A few years ago, Southampton University's Librarian, Bernard Naylor, sent round an email to his University Librarian colleagues, asking by which year each one thought he or she would be subscribing to just 20% of their periodicals as print rather than electronic journals. The replies duly rolled in, revealing that the consensus within this particular subset of the UK library profession was that 80% of journal subscriptions would be in electronic format by somewhere between 2005 and 2010. Bernard Naylor has been active again this year, leading a NetLinkS Discussion Area entitled 'Your chance to re-invent the scholarly journal' in which he asked thirteen questions on the future of the journal, based on the assumption that the electronic version was well on its way, and that this could change the whole basis of research literature provision. Few people took up this invitation. Inevitably, there were no unequivocal answers. It is abundantly clear that there is no magic pathway to the sunlit uplands of a future of quick and easy access to academic information.
Earlier this year, Ariadne ran a debate between Fytton Rowland of Loughborough University, and Steven Harnad, indefatigable advocate of free access, at the point of use, to the world's 'esoteric' research output. Although the former was supposed to be defending the value of print against the latter's championing of electronic access, it was interesting to note the level of agreement between the two. Both agreed on the functions of the journal literature and on the vital necessity of preserving peer review and some sort of quality hierarchy, whatever the output medium. They accepted that electronic production and output are here to stay. The main arguments centred more on future developments and on how the large commercial publishers could be circumvented. Although he no longer predicts any dates for their demise, Harnad regards the existence of expensive learned journals as fundamentally untenable in an electronic world where authors and readers are the same people with access to the same electronic resource. Rowland is not so sure that their extinction in all cases is so certain.
The previous paragraphs are an indication of the crucial changes taking place right now, of the difficulty of predicting the exact outcome of the changes, and the course and duration of the transition period en route to any new steady state. My crystal ball is as cloudy as anyone else's in the information business, and I do not wish to offer any predictions additional to the above, but what I can do is give an account of the present position of electronic journals (ejournals) in the UK higher education sector.
It is apparent that the Higher Education Funding Councils' Pilot Site Licence Initiative (PSLI) has been a real catalyst in encouraging the use and marketing of ejournals within UK universities. This is in spite of the fact that the PSLI was not initially seen as primarily a vehicle for ejournal promotion, but rather an experiment in trying to break the vicious circle of excessive journals inflation.
This initial emphasis may reflect its origins outside the orbit of JISC and eLib. This in turn may have limited some of the discussion on the future of, and any successors to, PSLI. The ejournals funded under eLib - Sociological Research Online, Journal of Information Law and Technology, Internet Archaeology, etc - have been very useful examinations of the process of setting up, running, and financing an ejournal. However, it is probably true that they, the rather different SuperJournal project and the PSLI have to an extent missed out on some promising cross-fertilisation by not coming under the same immediate umbrella. The same could be said of the future development of ejournals in general.
Pre-PSLI, a few libraries had been experimenting for a year or two in the mid 1990s by providing access to some of the free-of-charge ejournals. The pioneers started publishing in 1990, well before the World Wide Web, HTML, Netscape and Acrobat utterly transformed the ejournal scene. There was even very tentative adoption of one or two of the ejournals released by the pathfinder commercial publishers like Chapman & Hall in the UK and Johns Hopkins University Press in the US. BUBL was also doing its bit by providing access to some full text ejournals. However, the lack of a reasonable base of well-known journals meant that take-up, beyond a small band of enthusiasts, was low.
Until 1998 the PSLI provides the UK higher education community with full text access to all 170-180 journals published by Academic Press, and to all 35-40 Institute of Physics Publishing (IOPP) journals, providing the library subscribes to the print version of at least one title. Via BIDS it offers access to the full text of Blackwell and Blackwell Science titles where the library maintains a print subscription, with contents/abstracts information on the rest. Academics therefore suddenly acquired the ability to read, in the privacy of their own labs or offices, the current issues of around 250 well-respected peer-reviewed journals.
That is the good news, but that previous sentence requires some qualification. First, I use the word 'suddenly' loosely. Titles first became available in the spring of 1996. It took around six months after that for most Academic Press titles to appear online. IOPP titles did appear quickly, but BIDS JournalsOnline with the first Blackwell Science journals was launched in November/December 1996, while some Blackwell titles are still (autumn 1997) reaching BIDS for the first time. Many issues are still not appearing on the Web at the same time as the printed version is distributed, let alone a few weeks in advance which in theory should be possible, with timely publication often put forward as an advantage of electronic over print.
Second, the question of ease of access needs scrutinising. Even now, not all academics and researchers have everyday individual control of their own networked PC. The use of Adobe Acrobat PDF viewing software by all PSLI publishers, and many others, has also not been without its difficulties. Although Acrobat software is freely downloadable, action still has to be taken to achieve this, either by individuals or by those controlling particular servers or networks. This can take time to arrange or negotiate, and acts as a barrier to some. The recent upgrade to Acrobat 3.0, rendering some recent issues unreadable by previous software versions, means the whole process has to be repeated. Printing from Acrobat has been another controversial area. Fairly high specification laser printers, or printers configured in particular ways, are required to print Acrobat documents reasonably speedily, with some libraries initially reporting printing times of several hours for a single article. Other viewing software has been adopted by different publishers, for example Catchword's RealPage by Carfax, the Royal Society of Chemistry, Taylor & Francis, and others. This software may well have some advantages over Acrobat, but requires separate downloading and registration, creating another potential barrier.
Another aspect of ease of access is the use of usernames and passwords. Initially all three PSLI publishers required the use of passwords. Some institutions have announced these on restricted Web pages, but the majority have issued passwords individually as requested. This is a further restrictrion on widespread adoption of the new resource. Academic Press and IOPP have now moved to control through IP registration, while retaining passwords as an alternative in certain circumstances. This provides academics and students with transparent access, and is being adopted by more publishers, certainly simplifying both administration and use. An alternative may be the use of the new ATHENS3 authentification procedures. JournalsOnline is a slightly different case, in that researchers can use their existing BIDS ISI passwords, which will also come under ATHENS3, to enter the service.
There are other reservations. Even with 250 journals available, there are likely to be no more than one or two of interest to any individual researcher. Slow response times and server downtime cause frustrations here, as much as in other parts of the Internet. Yet it is still my belief that the PSLI has acted as an undeniable incentive to libraries to think out ways of making the ejournals available and publicising their existence, which they are then applying to other ejournals as appropriate. Many academics have vaguely heard of PSLI, and are gently pressing library staff to provide more information.
Libraries have advertised the PSLI, and ejournals in general, in various ways. Documentation has been written, and sent out to individuals, departments or committees; the library's website has been used; articles have been placed in university and faculty newsletters; demonstration sessions have been held for individuals and departments; 'electronic journal parties' have been held, where any pain inherent in learning about something new has been alleviated by a glass or two of wine; the publishers themselves have written to academics. One of our biggest surges in demand at Glasgow followed Academic Press's letter to individual staff. Despite all this, demand is still not overwhelmingly high. At Glasgow, one of the heaviest users in the UK of Academic Press's service, fewer than 400 articles have been downloaded from about 175 journals in the busiest month. However, use has to start somewhere, and will I believe continue to increase as a standard service develops, and more journals are issued in electronic format.
One service that has varied considerably from library to library in the UK has been the dissemination of information on which PSLI journals are available electronically. Over the summer, I have checked out the OPACs and websites of the majority of UK libraries. Only a small minority have entered PSLI journal details on their OPAC. Where a Web OPAC exists, the number of libraries that have made use of the MARC 856 field to provide direct access to a journal title, or even a publisher's home page, is probably in single figures. Very few OPACs even provide a note indicating that the journal is accessible via the Web. A similar number of libraries provide direct access from their OPACs to a few other ejournals: for example, only four out of the nearly seventy OPACs checked provide a link to the electronic version of Ariadne, although rather more than half list the print version. Of course, the majority of OPACs are still telnet rather than Web, but only a tiny number mention the electronic version, even if direct access is impossible.
Most library websites do recognise the existence of ejournals, including PSLI, although it is often necessary to move through various levels to discover entry points. More than half do not provide any listing of individual titles, however, only linking for example to IOPP's home page. Most of the rest that do list titles then connect again to the publisher's home page. This was very understandable when individual Academic title URLs, for example, changed with each reload, but that is no longer the case, and it is now possible to save the user time and effort by pointing directly to a title-level URL. We surely do not expect academics and students to search for journals by publisher, and although time is always at a premium, it is probably appropriate to invest a little more effort in informing users, through the library's OPAC and/or Web pages, which particular journals can now be inspected online.
An alternative popular with some is to concentrate on subject pages and/or to make use of general services like BUBL. These are certainly essential, and a good route to access sources like the excellent eLib broad subject resource discovery services such as EEVL, OMNI and SOSIG, but I suggest that traditional access via general catalogues and listings is also still important, and likely to be a part of various 'clumps' developments.
What of the future? There are so many developments at present. The serials agents are launching or developing their ejournal administration and control services such as Blackwell's Electronic Journals Navigator and Swets SwetsNet; OCLC is providing a similar service through Electronic Collections Online, emphasising its archiving functions; the US JSTOR project is tackling the problem of providing electronic runs of back volumes of prestigious journals; the eLib Open Journals project is trying to enhance the linking capabilities of HTML, and Acrobat, by constructing 'linkbases' that enrich many times over the content of an individual article or document; the initial evaluation of the PSLI has been published, and there is much interest in what if anything will replace it when the agreement expires at the end of 1998. Finally, almost all large STM publishers are now producing ejournals, all on offer under different charging mechanisms.
Each of these topics, and there are others, could be discussed at some length. The moral would appear to be that, whatever the future holds, and whatever paths we shall follow to reach that future, there is no shortage of activity, and no danger that anyone working in this area is likely to suffer from boredom.
 Higher Education Funding Councils' Pilot Site Licence Initiative (PSLI)
 Chapman & Hall
 BIDS JournalsOnline
 Acrobat software
 Initial evaluation of the PSLI