Lessons Learned in Publishing an Electronic Journal.

Angela Prior, Monash University Library, Wellington Rd, Clayton, Vic 3165 Australia

Zenon J Pudlowski, Terry Berreen and Roger Hadgraft, UNESCO Supported International Centre for Engineering Education, Monash University, Wellington Rd, Clayton, Vic 3165 Australia

Introduction

Electronic journals are proliferating on the Internet. In 1994 the number of electronic journals in the area of science, technology and medicine was 13. In 1995, this was 115 [HREF4]. In 1996, the number of journals will explode, as a number of government led initiatives such as Higher Education Funding Councils' projects, will start bearing results. It is important to remember however, that electronic journals are not a new concept. As early as 1977, research began on the possibilities and implementation of an electronic journal (Gaines, 1993) with the work of John Senders. The result was published in 1981 with the paper being titled "I have seen the future: and it doesn't work" (Senders, 1981). Other attempts, such as BLEND and the hypertext QUARTET from HUSAT (McKnight, 1995), although less pessimistic, did not survive. In general this was due to the technology being insufficient but also to user inexperience with technology. Important successes were also noted. In particular the BLEND project, noted the valuable use of extra facilities, such as on-line discussion and comment facilities.

The World Wide Web (WWW) has changed the electronic journal from being purely a research endeavour to being an on-line reality. However the process is still in its infancy. This is illustrated by the varying number of formats available. HTML is not necessarily regarded as a standard, particularly by those working with complex technical diagrams and figures [HREF4]. Acrobat PDF files are often used, as are Postscript and TEX files. Indeed the TULIP project, which started pre-WWW, is still using TIFF files, although conversion processes are taking place [HREF7].

If HTML is not being used then many of the advantages of a hypertext system will not be available. Hitchcock et al [HREF4] reported that the majority of electronic journals in the survey were not markedly enhanced by their format. Many were just electronic replicas of the paper version. A few journals were implementing intra-journal links, fewer still implemented world-wide links, listservers, search engines or multimedia capabilities. It may be some time before all the research possibilities, explored almost two decades ago, are integrated into the electronic journal.

A number of interesting research initiatives are continuing. It is worth describing a few of these briefly.

• Interjournal (Redi and Bar-Yam, 1995) is a project, which seeks to use a distributed model for publication where the journal only maintains a link to the paper in a distant URL. In addition, several different fields are considered under the one banner, allowing crosslinking of related but separate concepts.
• The Open Journal [HREF5], is looking to extend the value of a journal by providing easier ways for the users to move between the journal article and other on-line resources, such as gene sequence databases.
• QUEST - Query Environment for Science Teaching (Shneiderman et al 1994) - is an example of research into accessing electronic information, with an emphasis on how the users will query the electronic data. Most of this experimental work changes the role of the publisher, the library and even the reader.

Equally exciting is the number of journals being placed on-line. Bodies like the ACM (Association for Computer Machinery) have defined electronic publishing plans, which include the WWW and also define their position on copyright issues concerning hypertext links to other documents [HREF3]. Most major publishing companies have committed to some form of electronic publication. This commitment comes in different forms: sponsorship - John Wiley and Sons are sponsoring the CAJUN (CD-ROM Acrobat Journals using Networks) [HREF2]; collaboration, such as the ELVYN project between a library and a publishing company (McKnight 94); or direct support, such as Elsevier's TULIP project [HREF7]. The wide range of electronic material available underlines the reality of the electronic medium and the need for more tools to cope with the ensuing information overload.

The Monash initiative

In 1995, the Australasian Journal of Engineering Education (AJEE) [HREF1] was published on the WWW. This paper journal is published by the Australasian Association for Engineering Education which is hosted by Monash University. The electronic version was produced jointly by three groups within Monash University: the UNESCO Supported International Centre for Engineering Education (USICEE) in the Faculty of Engineering; the Unit of Medical Informatics, Faculty of Medicine; and the Monash University Library. The project was funded by a research grant provided by the Australian Vice-Chancellors' Committee and the Electronic Publishing Working Group from the National Priority Reserve Fund library infrastructure grant.

The aims of the project were clearly defined in the proposal. These were: "

1. to improve browsing of the contents of the journal through: hypertext, allowing on-linear presentation of the material; and electronic searching mechanisms, using keywords, to retrieve relevant information.
2. to improve general academic access through the widespread distribution via the Internet.
3. to make the submission and publication process more efficient
4. to set up an infrastructure so that this publishing technique can be utilised by other journals." (Monash University Electronic Journal Proposal).

Monash Medical Informatics' (MMI) role in the creation of the journal was two-fold. First, to perform tasks that required technical expertise, but would not require repetition. Such tasks included: establishing the web server; implementing the search engine for the journal; and preparing CGI scripts for the forms. MMI's other role was to train a librarian, Angela Prior, in HTML, so that the library could be responsible for the on-going publication of the electronic journal. The long term future of the journal was seen as a collaboration purely between the library and the content providers, in this case, the USICEE.

This paper will not focus on the final outcomes of the project. The issues of user acceptance are discussed in Groenewegen et al (1996). For further details on the This discussion will cover some of the problems in developing a process for publishing an electronic journal. All of these problems were resolved, but awareness of these pitfalls would have markedly streamlined the process. These problems are described below.

Limitations of the technology

An engineering journal, like other science journals, often contains specialised terminology, complex figures and complex tables. There were definite problems in reproducing the quality of the paper version in HTML. Compromise had to be made on a number of formatting issues, particularly reltated to equations, superscripts, subscripts and technical symbols. It was difficult for the library and technical personnel to understand the importance of quality appearance, as they were not familiar with the rigour of the editing and publishing process. Graphic files were used for most diagrams, and the HTML 3 tables were adequate for most tables. However common mathematical symbols in the text, posed some problems. In the end small graphic files were used, inserted into the text. A thorough search of the Web indicated that this was the only approach available. The desire for publishers to use Postscript or PDF formats, with an associated viewer is understandable as the HTML documents do not have the appearance of well published paper journal. If however, HTML is not utilised, then the many other electronic advantages of the electronic format are not available.

The other potential limitation is the delay time required to download and read the document. This does not appear to be a problem with the AJEE, although formal evaluation will enable better assessment of this.

Hypertext is not intuitive

One of the most striking problems was unintuitive nature of the medium. Those unfamiliar with the technology had difficulty in understanding the principles of hypertext. Structuring a journal, or a paper, involves some understanding of the medium. The non-linear nature of traversing the journal was not grasped until specifically demonstrated.

The librarian responsible for converting the documents from Word 6 for Windows into HTML did not know where to start in terms of designing the overall layout of the journal. She had found the acquisition of the HTML skills easy and fun, but still did not understand how to develop the `big picture'. This was a source of frustration. She consulted various authors and editors but they could not provide her with assistance due to their lack of knowledge of the medium. It was difficult for the MMI staff involved in technical support to assist, as the librarian was unable to articulate that she required a discussion of the structure and form of an electronic document. This is not a new problem. These difficulties have been observed for many years in the area of Computer Assisted Learning (CAL) where paper based material had also been converted to electronic form, often without thought to the new medium.

Several concepts needed to be conveyed. In hindsight, these should have been conveyed at the beginning of the process, rather than midway through. Many aspects of hypertext are second nature to those who have been working on-line but creating useable hypertext is not intuitive. Moreover, all those involved in the process, especially authors should be provided with some basic material to outline the concepts behind HTML. In the AJEE experience, the following ideas, which may appear entirely self-evident, should have been conveyed:

Information can be divided

A fundamental concept of hypertext is the concept of a node. This implies `chunking' or dividing the information into envelopes, which are not designed to be subdivided any further. For example, it was not immediately apparent that the biographies of the authors might appear separately from their papers in case the authors contributed further papers, or that someone might wish to access all the papers by one author from a single point.

It may be worth teaching Shneiderman et al's Golden Rules of Hypertext (Shneiderman, 1991). These Rules indicate conditions for creating hypertext and are:  "

• There is a large body of information organized into numerous fragments.
• The fragments relate to each other.
• The reader needs only a small fraction at a time."

Web pages have a structure

People may not immediately think to draw a diagram of the Web structure. It is not necessarily obvious that the Web can be structured. The use of a concept map (or semantic net) to illustrate and determine structures is well documented in the literature (Eklund 1995). It is worth explicitly explaining that 'chunks' of information are joined by links, and that these links represent relationships between the information 'chunks'. It took some consideration of the following concept maps, drawn as an example of the possible structure, for the HTML mark-up librarian to understand the issues at hand. Until this time, the Web had seemed amorphous. Figure 1 demonstrates the sort of simple diagram that was useful in describing the some of the concepts involved.

Concept maps do not have to contain new or radical relationships. Generally speaking the logical structure of the paper is regarded as being the most appropriate structure (Furuta, 1994). A logical structure reflects the standard, familiar, organisation of an academic journal article.

There are different modes of navigation

Readers should have to navigate as little as possible to get to the information they want. It may not be clear that various tools like search engines, or links to both authors and papers, are examples of increasing the readers' options of accessing the information.

Readers must be considered

It is important to indicate that the readers of the journal must be served by the journal structure. The structure may still be flawed, and may be adjusted after feedback, but the process of development must be user-centred. Any structure should be logical and clear. Even simple things like placing the required version number of Netscape at the head of the journal, are important.

Graphic design is not intuitive

The problems with Web page layout are obvious to anyone who surfs the Net. Even professional publishing sites contravene some of the basic rules of screen design. The availability of backgrounds, has resulted in unreadable text lying over busy patterns. If there is no funding or necessity for a graphic designer, then consider the basic rule of 'keep it simple'. An electronic journal will be read on the basis of its content, not on the basis of its groovy background.

Resistance from contributors

In a project such as this, where the contributors were primarily interested in submitting to a paper version, there may be some queries as to why electronic publishing is necessary. Authors may question the validity of the electronic form, or request impossible tasks, such as putting past LaTeX papers on line. In the AJEE experience, simple explanation was enough to resolve most problems. Problems seemed to arise mainly from inexperience with the medium. This is a problem which has been reported with earlier electronic journals (Gaines, 1993) and will probably continue for some years to come.

Solutions

In the AJEE project, the direct input was either technical, or content. There was no staff directly employed to oversee the structural and conceptual changes necessary to transfer paper to the WWW. This type of void had been reported in the literature and various solutions have been proposed. Examples include:

• A proposed style designer whose role is to specify the appearance of a logically structured document (Furuta, 1994).
• MITRE Corporation's knowledge management program for its digital library (Flynn, 1995). This position was created "in order to ease the transition and identify and meet the information needs of the technical staff during the transition and in the new environment".
• The Training Electronic Journal [HREF6]. This is probably the best example of a serious attempt to teach necessary skills to publishing and librarianship students. In this project, library and information studies students manage an electronic journal as part of their training. In this way they learn the necessary skills involved in the new medium and are introduced to electronic communications, hypertext and so on.

It may be difficult for the small publisher, such as Monash University, to find individuals with the correct type of training. For instance, the librarian involved in the AJEE process does not regard her role as structuring a document, but doing the physical work of creating the HTML. She states:

"I can do anything that they want me to do, but they have to tell me exactly what they want."

In some ways, this is appropriate, as creating a well-structured hypertext document often requires some understanding of the information (Shneiderman et al 1991) and should be done in consultation with the authors. Fortunately for this project, in creating the initial edition of the journal, templates have been set for further editions. Part of the purpose of the project is to document these structures and to make them available to further electronic publishing ventures.

In hindsight, two iniatives would have been sufficient to prevent problems.

Put an instructional designer on the team

In hindsight, the AJEE grant should have budgeted for an individual with instructional design or hypertext skills. Some of the problems being encountered are familiar to those working in the area of CAL. CAL designers are familiar with the problems of bridging the gap between medium and the message. They have been trained to tailor the design to suit the need of the readers. Fortunately for the AJEE project, these skills existed within MMI and the Faculty of Engineering, and informal communications were sufficient, although far less efficient, and possibly less effective.

Educate authors, editors and staff

An instructional designer might have also been responsible for a formal demonstration of the technology with an emphasis on previous research and educational advantages to electronic journals. This may well have prevented concerns and problems further into the project. Such a formal presentation would provide validity to the process, as there always seemed to be a lingering `why' as to the reasons for the new medium. A physical demonstration is always best when dealing with new technology, as electronic innovations are often very difficult to understand without concrete examples.

The continuation of the project has prompted the development of new guidelines for authors:

Microsoft Word template

A new template has been developed in Microsoft Word to improve the physical layout of submitted papers. This template encourages authors to get it right before they submit, with the inclusion of standard headings, graphics formats, table formats, caption styles and reference styles.

Guidelines for writing electronic papers

A more complex issue, only partly resolved, is the development of a guideline for writing a paper which is better suited to the electronic medium, but which will also be highly readable in printed form. It is expected that this will encompass a two level approach with each paper consisting of a one page mini-paper which sets out the paper's context, its methods and its conclusions, with a further 4-5 pages providing the detail where it is needed. The mini-paper replaces the brief Abstract which is currently used, and should allow a reader to grasp the paper's message, without reading the entire paper. In that sense, it becomes an effect Executive Summary, an essential attribute in these busy times.

Conclusions

The continuation of the AJEE journal, like other publications, is dependent on the de facto standards being set by the publishing industry. Although there is no question that the World Wide Web is here to stay, it is impossible to predict the exact future of electronic journals. Standards may work with the medium and enhance scholarly communications or they may simply be delivered through a more cost effective medium. Standards may never be implemented and the multiplicity of formats may grow rather than shrink. This is however, in essence, a non-technical problem.

There is no question that the technology is, in general, able to cope with the concept of the electronic library and the digital journal. Some details and concerns exist about bandwidth and quality assurance but the strength of the burgeoning industry is undeniable. On the other hand, in 1981, Senders reported about EIES

"many of the participants, not themselves experienced with computers, found EIES inconvenient or difficult to use".

This attitude was still encountered in 1996, where all the technical problems of 1981 had vanished. Unfamiliarity with the new technology was still apparent in the AJEE project, and probably caused the most difficulty in the publishing process. This experience confirms that greatest difficultly with electronic publishing lies with changing the current infrastructure and process. The technical changes are quick and easy but the social changes are the slowest and the most difficult to achieve.

Published References

J Eklund (1995) "Cognitive models for structuring hypermedia and implications for learning from the world-wide web" , Proceedings of AusWeb 95, pp 111-117 Ballina, May 1995, http://www.scu.edu.au:80/ausweb95/papers/hypertext/eklund/index.html

KM Flynn "The Knowledge Manager as a Digital Librarian: An Overview of the Knowledge Management Pilot Program at the MITRE Corporation", Proceedings of Digital Libraries `95, Austin, June 1995, http://www.csdl.tamu.edu/DL95/papers/flynn/flynn.html

R Furuta (1994) "Defining and Using Structure in Digital Documents", Proceedings of Digital Libraries `94, College Station, June 1994, http://www.csdl.tamu.edu/DL94/paper/furuta.html

BR Gaines (1993), "An Agenda for Digital Journals: The Socio-Technical Infrastructure of Knowledge Dissemination", Journal of Organizational Computing, 3(2) 135-193, 1993, http://ksi.cpsc.ucalgary.ca/articles/abstracts.html#DigitalJ

HW Groenewegen, B Cesnik, A Prior and ZJ Pudlowski (1996) "Experiences in electronic publishing: the Monash University electronic journal project" to be presented at the 3rd East-West Congress on Engineering Education, Gdynia, Poland, 15-20 September 1996.

C McKnight (1995) "Digital Library Research at Loughborough: the Last Fifteen Years", Proceedings of Digital Libraries 95, Austin, June 1995, http://www.csdl.tamu.edu/DL95/papers/mcknight/mcknight.html

C McKnight et al (1994)"ELVYN -- Publisher and Library Working Towards, the Electronic Distribution and Use of Journals, Proceedings of Digital Libraries", College Station, June 1994, http://www.csdl.tamu.edu/DL94/paper/mcknight.html

"Monash University Electronic Journal Proposal" (1994)

J Redi and Y Bar-Yam (1995) "InterJournal: a distributed refereed electronic journal", Proceedings of DAGS95: Electronic Publishing and the Information Superhighway, Boston 1995, http://www.cs. dartmouth.edu/~samr/DAGS95/Papers/redi.html

JW Senders (1981) "I have seen the future and it doesn't work: The electronic journal experiment." Proceedings of the Society for Scholarly Publishing 2nd Annual Meeting. Washington, Society for Scholarly Publishing,1981

B Shneiderman et al (1991), "Editing to Structure a Reader's Experience" in the "Hypertext/Hypermedia Handbook" (1994), McGraw-Hill, NY, 1991

B Shneiderman et al (1994) QUEST --Query Environment for Science Teaching, Proceedings of Digital Libraries 94, College Station, June 1994, http://www.csdl.tamu.edu/DL94/paper/shneiderman.html

Hypertext References

HREF1 Australasian Journal of Engineering Education, http://elecpress.lib.monash.edu.au/ajee/

HREF2 CAJUN Home Page http://www.ep.cs.nott.ac.uk/cajun.html

HREF3 Denning PJ, The ACM Electronic Publishing Plan, http://www.acm.org/pbus/epub_plan.txt

HREF4 Hitchcock S et al, A survey of STM online journals 1990-1995: the calm before the storm, http://journals.ecs.soton.ac.uk/survey/survey.html 1996a

HREF5 Hitchcock S, An Open Journal Framework: integrating Electronic Journals with Networked Information Resources, http://journals.ecs.soton.ac.uk/flyer.html 1996b

HREF6 Training Electronic Journal (TEJ) http://info.lut.ac.uk/departments/dils/tej/tej.html

HREF7 TULIP Newsletter no 4, April 1994, http://www.elsevier.nl/info/projects/newslet4.htm

Copyright