ACADEMIC CO-OPERATIVES FOR WEB JOURNALS: An Opportunity for Better Publishing Will Hopkins, Physiology & Physical Education, University of Otago, Dunedin NZ Sportscience News July-Aug 1997  http://www.sportsci.org/news/news9707/publish.htm

Web-based journals published by co-operatives of academics can improve on electronic carbon copies of conventional journals through faster turnaround of submitted material, a more accountable interactive review process, updatable articles and reviewing, retention of copyright by authors, and free access for the widest possible readership. Strategies to ensure high quality from the first issue include active participation by a high-profile advisory board, a no-risk draft first issue, and tiered levels of excellence within the journal. A journal based on these principles and devoted to sport science will be published at this site in 1998.

In the first issue of Sportscience News I wrote an article on the possibilities for a new Web journal devoted to sport science. Six months later my experiences of Web and conventional publishing have made me even more convinced that we should get this journal up and running.

In this article I update the issues from the perspective of Web journals generally. The article represents the substance of a seminar I gave in Cape Town and Perth. I have also produced a practical guide to Web authoring for a workshop I ran in both places.

I'll begin by highlighting the problems of conventional publishing. I'll then indicate how I think a Web-based journal published by academics provides an opportunity to overcome these problems. I'll finish by addressing the issue of quality.

Conventional Publishing
 
If you are a productive researcher, you will be familiar with the shortcomings of the conventional publication process. For the benefit of research students and other newcomers, here are the main problems:

• Slow turnaround of manuscripts. In my experience it averages a year, plus or minus six months.
• Flawed review process. The main problem seems to be rejection of worthwhile work.
• Substantial page charges for some journals.
• Upwardly spiraling journal subscription rates. As libraries cut back their subscriptions, the cost per issue increases.
• Slow access. If you or your library don't subscribe, copies of articles can take weeks to arrive.
• Authors have to sign away the copyright of their creative works.

Web publishing can get around all these problems. And it has several extra advantages: articles, and reviews of articles, can be updated; links can be added to take readers directly to related works in the same journal or other accessible sites; and you can expect your article to be read by more than just every researcher in your field.

You can already access some journals in our discipline on the Web. Problem: you have to be a paid-up subscriber. Other problems: these journals are just electronic carbon copies of the print versions, so it's the same flawed review process and slow expensive publication process. And you still lose the copyright. In short, publishers of conventional journals have not taken advantage of the move to the Web. Enter Web-publishing co-operatives run by the academic community.

Web-Publishing Co-operatives
 
The viability of the Sportscience website proves that academics can take virtually full command of the publication process. I say virtually, because running the server that connects your site to the Net is a job most academics can't do. But once the site is set up, those with access privilege can change documents at the site autonomously. You seldom have to liaise with the person running the server.

You need a sponsor to provide space on a server and to meet the costs of connection to the Net. In most parts of the world your academic institution will sponsor you. In our case the Royal Society of NZ offered to host the site and pay for the Web traffic. (The Royal Society has promoted science since the time of Newton.)

Every publishing house has an infrastructure of editors and reviewers. In the case of journal publishing, most or all of those positions are filled by academics on a voluntary basis. A similar voluntary co-operative of academics is needed for Web publishing. We've found that it takes time to get it together. By nature or calling, academics are skeptical independent individuals. So we still don't have a full complement for the editorial advisory board, but we are confident that we will end up with the right people.

A new role in the co-operative is the webmaster, a person who can convert a manuscript to a Web document and upload it to the site. You need several people willing to take on this job. As authors get more Web savvy, they will submit their articles as Web documents. Web-wise editors can also upload documents to the site. The webmasters will then have more time to add features to the site and improve its cosmetics and user-friendliness.

Other details... A short domain name ending in .org is well worth the investment (e.g. www.sportsci.org). You need a system of mailbags, preferably linked to the URL of the site (editor=AT=sportsci.org, webmaster=AT=sportsci.org...). A separate site, preferably physically separate for backup security, is also useful but not essential for preparation of the next issue.

An Opportunity for Better Publishing
 
OK, let's look at how a fresh start with a Web journal can overcome the problems of conventional journals and add new dimensions to the publishing process.

Faster Turnaround of Manuscripts
 
There's nothing to stop journals speeding up the review process. But when submission and correspondence is by snail mail, which can take weeks, the expectation is that reviewers and editors should be allowed a similar time to deal with the manuscript. They could usually deal with it immediately, but instead they leave it until the last possible moment. Months later, the author gets the manuscript, and promptly shelves it until the deadline for dealing with it (several months again). Then there's another complete iteration of the review process, and sometimes a third round. Finally, there's several months between acceptance and printing that cannot be shortened by conventional publishers.

Solution: the entire review and editing process is handled electronically. When the delay in reading a message is determined only by time-zone differences, you can expect everyone to deal with a manuscript in a matter of days, not weeks or months. Several iterations of the review process can be achieved in a week or two. And it's so much easier, when the work stays fresh in the minds of the authors, editors, and reviewers. The final delay--uploading to the Web--can be a day or two or a month or two, depending on the co-operative's policy on publishing either single items or complete issues.

Accountable Interactive Reviewing
 
Reviewing should be double-blind initially. I don't agree with the argument that it's easier to judge a work when you know the background of the authors or the lab. It just makes it easier to accept bad work from a big name and reject good work from a nobody.

After the manuscript has been accepted or rejected, the blinds should be removed for both parties. The decision to accept or reject is taken by the editorial board, not by the reviewers, so bad feelings about rejection will be directed mostly at the board. For articles that get accepted, the authors and reviewers agree to disagree on points of interpretation, and an agreed review under the reviewer's name is published along with the revised article. In short, accountable dialogue replaces anonymous diatribe.

This procedure would reduce the carelessness and nepotism of the review process. If your name is on your review for the public to see, you've got to get it right, and you've got to be seen to be fair. The bonus for reviewers is another line on a vita: your review is a creative work, however short it might be. The line in your vita is also objective evidence that you have been a reviewer. An annual mention in the back of the journal isn't good enough.

I foresee a higher rate of acceptance of manuscripts with this method of reviewing, simply because differences of opinion over interpretation will no longer be adequate grounds for rejection. Irreconcilable differences in opinion will be highlighted in the agreed review or reviews that are published with the article. Further debate is bound to follow and will be published as additional reviewers' comments. The author will always have a right of reply, of course.

What's stopping current journals from adopting this review process? The conservative side of human nature, which in this case is maladaptive.

Free!
 
Free access to your article means that many more people will read it. Nearly every academic and research student in the developed world has access to the Web. Many non-academic professionals who would otherwise not bother to track down your article in print will also read it on the Web.

There's no such thing as a free website, of course, even for readers. And if there are enough readers in an institution, it may cost the institution less to buy a single paper copy of the journal than to pay for the Web traffic. A way around this problem is for the institution to host a mirror image of the journal. Only one external download is then necessary. I predict that libraries will end up being electronic repositories of journals.

Updateable Articles and Reviewers' Comments
 
There is no good reason why an article or reviewers' comments should be fixed from the time of publication. Changes can obviously be highlighted appropriately. New reviews can also be added by people who have a valid point to make, and links can be added to point readers to publications that come out subsequently. None of these options are available to printed journals.

Access Full Articles
 
You can read the abstract or the full articles of a journal published by a co-operative. And you can do it from any institution you visit. If you have a laptop with a modem, you can join up with a local or international provider and access the journal from your hotel room.

You can print or download articles you are really interested in. Software for keeping track of electronic copies of Web pages is also becoming available. Reference managers like Endnote will become more Web friendly, too. For example, URLs will have their own field and be hot linked within your personal library of references. Electronic journals can provide text files of references for the current issue, and for all the references in each article, in a format ready to import into reference managers.

Speed of access is currently a problem for material on the Web. In some parts of the world the Web goes into gridlock in normal working hours. Mirror sites to your country or your institution is an interim solution.

Authors Retain Copyright
 
As an author, you may want to use a figure or use parts of an introduction or discussion verbatim somewhere else. Why on earth should you, the person who created the work, have to go cap-in-hand to the publisher for permission to use it? This practice can stop with electronic publishing.

The Issue of Quality
 
The editors of any proposed high-quality journal face the problem of convincing authors to submit their best work. For a journal on the Web, there is the extra challenge of convincing authors that the journal will be permanent. Here are my suggestions for getting the journal off to the right start:

• High-profile researchers on the advisory board. This obvious strategy reassures authors that their manuscripts will be dealt with effectively, and that the journal has the support of the best people in the business.
• Articles in the first issue by members of the advisory board not only guarantees high quality for the first issue, but also demonstrates that the board's support for the journal is more than just moral support.
• A no-risk draft first issue. All contributors to the first issue preview the final draft. They may withdraw their articles at that time, and the advisory board may cancel publication.
• Tiered levels of excellence within the journal. For example, studies with inadequate sample sizes or flawed design are flagged in their titles as pilot studies, preliminary reports, or letters, and are published as such in an appropriately labeled section of the journal.
• A policy of high originality would reinforce the message that the journal is not simply a research archive, provided that the advisory board can agree the meaning of originality.

What's Next?
 
The team at www.sportsci.org is now moving forward with the plan to add a peer-reviewed journal to the site, along the lines discussed in this article. The journal will be known as Sportscience, and it will specialize in human physical performance.

The current editorial advisory board is recruiting more members. If you are a top researcher in any of the disciplines of sport science, contact one of the existing board members personally via the Contact-Us pages, or email editor=AT=sportsci.org.

Authors should now consider submitting material for the first issue, to be published in January 1998. A full set of instructions for authors will be agreed by the editorial advisory board in the next few weeks. Meanwhile, contact editor=AT=sportsci.org about the suitability of your work.