We are still in the early days of electronic journals. This phase brings upon us a variety of formats and delivery methods. Formats include bitmaps, PostScript, PDF (Portable Document Format) as well as ASCII (American Standard Code for Information Interchange), SGML and HTML. Electronic journals might be delivered on CD-ROM, by electronic mail, or through the networks. Networked journals require that users actively access the publishers' sites; for convenience of the subscribers, new issues often are announced through e-mail notifications. Although publishers are still experimenting with various formats, the extensive use of the World Wide Web certainly will lead to a future of electronic journals that is network-based (Wusteman, 1996).
The increase in the number of so-called electronic journals is breathtaking. In the introduction to the 1996 edition of the ``Directory of Electronic Journals, Newsletters and Academic Discussion Lists'', Ann Okerson noted that ``in this sixth edition, the number of journal and newsletter titles (nearly 1700) has more than doubled since last year's and multiplied by over 15 times since the first edition (there were 110 listings in 1991, 240 in 1993, and nearly 700 last year)'' (Okerson, 1996b). Most of these journals are not ``true electronic journals'' though. In a widely distributed survey article by Steve Hitchcock, Leslie Carr and Wendy Hall (1996), the authors point out that currently most of the electronic journals are rather electronic versions of paper journals than fully operational e-journals; that is, they are confined to presenting bitmapped images or electronic formats intended for printing, like PostScript or PDF. In order to fully exploit the capabilities and advantages of electronic journals, they must be able to interlink sources and therefore need to go beyond print-oriented format as soon as possible.
The acceptance of electronic journals among the scientific community currently is not as high as the large number of electronic versions seems to indicate. According to Ann Schaffner (1994), in order for e-journals to be accepted by authors, it is necessary that they fulfill at least the basic functions of paper-based journals, which she defines as
Accessing electronic publications over the Internet can be unacceptably slow, and incorrect or out-of-date electronic addresses can turn locating documents into an extremely time-consuming task. Reliable links are the backbone of electronic services. A study by Harter and Kim (1996) revealed that 55% (71 out of 129) of the links to e-journal archives the authors tried to access did not work at first try. If electronic journals are to be effectively used, the reliability of electronic references must stabilize.
Harter and Kim also analyzed to what extent authors of e-journal articles (in whom one can assume a sympathetic attitude towards electronic media) are citing electronic journals and other online resources (electronic personal papers, newsgroup postings, electronic preprints etc.). Their findings showed that only 1.9% (83 out of a total of 4,317) of the citations in peer-reviewed articles published in 1995 referred to electronic sources. Indexing services can help increasing the acceptance of electronic publications but they are just beginning to include e-journals in their databases. Electronic journals must struggle to meet the criteria for inclusion applied by abstract services in order to be eligible.
One of the most important problems of electronic publications is archiving. Unless a stable system for future retrieval of publications is in place, authors will continue to be reluctant to publish in electronic journals.