Why berate publishing on the Internet for not mimicking traditional methods, asks Fred Nash, when it can offer so many unforeseen advantages?

Publication on the Internet is characterised by a distinctive feature which, for some, is also its major problem. It leaves no physical record of what is "published" on it; there is no actual and "preservable" "thing" with a unique ISBN reference and a copy at the British Library. From this issues the all important source citation problem.

For some (for example, Jim Smith, Multimedia, October 13 1995) that is precisely where the problem lies. But given that the Internet is a different medium of publication, we ought not to expect it to mimic the characteristics of the heretofore "permanent" pattern of publication.

Having just launched Political Science Discourse as the bulletin of the political science specialist group of the Political Studies Association of the United Kingdom (), I am conscious of the need for a set of protocols, which is more likely to emerge as a result of a reasonable convention being adopted by the majority of such publications.

ÖйúAƬ

Publication on the Internet portends far-reaching changes in the structure of scientific (at any rate, social science) community, and science as we know it. Importantly, bulletins on the Internet, such as the PSD, are not suitable conduits for the publication of long pieces: they are not a rival to books, but to articles. To appreciate the scale of the change and the nature of the potential, we must remind ourselves of the characteristics of hard copy publication of articles.

Articles are primarily the outcome of research, and they are valuable precisely to the extent to which they inform their research community, provoking reflection and reaction. Characteristically, a published piece is also a refereed piece: it is the end result of a long process of selection and judgement. And whereas science is not the property of any one individual scientist or group, the process of publication - subject, as it is, to the practice of refereeing and selection - necessarily privileges the views of some, allowing a select few to determine the direction and the shape, the very structure of knowledge in that discipline and, with it, its processes of accumulation and change.

ÖйúAƬ

Examined from this perspective, the "articles industry" is an odd one. A refereed article is only published if it conforms to some unspecified rules and norms. In getting past the editor, there is the gauntlet of unspecified referees - one's colleagues - who supposedly do not know the identity of the author whose work they are judging. Of course there are numerous examples of the referee knowing, not only the author's identity, but also him or her in person. But the author hardly ever knows the identity of his or her secret judges.

This secrecy is said to be necessary in order to enable them to give an honest academic opinion. And the point of the exercise is to ensure that the piece is "publishable". But this euphemism only means that a couple of supposed authorities have sanctioned it as a valid statement. The rewriting that follows may improve the quality of the outcome: comments are always relevant.

The point is that an article, when published, will present not so much an idea under consideration, but the reporting of an idea sanctioned by the "reputation" of a refereed journal. This is hardly anything less than a kind of covert paternalism exercised on author and reader, which Paul Ormerod has likened to the Spanish Inquisition (THES October 13 1995). There is also the possibility of "parafraud". as highlighted by Harold Hillman, in the same issue.

Internet articles present a different attitude, in contrast to this essentially antediluvian and authoritarian style of publication. But we must be clear about the limit of our expectations here. Given that the focus of interest for, say, PSD, is to encourage and enable scientific discourse within the discipline, the Internet can enable us to do at least four things which at this point are simply not allowed to happen.

It can liberate the scientific community from the financial considerations and limitations that the interests of a commercial publishing firm necessarily impose. It can accelerate the exchange of ideas by removing the vetting and publication lead time. A piece can be on the Internet in a matter of days, and responses to it can be added in a matter of minutes.

Some may consider the Internet more suitable for contemporary empirical pieces which will lose their relevance if published after a lapse of time and which require frequent updating. Of course the Internet has a significant potential as a source of information, but this is not an exclusive use for it. For example, PSD has a far more theoretical ethos.

It can enable us to generate genuine debate and real and open exchange of ideas between all who have something to say on the subject. Significantly, this will mean that we can see the actual process of reasoning and can examine the nuances of difference. The reader participates, instead of reading a polished report. It can enable more papers, and views, to be published: considerations of space - the thickness of an issue - and cost of production, printing and distribution are no longer relevant.

Such an open approach - in principle similar to the "open refereeing" process of the behavioural and brain sciences - will also mean that research, always work in progress, is now precisely that, with the added advantage that it is possible actually to engage with the argument. But this approach also raises at least three sets of issues.

ÖйúAƬ

ÖйúAƬ

The possibility of increasing the number and range of papers published may encourage the fragmentation of disciplines, a tendency already evident in the proliferating number of more specialised journals. But greater accessibility can invite more people to look up papers that are not in their narrow field of speciality. This puts further pressure on people's available research and reading time. For some, therefore, less restrained publication on the Internet may mean they run the risk of becoming overwhelmed by the sheer quantity of material. After all, the traditional vetting process also weeds out many papers.

But it also weeds out good material, while stifling creativity and innovation: it replaces risk with safety, and introduces a degree of conservatism into the process of change. The openness of the Internet will change all that. The gate-keeping role of the editor and the controlling function of the secret referee, which are the present instruments of creating and maintaining a balance between quantity and quality, will be demolished.

Exactly what the new balance will come to depend on is not yet clear. But it will certainly be easier to peruse a larger number of readily accessible papers and articles, or read the abstracts, even a few lines, and look at the references and footnotes in order to determine whether the article is relevant or not. Access is not limited by library opening hours, or the availability of the copy on the shelf. Looking up related articles - via hypertext links - will be easier. The reader can choose what to read from a menu offering a wide and varied choice.

Second, apart from articles, more literature reviews, short papers and comments can be published. At the moment very few people get to comment on books; and such comments often appear in print a good 12 months - or more - later. This area is an obvious candidate for change.

Third, there is concern about permanence, citation and security. It is not in the interests of, say, PSD not to offer a permanent record. As practitioners we work for the discipline and recognise that a record of publication is essential for research. Some may think of publication on the Internet as an addition to publication elsewhere; such a duplication would add the advantage of accessibility to traditional permanence. But an auxiliary role for the Internet will also stifle the potential that it has to offer. Unless the publication of an article on the Internet is recognised in its own right and taken seriously, the whole exercise will become self-defeating.

Articles on the Internet can be updated, revised, or added to. This is not necessarily a disadvantage, as such changes are always clearly stated. Meanwhile, a permanent record of all the articles and so on is always kept on a different site. The fear that an article on the Internet may not be there tomorrow, or that it may be different, implying that the original has disappeared, is simply mistaken.

Moreover, given that the editor of such a bulletin on the Internet has no direct interest in the articles, but does have a serious interest in the role an Internet bulletin can play for an academic subject, he or she can ensure it becomes a more permanent record of exchange of ideas. This is encouraged by the fact that bulletins are characteristically hosted where they truly belong, at universities, where no extraneous factors interfere with their role and future. And bulletins, can produce a secure CD-Rom annual record and deposit it at designated sites.

Hacking may be a problem. But then a piece can always be restored to its former condition: what is on the Internet is always backed up elsewhere. Or, which may be more of a problem, a hacker may intelligently change the message.

However, one's first thought is that any one sufficiently familiar with a discipline so as to be able intelligently to interfere with an article is likely to be a member of the profession, and therefore unlikely to act in this irresponsible manner. That said, there is a need to be vigilant.

As it stands today, the publication of articles is a closed process. The Internet can and will open this up to the advantage of science, the scientific community and society at large. The shape of science, famously determined by the invention of the printing press, is now set to change thanks to the infinite possibilities of electronic publishing. But we must recognise the challenge and tailor the system to suit our requirements.

ÖйúAƬ

The author is a research fellow in the department of politics, University of Southampton, and convener of the political science specialist group, PSA-UK.

Email psd@soton.ac.uk.