The views expressed in this paper are independent of the managing committees of the two major schemes described.
This paper has been downloaded
The Pilot Site Licence Initiative (PSLI) and the Electronic Libraries (eLib) research programme have been catalysts for dramatic change in journals publishing in the UK. Covering over 30 publishers and over 20 other e-journals and e-journal research projects, this report marks the extent of the change and reflects on how e-journals will develop next.
publishing, electronic journals, electronic libraries
E-journal developments in context
What is a 'UK' journals perspective?
Catalysts for change in the UK
PSLI: its wider impact
Beyond the PSLI
A note on journal formats
A note on pricing
Some other UK e-journals
Electronic journals: the eLib experience
eLib in perspective
Authors' note: why 'circa'?
List of tables
Table 1. Evolution of a species: the growth of peer-reviewed e-journals 1994-97
Table 2. Online journals produced by HEFC pilot site licence publishers
Table 3. Other UK-connected journal publishers on the Web
Table 4. E-journals on BioMedNet
Table 5. E-journals produced and distributed by Catchword
Table 6. E-journals produced and distributed by Bioline
Table 7. Formats used for e-journals
Table 8. Independent UK e-journals
Table 9. Original eLib e-journals
Table 10. Parallel e-journal projects within eLib
Table 11. Integrating services projects within eLib
Table 12. E-print projects within eLib
Table 13. Some random beliefs about e-journals exposed
Table 14. 'UK' e-journal totals
In the commercial world of electronic journals publishing the unicorn remains a shy and elusive creature; meanwhile the rhinoceros charges on. The terms are Schauder's (1994), the unicorn representing the goal of the fully paperless publishing cycle 'from writing to end-use'. The rhinoceros, in contrast, 'is as much identifiable by the swathe of organisational change which its rampaging progress leaves behind in the scholarly publishing industry, as by the technological characteristics of its horn'.
What this creature now leaves in its wake, as this paper shows, is a massive switch by publishers to embrace the idea of delivery of journals in electronic form. It is not a switch to paperless publishing because almost universally these e-journals from established publishers are based on page-by-page replicas of the paper originals. It might not reach the stage where every academic journal currently available in paper form will have an electronic analogue, but the vast majority will, and the switch in publishing resources to make this happen will have largely been accomplished in three years from 1996.
Given the speed with which change is happening, this paper
will present the bare facts of this change, a simple record of a short
period which may or may not, with greater analysis and hindsight, prove
to have been an important pivotal moment. As with any survey of a moving
target it must take a snapshot in place and time and this one, to reflect
this forum, considers 'UK' journals and views the horizon as it appeared
in October 1997.
E-journal developments in context
In one respect the speed of change today belies the fact that the development of early e-journals, from the viewpoint of some of the first experimenters, struggled to keep pace with technology (McKnight 1993). Yet it was approximately 200 years after the invention of print that the first scientific journals appeared in 1665 in London (Philosophical Transactions of the Royal Society of London) and Paris (Le Journal des Scavants) (Schauder 1994); it is just 20 years since the feasibility of an electronic journal was discussed by Senders (1977).
From a publishing perspective the importance of audience reach cannot be underestimated, and it was only with the emergence of the Internet as an international public utility that sustainable, networked e-journals became viable. It took the appearance of the World Wide Web as an easy-to-use Internet service to popularise this form of computer-based networked communication for a mass audience.
For e-journal developments, what might be called the 'Internet phase' lasted from 1990 to 1995. While this period may have produced relatively few journals in the overall scheme of academic publishing, it was a time of pioneering journals and significant diversity not just in terms of their appearance and content, but in the nature of the groups and organisations that published them (Hitchcock et al. 1996). By the end of that stage it was becoming clear that mainstream publishers were no longer going to remain experimenters or bystanders: 'A number of UK-based academic publishers have announced their intention to make all of their journals available online during 1996', we reported.
The Internet remains an important component of the infrastructure that supports e-journals, of course, but more specifically all e-journals are now effectively accessed via that single Internet service, the Web. That is not to say that the Web prescribes a single 'look and feel' for all content, but if familiar, refereed e-journals are all you look at, then that is how it might seem.
The popularity of the Web is founded on the freely available and remarkably effective browser interfaces that are becoming ubiquitous, certainly within new computer systems. Another important factor is the ready acceptance and stabilising influence of the hypertext markup language (HTML), the language of the Web that determines the appearance of documents as viewed in these browsers. It may seem strange, in the changing field of Web presentation formats - which now embraces 'dynamic' HTML, the extensible markup language (XML), which is beginning to define forms for metadata, as well as Java and various scripting languages used to manipulate Web content and give the appearance of movement on otherwise static pages - to refer to HTML as 'stabilising'. It is, though, undoubtedly a factor moderating the interests of commercial systems developers competing for the Web market, and it sustains the influence of the body that sets the standards for the Web, the World Wide Web Consortium (W3C).
It is this appearance of stability perhaps - certainly it is not due to HTML itself - together with the apparently growing demand for electronically-delivered information, that eventually lured journal publishers into putting real products on the Web. There had been major experiments during the Internet years, notably:
So while the Web grew dramatically in popularity from 1994 following the introduction of the Mosaic browser and then the appearance of Netscape, it was another two years or so before we entered what might now be called the 'Web journals' phase, when journals of all sorts were transferred to and reproduced in this new medium. How this migration happened during this period can be roughly charted from a number of surveys of e-journals, as shown in Table 1. In the aftermath of this transition we can now see Web journals publishing taking shape.
Table 1. Evolution of a species*: the growth of peer-reviewed e-journals 1994-97
|Authors||Data published||No. of
|Clement||Oct. 1994||25 (4 on Web)||Science, universal access, archived|
|Roes||Dec. 1994||39 (14)||Full-text|
|Hitchcock et al.||Jan. 1996||115 (115)||STM,
|Harter and Kim||May 1996||77||Scholarly|
|Hitchcock et al.
|Oct. 1997||c. 1300
(c. 3200 projected 1998/9)
What is a 'UK' journals perspective?
Academic journals are invariably international is scope with editorial boards and contributing authors drawn from all parts of the world, so it is probably worth a short note to explain what is meant when this paper refers to 'UK' journals. Despite the growth of networked publishing described here, the production of these journals indisputably remains rooted in the methods of paper-based publishing. That is, most journals have editorial offices that coordinate production from a particular geographical site. Publishers considered in this paper have editorial offices and a presence in the UK.
It is also true to say that many of these publishers are part of multinational corporations, so even though certain journals may be produced in the UK the whole journals programme of such a publisher is likely to be produced in more than one country. This paper does not differentiate journals produced in one country or another by multinationals. If a journal publisher fits the 'UK' criteria outlined, all of its journal titles are included in the figures.
None of this is vitally important because the broader picture is ultimately of most interest. The UK is not being singled out as an area of special merit in this field of publishing, but given the long history of journal publishing in this country and the role of English as the international language of science, it is probably not unrepresentative of wider developments. So although the totals given here do not represent the total world output of e-journals, given the presence of data from multinationals the figures may not be far short of current world totals, certainly not as far short as a stricter interpretation of UK publishing might suggest.
The justification for this summary approach is that if we want to monitor the development of e-journals, the survey method, as Table 1 suggests, is almost too cumbersome already. An objective survey should require that the journals covered are each viewed, but the Web is no faster than it was when we compiled our original survey in 1995, the number of journals is hugely greater and access to most is strictly controlled, a complete contrast to the conditions we encountered then. Having reached the number of e-journals reported here it is of little remaining interest to keep counting but more important to be able to see the broader picture and analyse the changing nature of the e-journals themselves.
One other point about the data collection. The figures given here are provided either directly by publishers or from data that can be gleaned from the Web. Any inaccuracies - or new and updated information - can be notified to the authors who will attempt to periodically update this Web version to complement this version of record. Where possible links are provided to sites from which data has been taken, so readers can obtain a more up-to-date picture by following selected links.
One of the difficulties of emerging fields is adopting phrases that have a consistent meaning for all readers over a significant period of time. In this field the prefix terms electronic- , networked- , online-, Internet- and Web- which describe the forms for disseminating journals are often used interchangeably. In fact, broadly they describe decreasing subsets of these forms. Thus, 'e-journal' has the widest interpretation and is used to describe disc and network distribution, and while the term 'Web-journal' is the most specific - almost invariably non-paper journals today are Web journals - it is the most recent and least stable of these terms and so is little used. This report uses the most general term e-journal, but through it we refer strictly to networked journals, not discs. Where used this term can be taken to subsume the terms below it in the above hierarchy.
Catalysts for change in the UK
Given the qualifications above, the UK has provided two distinct catalysts for change
Also funded by the HEFCs and managed by the Joint Information
Systems Committee (JISC), the eLib programme aims to apply some cutting-edge
technology while also being concerned with organisational and infrastructure
issues to support the concept of a digital library in its widest sense,
not just in the delivery of e-journals. The eLib e-journal projects described
below have in some cases attracted support from large numbers of publishers,
perhaps hoping to enhance the e-journal delivery programmes stimulated
by the PSLI.
PSLI: its wider impact
The intent of the PSLI scheme was not to blaze a trail
of new technology; rather, the real innovation was the speed with which
the technology was deployed within the companies across these large projects.
The e-journal outputs as a result of the PSLI are shown in Table 2.
Table 2. Online
journals produced by HEFC pilot site licence publishers
|Publisher||No. of online journals
of all titles
|Blackwell Publishers||134* (163 in 1998)||70% (85%)|
|Institute of Physics||33||100%|
Nor was technology the principal motivator as almost certainly the projects would not have happened without the financial underpinning of the PSLI scheme. Instead, remembering the omissions, as publishers might see it, of those Internet e-journal experiments, here is a business model compatible with existing journal subscription practices, in effect extending subscriptions from single titles across whole journal programmes, although with important differences to account for the different nature of e-journals.
So this is the real significance of the pilot scheme because, although restricted to subscribing organisations in the UK, this site licence model became the forerunner of site licences elsewhere, such as Academic Press' $1.1 million Ohiolink deal. In principle, a site licence buys a subscribing library access to a specified 'bundle' of a publisher's journals, usually a large number of journals, which can be made available at specified sites or terminals. In the Ohio case the arrangement covers all Academic Press journals at all college libraries within the state.
The advantage for the libraries is that they get more publications than they could afford in print, and there are fewer problems of access or copyright. External document delivery fees can be reduced. (PSLI Evaluation Team 1997) The advantage to users is that wherever they are in the state they can dial into the central library system and get full-text articles covered by these agreements. A researcher in a local college can have access just as if he or she were at a main university library.
According to Arly Allen, President of journal production company Sheridan Electronic Systems in the USA, this is a ‘political coup’ from the state government perspective. 'Constituents in rural regions who historically oppose spending on higher education because it draws money from their pockets into those at the universities miles away, can begin to see some benefit. Local junior colleges can have access to library resources and classes via long-distance learning centres. Educational opportunities are more evenly spread throughout the state. It is a system for which the state politicians will want to take credit.'
What has been described by Pieter Bolman of Academic Press as a 'win-win' deal for publishers, libraries and users has its detractors, however. Ken Rouse, head of the chemistry library at the University of Wisconsin-Madison was referring to a site licence contract from Elsevier when identifying the following drawbacks. Seen from the perspective of a librarian, Rouse (1997) said:
Apart from widening access to e-journals, the PSLI scheme has further shown how the user's experience of e-journals can be transformed, in particular by the range of online services that the Institute of Physics (IoP) offers to complement its full-text delivery. Among these facilities are free information services, including access to an online compilation of the full text of research 'letters' submitted to IoP journals, advance notice of abstracts of forthcoming papers and a product-finder service. There are personal services, such as a directory for contacting peers by email, and a jobs exchange. The whole service can be customised by users, who can also access text and services remotely when away from their offices. This service was recognised in the inaugural Charlesworth Group Award for e-journals in 1996 (Charlesworth is a UK journal producer and printer).
This is a strong offering even at this early stage, and it demonstrates the way that information providers on the Web must go: simply, for users there must be a compelling entry point, or focal point, to an environment that builds community of shared interests. While in principle this may seem no different from established publishing practice, on the Web it means giving away for free services which might otherwise have been considered for-pay (had these services been considered at all, that is). There is still significant progress to be made, because the for-free services will begin to encompass some full-text papers as well, as IoP's access to 'letters' shows, and this is where the alarm bells start to ring for other publishers.
The IoP's bold initiative betrays a lack of confidence in one respect, that its remote access facility will not allow the user to see full-text papers, and this user is a paid-up subscriber. As the service progresses it will become more likely that it, rather than the underlying papers, acquires the important publishing added value, a trend which can only accelerate as journals have to cede exclusivity on individual papers and their contained data as multiple forms of any paper, in multiple media, prevail.
In other words, what users will pay for in paper form does not translate directly to the Web, but through schemes such as the PSLI, publishers and libraries are beginning to discover what does.
There is another way to look at this issue: pricing. Fundamentally what is at issue here is access to information, in particular the need to reduce barriers to that information. Pricing is the most important barrier of all. High pricing reduces circulation and access, exactly the process that is diminishing paper journals publishing, the so-called 'serials crisis' because library budgets cannot keep up with journal price increases year-on-year and subscriptions numbers have eroded. On the Web information providers want positively to encourage users to their information rather than appearing to deny it as journals pricing does at present. So those services that command user support will be those that provide the best-tailored services, the optimum means of discovering the information the user needs and, in the era of non-exclusivity, perhaps the best version of that information as well.
The serials crisis at its worst leads to journal cancellations, at best it is a problem of perception for publishers. The model of e-journal publishing must address this perception and tackle the issues of both pricing and access. One of the strengths of the PSLI scheme is that it attempts to do this.
A successor scheme to the PSLI is under review following evaluation of the first phase (HEFCE 1997).
Beyond the PSLI
In the wake of the PSLI, over 18 months on from those first services from Academic Press and IoP we can see many other publishers on the brink of joining them by introducing online access to current issues of their complete journals catalogues. This section briefly shows what progress has been made by these publishers, or what progress is intended. Signs are that there may have been slippage in the original schedules for some journal programmes, but the overall plan to provide access to online versions of all major journals is clear, and it will happen in 1998/9 if not before (Table 3).
Table 3. Other UK-connected journal publishers on the Web
|Springer-Verlag||c.220||400 e-journals in Link service by 1999|
|International Thomson||>140||reducing to c. 110 due to planned sale of some titles|
|MCB University Press||98 (100%)|
|Gordon & Breach||49 (c.15%)||13 more titles imminent|
|Cambridge University Press||30 (20%)||intends to make 60 more titles available in 1997/8|
|OUP||10 (6%)||23 more due within months|
|Priory Lodge||9 (45%)|
|**Elsevier||8 (1%)||c.1100 journals will be accessible via ScienceDirect; early release phase to selected institutions began June 1997|
|Wiley Interscience||5||50 journals to be piloted online in October 1997: c. 350 more to be phased in during 1998|
|**Kluwer||116 journals to selected libraries from July 1997; c.320 to be available through OCLC|
The advantage, now apparent, conferred by the PSLI was not available to some publishers and not grasped by others. Smaller publishers with fewer than 35 journals were denied participation. One of the weaknesses of the PSLI is that it fails to address the role of small and emerging publishers of Web journals, with the potential innovations that could have introduced. For larger publishers there was the issue of international markets. By being restricted to the UK the scheme was seen as being of little relevance by some of those publishers for which the UK represents less than 10 percent of their market.
Yet new, service-oriented online publishers and producers have emerged in the UK, most notably BioMedNet Ltd, Catchword and Bioline. These are not identical services; what they have in common is that they have brought publishers and journals to the Web faster than may have happened otherwise. Where BioMedNet is part of the Current Science Group, an established publisher, and Bioline is a publisher and publishing service, Catchword is founded on production and subscription services rather than being a publisher itself.
BioMedNet and Bioline are online-only publishers, a distinction which in world terms they share with Highwire Press in California. BioMedNet, which is described as an online club for those working in biology and medicine, does not produce any paper journals, and while this puts the company high on the innovation/risk curve, it also provides the flexibility to respond more quickly to new user demands for access to information. The philosophy and practice of BioMedNet was discussed candidly by Hitchcock et al. (1998). The publishers for which BioMedNet produces online journals are shown in Table 4.
Catchword aims to provide a 'complete electronic scholarly publishing environment on the Internet'. It hosts publications on a global network of servers and provides support for publishers to manage subscriptions and collect money. Catchword produces over 130 e-journals for publishers, as Table 5 shows. Unique to Catchword is RealPage, a proprietary portable document format (PDF).
Bioline Publications is another electronic publishing service for bioscientists. In collaboration with Base de Dados Tropical, a bioinformatics database service in Brazil, it works with publishers to make established journals and new online-only journals, as well as other sources of scientific information, available 'more easily and more cheaply' to end users (Table 6).
Table 4. E-journals on BioMedNet (Note. Registration required to follow this link)
|Y||Company of Biologists||4||4|
|Others||6||1||4 (1 HTML)|
Table 5. Online
journals produced and distributed by Catchword
|Taylor & Francis||37|
|The Psychology Press (T&F)||15|
|Royal Society of Chemistry||13|
|Institution of Chemical Engineers||3|
|Royal Society of Medicine Press||2|
|Others (inc. Plenum Corp.)||31|
Table 6. Online
journals produced and distributed by Bioline
|Taylor & Francis||1|
(11 international publishers)
A note on journal formats
Discussion of these e-journal producers is an appropriate point at which to take a broad overview of the formats being applied to the production of e-journals. Given that HTML is the formatting language of the Web and that Postscript is the de facto standard for printing copy from computer-generated pages, it is inevitable that formats based on these standards will dominate e-journal production.
Adobe developed its Acrobat PDF application, which can be produced from Postscript, to improve the transmission and distribution of documents over networks and via CD-ROMs. A significant development were the file compression options to reduce file sizes. Large files seriously limit page image formats, as the Tulip project discovered. Although there are other PDF applications, Acrobat has come to dominate in the production of online journals, as Table 7 shows.
Catchword's RealPage also converts from Postscript delivering, it is claimed, 'a much smaller file size much more quickly'. Unlike the widely-marketed Acrobat, RealPage is targetted directly at journal publishers.
PDF and HTML support distribution to all popular computer platforms through freely available viewers, and both support hypertext linking and text search, although the ease and cost with which these facilities can be offered varies significantly.
It is possible that HTML generated from SGML 'on the fly' will become the major e-journal format in terms of journal numbers based on the announced intentions of Elsevier. This format is already used by BioMedNet. SGML is the preferred format for long-term storage of information in electronic form and offers the most power and flexibility in terms of output formats, in contrast to HTML which is essentially a communication format. The problem with SGML is that it is complex, and maintaining SGML publications is labour-intensive (Wusteman 1997). Ironically it is the simplicity of HTML which is motivating renewed interest in SGML.
Before the Web and browsers prevailed SGML presentations required a formatter or front-end translator in order to be viewed. The complexity of SGML derives from the fact that, despite the inference of its full title, Standardised Generalised Markup Language, it is not a markup language: it prescribes how markup should be specified, not what that markup is (Barron 1989). Thus an SGML document can be presented to a viewer in different forms, the actual appearance of the document being determined by the document type definition (DTD) specified for the presenting application. HTML for example, which is an instance of an SGML DTD, prescribes a set of tags, defined in the recommended HTML standard of the moment, which are recognised by a Web browser and rendered on the screen.
Intermediate between the extremes of SGML and HTML is a new markup language not yet seen in journal applications, the Extensible Markup Language (XML). Described as a simplified subset of SGML, and claimed to be easier to use, XML will allow content creators to define new tags. This is a work in progress as far as standardisation and industry adoption are concerned, but its prospective importance for journal publishers is that it may be supported by more browsers than SGML; compared with HTML it might handle technical notations more effectively, and it will offer greater support for validation of document coding or formatting. Above these, XML is being promoted for more sophisticated Web publishing applications, such as integrating database services with document delivery functions, providing more sophistication for features such as hypertext linking. Not forgetting that XML is the fulcrum around which standards for document metadata, another important goal for publishers, are being defined.
Against this background HTML continues to evolve and will no doubt subsume some features planned for XML. Cascading style sheets, another W3C standard, provide DTP-like functions and can be used to modify and control the appearance of HTML documents by specifying presentation features for a document or for a range of documents (Lie and Bos 1997). HTML will continue to dominate Web-based presentation for the foreseeable future
Whether a journal adopts HTML or PDF typically depends on the origins of that journal and existing arrangements for its production, with HTML used for electronic-only journals where there are no problems with settings for mathematical or scientific notations, and PDF used for journals originated for publication on paper. Increasingly papers are becoming available in both formats to support screen reading or local printing, although the use of either format for journals has been questioned (McKnight 1997). Proprietary formats such as Postscript and Printerleaf are retained in cases of complex or special settings and in fields - maths and physics are obvious examples - which have had a longer affinity with online information services and pre-date more recent formats.
Table 7. Formats used for e-journals
|Format||No. publishers using
|No. journals using
|19||c. 1050 (c. 800)|
|RealPage||14 all via Catchword||131|
(HTML ‘on fly’)
|5 via BioMedNet
|Other (Postscript, Printerleaf)||2|
A note on pricing
Needless to say, it is not maturing formats for e-journals that have motivated publishers to release e-journals on the scale described here. True, the ability of PDF to deliver high-quality facsimiles of journal pages is a factor because it incurs little change in the appearance, hence in the identity, of a published product. More important, though, is the availability of systems, such as Commands from ICL/Fujitsu (used by Academic Press) and Zuno's Digital Publisher (to be used by Wiley) to handle subscription transactions online together with the apparent market acceptance of adapted subscription models, such as site licences, as demonstrated by the PSLI.
Thus, pricing schemes for e-journals fall into three broad categories, of site licence subscriptions, per-journal subscriptions and per-article fees. Typical pricing examples for each category are:
Some other UK e-journals
While many scholarly journals are little known outside their specialist fields, a few prestigious research journals, typically in biomedicine, transcend this market segmentation to become familiar names. Of these journals the pioneer of online texts is perhaps Science, although from a UK perspective the best known titles - The British Medical Journal (online issues from March 1995), The Lancet (from June 1996), and Nature (full online service imminent) - offer more limited services with some free 'magazine' content including news, letters and selected retrospectives. Full research papers can be hard to find, and where they are available they may be from another service - The Lancet papers from Ovid, for example.
In contrast to the lead, in world terms, taken by established
UK journal publishers in releasing e-journals, this country has not proved
so prolific in terms of producing new and independent (of established publishers)
e-journals. It has been anticipated that the switch to electronic publishing
would stimulate new producers, and during the 'Internet journal' years
it was the case that a relatively high proportion of e-journals were independent
(Hitchcock et al. 1996). Revealingly,
with one exception those e-journals originated in the UK and listed in
Table 8 have so far barely attracted enough papers between them to fill
a single issue of a conventional journal. The one journal in Table 8 to
have established a substantial literature, Psycoloquy, originated
not in the UK but in the USA, although its editorial office is now based
at Southampton University.
Table 8. Independent
|Journal||Launch date||No. original papers|
|Psycoloquy||1990-||1990: one vol., 16 issues
1991-: seven vols, 338+ original articles and commentaries
|J. Corrosion Sci. and Engng||1995-||7|
|Int. J Small Satellite Engng||1995-||5|
|Brit. Soc. Gynaecological Endoscopy Internet J.||1995-||4|
|J. Interactive Media in Education Charlesworth commended||1996-||4|
|Internet Journal of Science - Biological Chemistry||1997-||8 (hosting an e-conference Nov./Dec. 1997)|
|Geometry and Topology||1997-||3|
|Journal of Digital Information||1997-||2|
Psycoloquy was one of the first e-journals to appear on an Internet service. Its innovation was to pioneer open peer review in an online environment - the journal's editor Stevan Harnad refers to this as 'scholarly skywriting' (Harnad 1990) - based on methods proven in the print medium. Although Harnad has long argued that freely networked e-journals, with the potential for virtually instantaneous dissemination, are the natural medium for open peer review, his high ambitions have by admission yet to be fully realised.
Indicating that innovation is a vital feature if a new
journal is to make an impact, another independent UK e-journal of note
is the Journal of Interactive Media in Education (JIME), commended
in the Charlesworth awards for 1996, this recognition being in respect
of the journal's novel presentation rather than its content. Developed
at the Open University, JIME exhibits a frames-based structure for
managing content and presentation, with distinctive results.
Electronic journals: the eLib experience
The Electronic Libraries programme (eLib), which began in 1995 with a budget of 15 million pounds, is a research programme with a difference which ought to be noticeable among journal publishers and those concerned with journals. A primary objective of eLib has been to base its projects within the broad community it serves, through collaboration within and between projects, but more importantly by promoting continuing evaluation of projects and its emphasis on projects developing so-called 'exit strategies' and preparing for when funding ends. In other words, projects have been exposed to influences beyond academic research, have to report publicly, and have had to focus on ideas and applications that have the potential to become relevant and applicable beyond the research environment.
The Tavistock Institute has been instrumental in setting up, coordinating and developing the evaluation framework. Its first report (Tavistock Institute 1997) synthesising the results of projects to August 1996 reveals the successes and difficulties of imposing this approach: 'We get a sense from the reports that whilst evaluation is one of the activities posing additional demands on projects, there is at the same time a new appreciation of its centrality to the innovation process'. A second report due in November 1997 will give a better picture of progress in the evaluation and exploitation process.
In this way eLib is not overtly a research programme: 'its main remit is to provide a body of tangible, electronic resources and services for UK higher education, and to affect a cultural shift towards the acceptance and use of these resources and services in place of more traditional information storage and access methods'.
The programme supports around 60 projects under 11 themes
including, quite conveniently for this purpose, 'electronic journals',
but since it is a forward-looking programme we should also be concerned
with project areas which might influence future e-journals in some way,
in particular e-print services. Within the eLib e-journal strand itself
projects can be categorised in three ways: original e-journals, 'parallel'
e-journals (i.e. those that are based on existing publications), and integrating
services which are concerned less with content generation but more with
repackaging, enhancing content and improved means of delivery.
Launching a new journal, any journal, whether paper or electronic, is not easy. In this respect it is fair to say that the original eLib journals have performed markedly better, as can be seen in Table 9, than most of the UK independents described above with which they share a lack of established publishing support. Innovation in these journals is largely confined to presentation as they explore the potential of the electronic medium, and the publishing models adopted are based firmly on commercial paper journals publishing, as too seem to be the preferred exit strategies.
Table 9. Original eLib e-journals
|Journal||No. issues||No. original papers||Features|
|Internet Archaeology||3||12||Supports longer papers, e.g. book
Maps, colour photos
|J. Information, Law and Techn. Charlesworth winner||5||47||Innovative interface
|Sociological Research Online||7||45||Full-text refereed papers
Search, announcements, training and directory services
Judging solely from the range of contributions, these eLib journals appear to be developing new communities. Each one has demonstrated features that are not always feasible on paper: Internet Archaeology, for example, has published book-length theses, colour maps and photographs. In contrast Journal of Information Law and Technology presents research at magazine-style speed with a mazy range of features but brilliantly organised so that even the novice e-journal reader should not get lost, for which it was the overall Charlesworth winner last year. Sociological Research Online is more conventional and demonstrates the curve of a maturing journal, not untypical of a print journal, growing in visibility and authority.
The question is whether the journals can translate community
into viable markets. The danger for all in this scenario is that, despite
the content, the publishing model will not support this transition, not
without compromising content, and that while the market wills this new
form of publication, it may not yet be prepared to embrace a radical means
of funding it.
Progress of parallel journals can hardly be measured by volume of output as above. Instead, these projects might be viewed more as trialling new technologies, especially formats, in a real publishing environment (Table 10). Thus the CLIC project, which is working with the Royal Society of Chemistry to build an enhanced online version of Chemical Communications, is experimenting with new formats such as VRML (virtual reality markup language), but its real significance is the overall scope of its work and its ability to set the agenda and influence standards for electronic information within the field of chemistry. Take into account chemical MIME (multipurpose Internet multimedia extensions), chemistry Java, and you begin to get an idea of this project. Add the chemistry markup language (CML), the group's advocacy of XML and the specially-developed viewer, equivalent to a Web browser, which displays CML within XML structures, and this is a project of international stature.
Parallel editions are typically the most conservative form of e-journals. Ask Henry Rzepa, the project leader, why a research chemist should be interested in such developments and he will reply that what are emerging are not just new publications but new instruments for chemists. CLIC provides the evidence that even 'parallel' e-journals will not be journals as they are known today.
Where CLIC is forward-looking, ambition in the Parallel
Processing of Transactions project extends in the opposite direction along
the timeline. Beyond its production of PDF-based electronic editions of
current issues of the Transactions of the British Institute of Geographers,
the project has announced plans to produce digital copies of every issue
from the first in 1935. Copies will be scanned and processed into PDF and
the complete archive is expected to be available this year.
Table 10. Parallel
e-journal projects within eLib
||Activities of project|
|CLIC||Royal Society of Chemistry||Enhanced Chemical
Chemistry markup language
Latex to HTML
|Parallel Processing of Transactions||Royal Geographical Society
Institute of British Geographers
|Building a parallel e-version
of the Transactions of the Institute of British Geographers
Also with pre-prints, abstracting and indexing services, presented with enhanced Adobe Acrobat technology.
Plans to digitise the complete journal archive from 1935
Publishers recognise that e-journals are dynamically changing products, otherwise why would so many publishers be involved in these projects referred to here as 'integrating services' but which, as Table 11 shows, are more diverse than this label suggests? If publishers are active to an appreciable extent within the eLib framework it is in these projects. We, the authors of
this paper, have to declare a vested interest in one of
the listed projects, the Open Journal project, but in these three projects
lie the clues - information filtering, agents, links, multimedia - not
just to the next generation of e-journals but to the emerging shape of
the digital library. Clearly these projects will not provide all the answers
or the tools, but they are good starting points to understand how, also
why, e-journals will change.
Table 11. Integrating
services projects within eLib
|Project||Project description||No. attached publishers|
April 1996- March 1998
|'Personalised' information services – information filtering, software agents, metadata, etc. – for the library community. Pilot service planned Nov. 1997||2|
May 1995- May 1998
|Hypertext links for collected resources – journals, books, databases, essays – in biology, cognitive science and computer science. Demonstrations released for evaluation; open access by end 1997||11|
Oct. 1995- Dec. 1998
|Exploring value-added features – search, display, formats, multimedia – by creating subject 'clusters' of journals. Three clusters containing over 2,400 articles from over 30 journals released to test sites in first-half 1997||20|
For those who believe that the success of Paul Ginsparg's physics e-print archive based at Los Alamos is an isolated case, eLib's three e-print projects do not yet provide compelling evidence to the contrary (Table 12). Raw e-print archives are all about access and volume of papers, not the trimmings of value-added commercial journal publishing. Judged in this context, the one project to launch a service so far, Education-Line, made a solid start with 166 papers in the first month. Judged against e-journal projects with similar resources, this was a startling debut. Progress has been slower since, but from either viewpoint the project has justified eLib's decision to explore this area.
Of the projects still building technical frameworks and
yet to launch user services, CogPrints is notable because it is being built
on Ginsparg's system. Should the service attract a critical mass of papers,
not only will the service look and feel like the Ginsparg archive, there
could be moves to converge and ultimately merge the two. Where new journals
have traditionally required the support of prestigious publishers to achieve
visibility and authority, so might e-print archive services cluster around
their famous progenitor for the same reasons.
Table 12. E-print
projects within eLib
|Project||Launch of service||No. papers||Comment|
|Education-Line||Jan. 97||343||Includes searchable archives of
recent conference proceedings
Target: 2000 papers/year
eLib in perspective
The eLib programme is closer to its conclusion that its
launch and judgement draws nearer, perhaps measured in terms of the number
of successful exit strategies, or in the longer term on its wider impact
in the development of the digital library. Whichever it is, that time is
not now and this paper does not seek to judge it here, not least because
the programme is much larger than is described here. In aiming to contribute
to a framework for electronic libraries, eLib cannot be judged in terms
of the number of e-journals it produces.
Journals publishing has changed more dramatically in the
last two years than could reasonably have been anticipated. Such is the
extent of the change that some believe the end of the transition is in
view, with publishers hoping for stability in the market for electronic
access by 1998 (HEFCE 1997). Publishers have not been
the only ones to confound expectations, however, as random analysis of
some common beliefs about e-journals indicates (Table 13).
Table 13. Some
random beliefs about e-journals exposed
|The Web and HTML don’t support maths and complex symbols||Physics produces most online papers|
|Materials published on the Web lack integrity||Online biomedical journals are among the best supported|
|Electronic journals will cost less||The PSLI apart, most online journal packages cost more than print|
|Publishers do not want to be online||They do if current subscription models prevail|
|Publishing will change||When subscription agents' online services become available as well, nothing will have changed!|
In the change to e-journals publishers have proved more responsive, robust and more in demand than their detractors would admit, but those publishers hoping for stability are too optimistic - the market is hardly more than embryonic - and are not alert to the new opportunities. The truth is that technological change is rapid and there are now too many vested interests to let it slow. Every new participant in the process thus accelerates change. As far as e-journals are concerned publishers have leapt ahead of the market - whether wisely or briefly, they had no choice - but in awaiting the view of the market they must recognise this is the point at which the real race begins. In the light of recent publisher mergers and acquisitions, market repositioning is already evident.
The task of this paper was to view the current status of e-journals in the UK, which leaves little scope for analysis or prediction. Nor, frankly, is the evidence here alone a suitable basis on which to predict the future, but there are a number of reasons for arguing that further change is inevitable. Here is just one. The Web represents a new medium in which electronic information will diverge from its print counterpart.
There is one safe bet: we are approaching a time where every piece of electronic information will be viewed on the Web and shaped by society on the Web or its successor or like image. Early, pre-Internet e-journals failed for two reasons: poor interfaces for users, and the lack of coordinated international telecommunications networks (Freeman 1987). With the Web interface to the Internet, ubiquitous browsers and adoption of HTML/XML, users will increasingly find a uniform view of, and access to, all network resources.
For e-journals the belated arrival of established publishers
and a critical mass of journals online (Table 14) marks start of the race
to reinforce market positions, not its end. The prevailing wisdom is as
likely to be confounded as earlier expectations.
Table 14. 'UK'
|No. available in October
|Forthcoming (e-journals announced
|Projected total during
Authors' note: why 'circa'?
Counting e-journals is not easy, as we have said before.
The numbers change rapidly, and the journals publishing business is in
a greater state of flux than for some time. Even individual publishers
can be unsure of current totals: titles have been 'bought, sold, merged
and split ... there are titles which we no longer publish but to which
we hold the rights for previous years. Most of our recent publicity material
includes the word "about " whenever numbers are quoted for that reason'
said one. The overall picture, rather than precise numbers, is what we
are trying to convey.
Barron, D, 1989, Why use SGML? Electronic Publishing, Vol. 2, No. 1, April, 3-24
Bekhradnia, B, 1995, Pilot national site licence initiative for academic journals. Serials Vol. 8, No.3, November, 247-250
Borghuis, M, et al.,
1996, The TULIP project final report
Clement, G, 1994, Evolution of a species: science journals published on the Internet. Database, October/November, 44-54
Durniak, A 1997, Rationale
for American Chemical Society Web journal prices. Newsletter on Serials
Pricing Issues, No. 193, September 22
Entlich, R, et al.,
1995, The Chemistry Online Retrieval Experiment: a summary of the CORE
project (1991-1995). D-Lib Magazine, December
Freeman, D T, 1987, The false start of the electrical journal: a look at human factors and automation. Proceedings of the American Society for Information Science, Vol. 24, 79-82
Harnad, S, 1990, Scholarly
skywriting and the prepublication continuum of scientific enquiry. Psychological
Science, Vol. 1, No. 6, 342-343
Harter, S P, and Kim. H J,
1996, Electronic journals and scholarly communication: a citation and reference
study. Proceedings of the ASIS Midyear Meeting, San Diego, CA, May,
HEFCE, 1997, Report on Phase
I of the Evaluation of the UK Pilot Site Licence Initiative. HEFCE Ref
M 3/97, April
A summary of the report is at http://back.niss.ac.uk/education/hefce/pub97/m3_97.html
Hitchcock, S, et
al., 1998, Linking everything to everything: journal publishing myth
or reality? Serials Review, to appear
Hitchcock, S, Carr, L, and
Hall, W, 1996, A survey of STM online journals: the calm before the
storm. Directory of Electronic Journals, Newsletters and Academic Discussion
Lists, edited by D. Mogge, 6th edition (Washington, D.C.: Association
of Research Libraries), pp. 7-32
Lie, H W, and Bos, B, 1997, Cascading
Style Sheets: Designing for the Web (Harlow, UK: Addison-Wesley).
Web updates http://www.awl.com/css/
Lucier, R E, and Brantley, P,
1995, The Red Sage Project: an experimental digital journal library for
the health sciences. D-Lib Magazine, August
McKnight, C, 1997, Designing the electronic journal: why bother? Serials, Vol. 10, No. 2, July, 184-188
McKnight, C, 1993, Electronic journals - past, present...and future? Aslib Proceedings, Vol. 45, No. 1, 7-10
PSLI Evaluation Team, 1997, UK pilot site licence initiative: a progress report. Serials, Vol. 10, No. 1, March, 17-20
Roes, H, 1995, Electronic journals: a survey of the literature and the Net. Journal of Information Networking, Vol. 2, No. 3, 169 -186 http://cwis.kub.nl/~dbi/users/roes/articles/ej_join.htm
Rouse, K, 1997, The serials
crisis in the age of electronic access. Newsletter on Serials Pricing
Issues, No. 177, May 7
Schauder, D, 1994, Electronic publishing of professional articles: attitudes of academics and implications for the scholarly communication industry. Journal of the American Society for Information Science, Vol. 48, No. 2, 73-100
Senders, J W, 1977, An online scientific journal. Information Scientist, Vol. 1, No. 1, 3-9
Smith, P N, et al.,
1993, Journal publishing with Acrobat: the CAJUN project. Electronic
Publishing-ODD, Vol. 6, No. 4, December, 1-13
Tavistock Institute, 1997,
Evaluation of the Electronic Libraries Programme - synthesis of (first
year) annual reports
Wusteman, J, 1997, The evolving use of SGML in electronic journals. ICCC/IFIP conference on Electronic Publishing '97: New Models and Opportunities, Canterbury, UK, April