■ Visibility and promotion of the research output of institutions;
■ more complete records of scholarship;
■ access for the wider public including teachers, students, tax-payers;
■ alleviation of the access problems of the developing world;
■ creating fl exible and free re-use options for articles: open access articles may be read, downloaded, copied, distributed, printed, searched, linked to, translated as long as authorship is acknowledged and they are not used for profi t making. This will lead to a reduction in royalties paid to foreign publishing companies operating offshore, even for articles describing local research;
■ international access to research generated in developing countries which is currently invisible to the developed world and which is important for research on global problems;
■ alleviation of the budgetary problems of libraries all over the world, especially in the developing world;
■ contribution to the digital preservation of scholarship;
■ a shift from competition for authors rather than for subscribers, creating a healthier model;
■ facilitating peer review and other forms of quality control by giving reviewers, supervisors and examiners easy access to all the papers referred to in a new manuscript (Harnad 2005, Chan, Kirsop & Arunachalam 2005);
■ failed research can be reported, and may lead to the identifi cation and subsequent avoidance of blind alleys (Houghton 2005).
Not all role-players are enthusiastic, however. In November 2005 the Royal Society (London) issued a statement (http://www.royalsoc.ac.uk/page.asp?id=3882) cautioning against over- hasty pressurizing of researchers to publish in accessible archives, claiming that the business models on which some of these are based are not proven and urging restraint while more cautious investigation is carried out. That 61 Fellows (including 5 Nobel prizewinners) promptly issued an open letter (http://www.frsopenletter.org/) criticizing the Society’s stand on the matter, confi rms that Open Access Archiving is receiving acceptance from key stakeholders, but also that the debate is not yet over!
Figure 4: Maximized impact with Open Access (Harnad 2005) CHAPTER 5: GLOBAL ERESEARCH TRENDS AND THEIR IMPLICATIONS FOR SOUTH AFRICAN RESEARCH PUBLISHING
In November 2005 the Royal Society (London) issued a statement cautioning against over-hasty pressurizing of researchers to publish in accessible archives, claiming that the business models on which some of these are based are not proven and urging restraint while more cautious investigation is carried out
Impact cycle begins:
Research is done
Researchers write pre-refereeing
“Pre-Print” Pre-Print is self-archived in University’s Eprint Archive Submitted to Journal
Pre-Print revised by article’s Authors Pre-Print reviewed by Peer
Experts – “Peer-Review”
Refereed “Post-Print” Accepted, Certifi ed, Published by Journal
Researchers can access the Post-Print if their university has
a subscription to the Journal
New impact cycles:
New research builds on existing research
New impact cycles:
Self-archived research impact is greater (and faster) because access
is maximized (and accelerated Post-Print is self-archived in University’s Eprint
Archive
12-18 Months
A 2004 study by the Thomson Corporation, publishers of the ISI citation databases, described a study of the 239 OA journals in the Web of Science (WoS) and the Web of Knowledge (WoK).
These journals constituted 20% of the then universe of 1190 OA titles but only 2.6% of the WoS titles and 1% of the WoK titles. Generally, they ranked in the lower 50% in terms of impact factors, although 14 journals ranked in the top 10% of particular categories. OA journals fared better with regard to immediacy of impact where 20 titles were in the top 10%. This last fi nding confi rms the view that OA journals may infl uence the research cycle by diminishing the time lag between completed research and the results being used and cited (Mc Veigh 2004).
At the article level, advantages appear to be more pronounced. Brody et al. (2004) found that self-archived articles have 50-250% more citations than others otherwise matched with them. In a study of Philosophy, Political Science, Electrical and Electronic Engineering and Mathematics journals, Antelman (2005) found increases in citation rates ranging from 45-91%. A research team from the Université du Québec à Montréal, Southampton University and Universität Oldenburg, in collaboration with ISI, are in the process of studying a 10-year sample of 14 million articles across all disciplines to see whether these fi ndings will hold up in the system as a whole.
Other measurement tools are in the process of development, and these will infl uence the evolving scenario. Online article downloads can easily be measured and there seems to be a correlation between downloads and citation (Harnad & Brody 2004, Perneger 2004) which may make it easier in future to measure the impact of OA. CiteSeer or ParaCite assesses the impact of individual articles. A partnership between ISI and CiteSeer has been set up to create a new citation measurement tool. (http://www.isinet.com/press/2004 /8217120).
It is well known that many publishers and authors exploit the citation system in order to increase the impact of their journals and articles (Monastersky 2005). This practice, also known as ‘gaming the system’ becomes inevitable when impact factors are regarded as the dominant quality measure. There is a rising need for a multidimensional model to measure infl uence and impact (Academies of Science 2004). The Public Library of Science journals have recently come out in favour of content which does not necessarily have the potential for high citations but which will have an impact on public awareness and healthy debate (Parthasarathy 2005)
Open Access has a lot in common with disruptive technologies: it has put in motion processes that may change scholarly communication profoundly. The next meaningful step has already been taken by institutions such as the CCLRC Rutherford Appleton Laboratory who link an OA repository with research data in the GRID + ambient computing environment (Jeffery, 2005).
The slow uptake of Open Access
In light of these advantages, it is amazing that the uptake of Open Access has been so slow.
Bjork (2004) identifi ed six barriers to OA success and plotted them against the three major OA modes: OA journals, institutional repositories and archiving on personal web sites. In the case of OA journals, the major problems seemed to be:
■ business models – where the cost of the journal has shifted from subscription cost to publishing cost and has to be borne by the researcher or his organisation – this seemed to be a particular problem for research organizations with high output;
■ academic reward system – until OA journals are included in the ISI system on the strength of the impact of their articles or until they are deliberately selected by the reward system, researchers will probably choose other journals to publish in;
■ marketing and critical mass – researchers will have to come to know OA journals in their fi elds and trust their content before considering publishing in them.
A recent study by Rowlands and Nichols (2005) revealed, however, that authors’ awareness of OA has increased by 10% over the last year while the percentage of authors publishing in OA journals has risen from 11% to 29%.
In the case of institutional repositories, the problem appears to lie with:
■ indexing services and standards – which explains why good harvesting and value-added services are so important,
■ marketing and critical mass – merely creating the capability (mostly technical) to run an institutional repository will not be enough. Awareness, support, skills, advantages and policies will have to be used to make this the logical conclusion of the scholarly process, and to a lesser extent
Open Access has
a lot in common
with disruptive
technologies: it
has put in motion
processes that may
change scholarly
communication
profoundly
■ copyright – which will be discussed separately.
In the case of repositories, by far the most effective strategy seems to be to mandate the activity.
Institutions who have done so, such as University of Southampton, Max Planck Society, Lund University and University of Amsterdam, have seen dramatic increases in server content without having to waste resources on unnecessary marketing and lobbying. This is supported by the fi ndings of a recent JISC-funded worldwide study by Swan and Brown (2005) where 81% of researchers were positively inclined and would comply willingly with a mandate from their employer or research funder to deposit copies of their articles in an institutional or subject-based repository. A further 14 per cent would comply reluctantly, and only 5 per cent would not comply.
Open Access archives and copyright
Authors often use copyright as an excuse for their reluctance to self-archive, claiming that it will harm their relationship with conventional publishers and jeopardize future publishing opportunities. This stems from a lack of awareness of publishers’ policies although these are readily available on the Romeo and SHERPA web sites (http://www.sherpa.ac.uk, http://
romeo.eprints. org/stats.php). More than 79% of publishers allow postprint archiving and 13% allow preprint archiving without the need to apply for permission. For the remaining 8%
it will be necessary to negotiate with the publishers concerned.
To be able to interpret publishers’ policies correctly authors need to take note of the following:
■ Copyright to a pre-refereeing preprint belongs to the author who may self archive it if the publisher permits.
■ Once the article has been peer-reviewed, copy-edited and accepted for publication (the so- called post print), it may be self-archived only with the permission of the publisher if the author assigned exclusive copyright to a publisher.
■ The online version of the article may not be used as a postprint unless explicitly permitted.
■ Authors should be aware of the policies of publishers and should [at least try to] exercise their right to have self-archiving restrictions removed from their contracts.
■ Authors and repository owners need to be aware that the respective rights of co-authors and their employers need to be clarifi ed (Houghton 2005).
Well aligned with the Open Access movement is Creative Commons (CC) (http://
creativecommons.org/) which is working on a broad front to develop licences that will protect the rights of authors according to their specifi cations without restricting reuse. CC licences are used by the Public Library of Science, BioMed Central and Springer OpenChoice™. Taking this even further is the CC subsidiary, Science Commons (SC), launched in early 2005. SC will expand these activities to the making of tools and data available for reuse, associating research articles and data and standardizing metadata associated with both. (Wilbanks 2005).
The most exciting initiatives with real potential to advance research come from bold experiments where the best options provided by these new technologies are combined within a climate of openness to create knowledge environments.
■ The Social Science Research Network (http://www.ssrn.com/home_hd.html) is collaborating with a variety of role players to provide a real participative environment for nine groups of social scientists.
■ The Signal Transduction Knowledge Environment (STKE) (http://stke.sciencemag.org) provides a community environment with relevant tools and resources for the creation of new knowledge and collaboration. The STKE virtual journal gives full-text access to articles from nineteen different publishers.
■ Faculty of 1000 (http://www.facultyof1000.com) is a next generation literature awareness tool, based on the recommendations of the leading researchers, with the potential of reducing information overload.
Open Access implications for South African research journals
■ OA journals (the so-called OA Gold option), a subset of e-journals, are emerging as a viable alternative to commercial e-journal publishing. Access is free to all and the cost of publication is covered by other means such as author charges. Some of these journals are already included in the ISI databases.
CHAPTER 5: GLOBAL ERESEARCH TRENDS AND THEIR IMPLICATIONS FOR SOUTH AFRICAN RESEARCH PUBLISHING
The most exciting
initiatives with
real potential to
advance research
come from bold
experiments where
the best options
provided by these
new technologies are
combined within a
climate of openness
to create knowledge
environments
■ OA repositories with archived copies of articles that are published in regular journals (the so-called OA Green option), whether subject specifi c, such as arXiv, or institutional, provide additional access and visibility. This has advantages for readers who do not have subscription access or who cannot pay the pay-per-view fee, as well as for authors who achieve greatly enhanced visibility.
■ OA repositories have to be created, branded, managed, marketed and maintained. The costs to archive are minimal, but ongoing curation costs will accumulate. The barriers for changing the commercially oriented communication system have been greatly underestimated and need to be managed.
■ OA is signifi cantly devalued where harvesting of the published material is sub-standard: this very important activity should be planned for if impact is to be maximized.
■ The barriers to changing the commercially oriented communication system were greatly underestimated and this needs to be managed.
■ Vested interests are emerging – the ‘commercial’ publishers have a business with a profi t motive, while governments/other funders are resisting ongoing copyright “raiding” by publishers.
■ Parallel print with electronic (delayed or immediate) OA has been demonstrated as no real threat to conventional publishing from the viewpoint of declining subscriptions, and has shown evidence of increased citation.