals are aware of the initiative.[25] With all those measures in place, the initiative is likely to have a major impact.

Content coverage

7.50. In considering extensions to current licensing arrangements, it is important, as we noted earlier, to consider the amounts and proportions of content, as well as the sectors, that are covered. There could be a natural tendency in seeking to extend he numbers of people who have licensed access to focus attention on the larger publishers who control the majority of content in the form of journals and articles. It will be important, therefore, to put measures in place to protect the interests of large numbers of smaller publishers—with journals that are valuable in their fields—who would find it more difficult than their larger colleagues to engage in negotiations on extending their licence agreements to cover more people and organisations; and to make sure that as many people as possible have access to as wide a range of journals as possible, including those published by the smaller publishers.

Costs

​ 7.51. The costs of extensions to current licensing in the UK would depend on the scope of the extensions. Our estimate is that licences for access to the great majority of journals for the whole HE sector in the UK would cost £6-12m a year on top of what is currently being paid by universities and other HEIs; and that licences for relevant journals for the whole NHS would cost £1-2m in addition to what is currently being paid.

7.52. We have not attempted to estimate the additional costs—on top of the c£35-40m currently being paid by organisations outside the HE sector—of licences to cover other sectors such as Government and the public sector; voluntary organisations; or business in general and small and medium-sized enterprises (SMEs) in particular. We believe, however, that the costs could be relatively high, not least because publishers would seek in their pricing for sector-wide licences to protect themselves against the loss of potential for additions to their revenues from individual organisations within the relevant sectors. Hence it is important that key representative bodies for the public, business and voluntary sectors should work with publishers to identify the business case for sector-wide licences, including the possible sources of funding to support them. And we believe that there may be scope for negotiating trade-offs between increased revenues for publishers from extended licences on the one hand, and the amounts paid in APCs for articles published open access on the other. (See Section 8)

7.53. The costs to the public purse of the two proposals that have emerged from our discussions, however, would be minimal. For the provision of access to micro enterprises via universities, individual universities would be able to recoup the cost by charging a fee for access that would be free at the point of use. For the public library initiative, most of the costs would be borne by the publishers, but there would be a need to produce some guidance and promotional material to raise awareness of what is being provided.

Repositories

Institutional repositories

7.54. Repositories come in a number of forms, as we noted in Section 4. Most universities in the UK now have an institutional repository, though there are considerable differences in size and scope of holdings, and levels of usage. The policies of neither research funders nor universities themselves have yet had a major effect in ensuring that researchers make their publications accessible in institutional repositories as a matter of routine: levels of deposit as yet remain low, and for journal articles in particular, most of the records in institutional repositories tend to consist of metadata rather than full text.

7.55. Hence the impact of institutional repositories in increasing access to research publications has so far been limited, despite the best efforts of repository managers and others; and without further active measures from funders and universities, that seems unlikely to change. Such measures could well be warranted, however, since ​set-up costs have already been incurred, and the evidence suggests that operating costs are modest. And for universities, there are benefits in providing a showcase for their research, and a mechanism for creating a central record of publications and other outputs.

7.56. Institutional repositories make use of a number of different software platforms, which means that users encounter different platforms and interfaces, and that cross-searching and navigation can be difficult. Most UK repositories nevertheless comply with the Open Access Initiative Protocol for Metadata Harvesting (OAI-PMH), which allows basic metadata to be harvested to support discovery and navigation services. Repository content is increasingly exposed to and harvested by Google and other search engines; but the kinds of search and navigation functionality provided by services such as Web of Knowledge[26] or SCOPUS,[27] and by other more specialised services, are not available for those seeking material in repositories; search and navigation facilities are very limited by comparison.

7.57. There are a number of international initiatives to improve interoperability between repositories, through organisations such as the Confederation of Open Access Repositories (COAR)[28] and DL.org.[29] At a European level, the Driver project,[30] the second phase of which ended in December 2009, established a pan-European infrastructure for digital repositories, offering a range of sophisticated functionalities for researchers. Driver sought to “establish the successful interoperation of both data network and knowledge repositories as integral parts of the E-infrastructure for research and education in Europe.” Building on this, the OpenAIRE[31] initiative supports the development of a network of repositories; it provides a portal for access to resources stored in these repositories, and guidance to ensure that repositories are compliant with a set of Europe-wide standards, especially relating to metadata (in order to facilitate cross-searching and harvesting). It works within the context of the EU’s open access pilot in the FP7 Framework programme, and the European Research Council’s Guidelines for Open Access.

7.58. In the UK, JISC has funded[32] demonstrator projects on interoperability, aimed at developing “realistic scenarios regarding repository use [with] a range of computer-computer interfaces between repositories and related services and ​systems.” In 2009, JISC also funded a project[33] to investigate interoperability between repositories and online library catalogues. More recently, it has funded an Open Access Repository Junction[34] “to scope, build and test a deposit broker tool to assist deposit into, and interoperability between, existing repository services.” This is intended to simplify workflows for authors and publishers who wish to deposit material in more than one repository. JISC has also worked on interoperability issues with analogous bodies in Germany, the Netherlands and Denmark through the Knowledge Exchange, which includes an interoperability of digital repositories (IDR) working group.[35]

7.59. The repository metadata landscape remains confusing, however, and the UK repository community in universities does not have a clear understanding of the requirements arising from initiatives such as OpenAIRE, and the Common European Research Project Information Format (CERIF). JISC is therefore working with RCUK and others on guidance to institutional repositories on an enhanced metadata set.[36]

Subject repositories

7.60. The issues for subject-based repositories tend to be rather different. They have had a significant impact in a number of subject areas including physics (through ArXiv), and the life sciences and medicine (through PubMedCentral and UKPMC). The most successful repositories have been able to develop good search and navigation facilities, but these remain a challenge for others that have fewer financial resources to invest in such services. Overall there remain many gaps in the provision of subject-based repositories; many subject areas lack them entirely, or have only small-scale repositories which have not reached the critical mass to make them effective routes to access for more than a relatively small band of enthusiasts.

7.61. It is important, moreover, to note the characteristics of the most successful repositories. In physics, ArXiv operates in the main as a pre-print repository, where researchers deposit and gain access to draft papers before they are submitted to journals for peer review and publication. The repository and the journals thus co-exist, each with their distinctive roles. In medical and biological sciences, PubMedCentral and UKPMC have been established by the major research funding agencies in their domain in the UK and the US as key mechanisms to support their access policies; and for the funders of UKPMC, that service sits alongside their support for open access publishing. Again, the repository operates in tandem with ​publishers, who undertake the bulk of the work in depositing their publications in UKPMC.

Funders’ policies

7.62. The UK Research Councils and other major funders such as the Wellcome Trust require any peer-reviewed publications arising from work they fund that are not published in open access or hybrid journals to be made accessible via a repository as soon as possible. The policies vary in detail, and among the Research Councils, only the MRC currently specifies a maximum embargo period, of six months; the other Councils require deposit and access in compliance with licensing and copyright arrangements. There are differences also in requirements as to precisely what is deposited: the submitted manuscript, the accepted manuscript, or the published paper.

7.63. Research Councils’ current policies have been in place since 2006. Given the timing of applications and awards, the average length of grants, and the delays before publication, it is only in the last couple of years that assessing compliance has been feasible on any kind of systematic basis. The Councils have recently established systems for comprehensive reporting on publications and other outputs,[37] and it should be possible to check compliance levels systematically from next year. The evidence suggests, however, that rates of compliance are at present generally low. The Councils have been considering how to consolidate their policies, and they have recognised—as they have been required to do by Government[38]—the need to make more efforts to raise awareness of their policies across the HE and research communities. Proposals have been circulated which include a requirement that publications should be made freely accessible either immediately upon publication, with unrestricted rights of use and re-use, where an APC is paid; or, where an APC is not paid, within six months (twelve months for publications arising from work funded by the AHRC and the ESRC, at least for an interim period,).

Publishers’ restrictions

7.64. For open access publishers which receive their revenues in the form of APCs before articles are published, repositories arouse few concerns: they regard them as complementary channels for disseminating the articles, and hence allow access to them via repositories as well as via their own publishing platform. Subscription-based publishers, on the other hand, tend to regard repositories as rival channels and as a threat to their subscription revenues and thus to the viability of their journals. They have therefore responded to the rise of repositories with a range of policies that reflect those concerns. Most impose a range of constraints, in the form of embargos, restrictions on what version of a paper can be deposited, and on the ​uses that can be made of it. Those restrictions serve to limit—as they are intended to do—the usefulness of what is made available to readers via repositories. Hence as we noted in Section 4, evidence as to any potential impact on the viability of journals arising from the access provided via repositories under current restrictions is as yet not clear; journal publishing has continued to grow in recent years.

7.65. As to what version of a paper that can be deposited, relatively few subscription-based publishers allow the version of record—that is, the version finally published, with the functionality associated with links and semantic mark-up—to be deposited and made accessible. Those few that do—as, for example with the British Medical Journal—allow such deposit only after an embargo period.[39] Most other publishers allow either the submitted or the accepted (after peer review) manuscript to be deposited; and policies vary as to which of those two it should be.[40] Some of the major publishers co-operate with the NIH by depositing in PubMedCentral versions of the articles they publish, but with a disclaimer making clear that what is accessible there is not the version of record, which remains accessible only from the publisher’s site. Highlighting the status of different versions of the article in this way is now complemented by the CrossMark service[41] which puts a kitemark on the version of record in its most up-to-date form.

7.66. Funders have in general sought embargo periods of twelve months, and publishers of subscription-based journals are very concerned at any moves to reduce that period, believing that it would lead to a loss of subscriptions that would put the viability of their journals at risk. The concerns focus on the half-life of journals in terms of downloads: the length of time it takes the articles in each volume to reach half the number of downloads they will reach in total. Some major publishers have supplied us with figures which indicate half-lives varying from two-and-a half years in fast-moving fields such as computer science to eight years in mathematics.

7.67. Publishers have also noted that the availability of articles via the large subject-based repositories such as ArXiv and PubMedCentral tends to reduce the number of downloads from publishers’ own platforms. That tends to increase the cost-per-download ratio for universities and others who pay for subscriptions for licensed access to the relevant journals via the publisher’s platform; and since that ratio is being used increasingly when universities review the journals to which they subscribe, some publishers are nervous about loss of subscriptions. On the other hand, evidence from the PEER project suggests that providing access to articles via repositories with high-quality metadata may lead to a marginal increase in ​downloads from the publisher’s site.[42] Nevertheless, a survey of librarians conducted by the Association of Learned and Professional Society Publishers[43] indicates that if embargo periods were to be reduced to six months, 10% of them would cancel all science, technology and medicine (STM) journals, and a further 34% would cancel subscriptions to some of them; the figures for arts, humanities and social science journals were 23% and 42% respectively.

7.68. Such evidence has reinforced the concerns of subscription-based publishers who may be, for whatever reason, unable to make a rapid move to open access publishing, that a reduction in the allowable embargo period to only six months, especially if combined with a requirement to eliminate any restrictions on use and re-use, would put the viability of their journals at severe risk.[44]

Use and re-use rights

7.69. We noted earlier that access is not just about the ability to read a publication, but about what users can do with the content: to analyse and manipulate it; to shift it from one format to another; to re-use and re-purpose it in many different ways to facilitate the creation of new knowledge. Use and re-use rights depend to a significant extent on the formats in which content is made available: the range of potential uses of a PDF file, for example, tend to be more limited than for content that is made available in HTML or XML. Word-processed text files in repositories may thus be much less ‘useful’ to users than more advanced formats. The key for researchers and many other users is that published content should be accessible in formats that are as easy to manipulate as possible; and that any restrictions on what they can do with the content should be minimal, if they exist at all. Researchers want the maximum freedom to use the latest tools and services to make the best use of the information to which they have access.

7.70. But for subscription-based publishers, re-use rights may pose problems. Any requirement for them to use a Creative Commons ‘CC-BY’ licence,[45] for example, would allow users to modify, build upon and distribute the licensed work, for commercial as well as non-commercial purposes, so long as the original authors were credited.[46] Publishers—and some researchers—are especially concerned about allowing commercial re-use. Medical journal publishers, who derive a considerable part of their revenues from the sale of reprints to pharmaceutical companies, could face significant loss of income. But more ​generally, commercial re-use would allow third parties to harvest published content from repositories and present them on new platforms that would compete with the original publisher.[47]

Costs

7.71. Since most universities in the UK have now established a repository, the costs of so doing may be regarded as sunk, although there will be a continuing need for investment and improvement.[48] A recent estimate of the annual operating costs puts them at between £26,000 and £210,000,[49] depending on the size of the university and its research community. As to subject repositories, the 2012 budget for ArXiv in the physics community is $589,000,[50] and the current cost for the NIH’s administration of PMC is put at $3.5-4.0 m. The annual running costs of UKPMC (excluding the support it receives from the US National Library of Medicine in ingesting articles from publishers) are c£600,000. It is important to note, however, that cost effectiveness depends critically on usage: the numbers of items uploaded into the repository, and downloaded from it. And it has recently been noted that “the limited resources devoted to repository management make it difficult to enhance services” with a detrimental effect on both efficiency and effectiveness.[51] The costs to universities of running fully-effective repositories thus remain unclear.

  1. ↑ Some Hindawi journals charge no APC at all. http://www.hindawi.com/apc/
  2. ↑ http://www.cell.com/cellpress/FundingBodyAgreements
  3. ↑ Mary Waltham, The Future of Scholarly Journals Publishing Among Social Science and Humanities Associations, 2009
  4. ↑ Confidential information provided to the secretariat.
  5. ↑ HINARI in medicine, AGORA in agriculture, OARE in environmental sciences, and ARDI in development and innovation. http://www.research4life.org/
  6. ↑ Mark Ware Consulting Ltd, Submission Fees: a tool in the transition to open access? , Knowledge Exchange, 2010
  7. ↑ Elsevier, International Comparative Performance of the UK Research Base 2011: a report for the Department of Business, Innovation and Skills, 2011.
  8. ↑ Elsevier estimate based on analysis of the SCOPUS database
  9. ↑ There would of course have to be clear guidance on the issues we highlight in paragraph 7.12 above.
  10. ↑ Swan, A (2010) Modelling Scholarly Communications Options: Costs and Benefits, JISC, 2010.
  11. ↑ Paying for Open Access Publication Charges, RIN and UUK, 2009.
  12. ↑ It is important to note that the profile of journal titles and articles is skewed: roughly a third of titles are responsible for 80% of published articles; and 50% of titles for 90% of articles. Steven Hall (2010): A commentary on ‘The economic implications of alternative publishing models’, Prometheus: Critical Studies in Innovation, 28:1, 73-84
  13. ↑ See, for example, OhioLink in the US (http://www.ohiolink.edu/about/ ), and the BIBSAM consortium in Sweden (http://www.kb.se/bibliotek/centrala-avtal )
  14. ↑ John Cox and Albert Prior, Bloc Payment Methods for Online Journals Agreement: Models for redistribution of costs, 2010( http://www.jisccollections.ac.uk/Documents/Reports/BLOC%20PAYMENT%20APPORTIONMENT%20REPORT%20(PUBLIC%20VERSION%20FOR%20WEBSITE).doc )
  15. ↑ http://scurl.ac.uk/WG/SHEDL/about.html#shedl_about . For an early evaluation of the SHEDL initiative, see One Year On: Evaluating the initial impact of the Scottish Higher Education Digital Library, RIN 2010,
  16. ↑ In Ireland, Science Foundation Ireland (SFI) and the Higher Education Authority (HEA) provide funds to support a national electronic library (IReL) (see http://www.irelibrary.ie/about.aspx ); in Germany, the Deutsche Forschungsgemeinschaft (DFG) has since 2004 given financial support for the purchase of licences for journals and a range of databases and e-book collections for the whole HE sector (see http://www.dfg.de/en/research_funding/programmes/infrastructure/lis/digital_information/library_licenses/index .html ; and in Canada, the Canadian Research Knowledge Network operates as a consortium of 75 research-led institutions, whose members are committed to licensing a broad portfolio of research content from multiple vendors (see http://www.crkn.ca/home )
  17. ↑ For an analysis of the cost-apportionment models used by various consortia, see John Cox and Albert Prior, op.cit
  18. ↑ Heading for the Open Road, RIN, JISC, Wellcome Trust, PRC and RLUK, 2011
  19. ↑ Ibid.
  20. ↑ The data on turnaways is difficult to interpret, but the numbers are very large.
  21. ↑ See http://www.hvar.is/sida.php?id=154
  22. ↑ Such a definition might start from the EU definition of a micro enterprise, with 10 staff or fewer, and a turnover of 2m euros or less. (http://ec.europa.eu/enterprise/policies/sme/files/sme_definition/sme_user_guide_en.pdf )
  23. ↑ Sir Tim Wilson, A Review of Business-University Collaboration, 2012, http://www.wilsonreview.co.uk/wilson-review/wilson-review.pdf
  24. ↑ Publishers are required to allow walk-in access through university libraries as a requirement of NESLi2 licences; but access provision by libraries is patchy, and take-up small. See Public access to licensed journals held in academic libraries, RIN 2006.
  25. ↑ Experiments in the Netherlands to provide access through libraries to SMEs have proved disappointing mainly, it is thought, because they received very little publicity.
  26. ↑ A citation indexing and search service with web linking provided by Thomson Reuters. For UK education, the service is mediated through MIMAS at the University of Manchester, http://wok.mimas.ac.uk/
  27. ↑ A bibliographic database service provide by Elsevier, http://www.scopus.com/home.url
  28. ↑ http://www.coar-repositories.org/
  29. ↑ http://www.dlorg.eu/
  30. ↑ http://driver-repository.eu/
  31. ↑ http://openaire.eu/
  32. ↑ http://www.jisc.ac.uk/whatwedo/programmes/reppres/interoperabilitydemos.aspx
  33. ↑ Duncan Birrell et al, Online Catalogue and Repository Interoperability Study (OCRIS), http://ierepository.jisc.ac.uk/430/1/OCRIS_Report.pdf
  34. ↑ http://edina.ac.uk/projects/oa-rj/index.html
  35. ↑ http://www.knowledge-exchange.info/Default.aspx?ID=290
  36. ↑ RIO Extension: Mapping Repository Metadata Requirements: http://www.jisc.ac.uk/whatwedo/programmes/di_researchmanagement/repositories/rioextension.aspx
  37. ↑ Some Councils, such as ESRC, established such systems some years ago.
  38. ↑ Innovation and Research Strategy for Growth, Department for Business Innovation and Skills, 2011, page 77. http://www.crossref.org/crossmark/index.html
  39. ↑ Hence relatively few versions of record are available directly through institutional or subject repositories.
  40. ↑ The Royal Society of Chemistry, for example, requires deposit of the accepted manuscript, while the Royal Statistical Society requires it to be the submitted manuscript.
  41. ↑ http://www.crossref.org/crossmark/index.html
  42. ↑ See Ian Rowlands et al, PEER usage study findings, presented at PEER End of Project Results Conference, Brussels, 29 May 2012. One suggestion is that the improvements to metadata involved in the project meant that articles were easier to find through search engines and other gateways.
  43. ↑ Linda Bennett, The potential effect of making journals free after a six month embargo, ALSPS and the Publishers Association, 2012
  44. ↑ They note further that the draft regulations recently laid before Parliament under Section 11(6) of the Legal Deposit Libraries Act 2003 allow access to digital works deposited in the legal deposit libraries to be delayed for up to three years.
  45. ↑ For an explanation of Creative Commons licences, see the Glossary.
  46. ↑ See http://creativecommons.org/licenses/by/3.0/
  47. ↑ That might offer scope for innovation by third parties including new entrants; but such innovation would put at risk the journals on which any new innovative services would depend.
  48. ↑ A recent study for the PEER project found it impossible to determine set-up costs for repositories, since there were large degrees of internal cross-subsidisation, and adequate records had not been kept. But since set-up costs for institutional repositories have largely been met for the HE sector in the UK, the lack of adequate cost information may be of little relevance. See, Centro ASK, Universita Bocconi , PEER Economics report, 2011 , available at http://www.peerproject.eu/fileadmin/media/reports/PEER_Economics_Report.pdf
  49. ↑ Swan, A, op cit. A survey by the Association of Research Libraries in North America puts the annual running costs of institutional repositories at between $8,600 and $500,000.
  50. ↑ Ricky Erway, Lasting Impact: Sustainability of Disciplinary Repositories, OCLC, 2012
  51. ↑ Centro ASK, Universita Bocconi op. cit.
Retrieved from "https://en.wikisource.org/w/index.php?title=Accessibility,_sustainability,_excellence:_how_to_expand_access_to_research_publications/Access_Mechanisms&oldid=10591674"