Internet Archaeology Archive: Internet Archaeology: Where Next? Paper presented at SAA, April 2000

Introduction

Internet Archaeology (http://intarch.ac.uk) is the world's first fully refereed electronic journal for archaeology. It was founded in 1995, with funding from the UK Electronic Libraries eLib programme. The project is run by a consortium of UK universities working in partnership with the Council for British Archaeology and the British Academy. It is based in the Department of Archaeology at the University of York. Our first issue was published in Autumn 1996 and we are now publishing Issue 8. We have set ourselves the task of publishing articles of a high academic standing which also try to utilise the potential of electronic publication. At the time of writing we have over 23,000 registered readers, but this figure continues to grow at an exponential rate; we publish articles on archaeology anywhere in the world, our readers are drawn from over 120 countries. This paper will review our first four years of operation in order to evaluate the success of the project in establishing a fully electronic Internet journal for archaeology. It will examine reader profiles and explore the further potential of the digital medium. It will outline our plans for turning a research and development project into a self-supporting journal. Issues of copyright, preservation, and user authentication will each be explored in order to analyse the continuing viability of online digital publication.

Content

Although, as a UK-based publication, there has so far been a bias towards the research interests of British archaeologists, our editorial policy is to publish quality articles on world archaeology. As well as covering the British Isles we have published papers on the archaeology of Australia, Hungary, Luxembourg, the Netherlands, Spain, and Senegal. We have forthcoming papers on Beirut, Ecuador, and the Near East. We also publish articles in any language, but if articles are in a foreign language we provide abstracts in English, French and German. The periods currently covered in the journal range from Neanderthals to Historical Archaeology.

We have also published a number of methodological articles, including several discussing aspects of electronic publishing media in archaeology. Over the years, and as we are offered more high quality copy, we have been able to work towards themed special issues, such as a recent issue on e-publication, and a forthcoming issue on visualisation. We have not yet published many full excavation reports, although we have published fieldwork reports with funding from English Heritage, Historic Scotland and York Archaeological Trust. Nonetheless, there is a realisation that electronic dissemination has to be planned into the initial stages of a project and therefore there is an inevitable timelag for such projects to come to fruition. There is also a recognition that full multimedia publication, with "bells" and "whistles" is expensive. It is not that e-publication is more expensive than its print equivalent when like is compared with like. Indeed we have evidence which suggests that if traditional reports are prepared electronically there are considerable savings in distribution costs, and the reports reach a much wider readership. However, when faced with the exciting possibilities of the media authors - and readers - rightly expect value-added, including more colour photos, more access to data, and sophisticated interfaces.

Some 40 refereed papers have so far been published in Internet Archaeology. These exploit a range of multimedia techniques, including on-line click-able maps and time-lines, searchable databases, video, and VRML and QTVR visualisations, as well as large numbers of colour images. Each paper makes extensive use of hypertext links and several of our contributions would, if printed by traditional means, amount to substantial monographs.

Unlike many electronic journals we do not have a print equivalent. Indeed, it would be impossible to create one as much of our content relies on the value-added from on-line presentation. This also means that it is impossible to download much of the content and run it on a local server as it requires specialist software such as database and map servers. This has also prevented us from providing CD-ROM editions as considerable effort would have to go into authoring CD equivalents.

One of the reasons we opted for on-line publication rather than CD-ROM was the ability to provide access to data archives, allowing users to "drill down" from authors' interpretations to supporting data. Another reason was the hope that it would be possible to run dynamic discussion lists alongside the journal, allowing readers to challenge and debate with the authors. In general this has not happened without external orchestrating, and our success stories in this area have had to be carefully "seeded" by the Editor.

A third reason for choosing Internet delivery was the greater ability to reach a large number of readers. At the time of launch many questioned whether Internet access would exclude many readers. As we all know, access to the Internet continues to grow and our access statistics have demonstrated that we are reaching a far wider audience than would have been the case with CD-ROM. Indeed, international availability is one of the most common reasons given by Internet Archaeology authors for why they chose to submit articles to us.

Readership

The number of registered readers continues to grow at an impressive rate. By the end of February 2000 the number exceeded 24,000. Our annual rate of increase is more than 100%. Although our evaluation report indicates that many of these are casual browsers rather than regular readers, we believe there are around 4-5,000 readers who come back to the journal on a regular basis. This compares very favourably with print equivalents. In the year from 1 April 1998 to 31 March 1999 the journal's web pages received over 459,000 successful requests for pages, an increase of some 230% over the previous year. Monthly hit figures are now running at an average of 45,000 per month.

Approximately 25% of readers are drawn from the UK academic sector, with about 5% from the North American .edu domain. The rest are drawn from over 120 countries around the world, from Albania to Zimbabwe. A high proportion of our readership have commercial IP addresses in the .net or .com domains. This is partly made up of contract archaeologists and academics reading us from home, but also reflects the huge popular audience for archaeology on the Internet.

Copyright

One of the fears about making material widely available on the Internet is copyright. To be honest this is not as critical in archaeology as in some other areas. Few archaeological publications are potential major revenue earners. Archaeological authors want their work to get as much exposure as they can and are frequently willing to assign copyright for research and education purposes. What would worry them was if someone were to plagiarise them and steal their intellectual copyright. Internet Archaeology does not claim copyright of content of the papers we publish. This resides with the author or funding body to exploit in as many other ways as they see fit. We do claim copyright over our added value - the multimedia delivery of the product, and there have been a small number of occasions when we have had to ask others to remove material from other web servers where this breached our copyright. Fortunately, we have so far managed to protect our copyright without needing to resort to the courts. However, since what is the real "value-added" rests in searchable databases and similar features that run only on our server, there is really little opportunity for theft.

Quality control

A major challenge in encouraging culture change was the perception that the Internet was a dumping ground and that reputable archaeologists did not publish there. Although we have always insisted that articles are peer-reviewed, and reject a high proportion of the proposals submitted to us, worries were expressed that e-publication would not count towards promotion and tenure. In the UK research publication is geared towards the five-yearly Research Assessment Exercise, or RAE, according to which every academic receives a score from 1 to 5 according to the quality of their best 4 publications in the previous 5 years. These scores are of great importance in determining funding awarded to institutions. We have been extremely fortunate that the review panel for our discipline has been keen to encourage electronic publication and have left contributors in no doubt that their work would count in the RAE, irrespective of the publication medium. In a recent communication with the journal, Professor Michael Fulford, Chair of the Archaeology RAE Panel, has written:

Colleagues are urged to exploit the exciting possibilities of electronic publication. They can be assured that the RAE panel will be concerned with evaluating the content of publications. The medium of publication, whether electronic, or in traditional, printed form will be of no account.

Electronic media have already made a significant and welcome impact on academic publication, though we are still far from exploiting their full potential, and their importance will continue to grow in the immediate future. There were in fact electronic publications submitted to the Classics panel in the last RAE, and these were evaluated alongside, and according to the same criteria, as all other publications. I expect that electronic publications will be a more marked presence in the 2001 RAE. Our interest is in the merit of the research output submitted, not the medium, and electronic publications will be treated by the RAE panel in exactly the same way as those in more traditional media.

Preservation

Another frequently expressed concern is the question of longevity. This is again a cultural issue stemming from the transient nature of content that too often appears fleetingly somewhere on the Internet and then changes or disappears. We have deliberately taken the policy of not changing content once it has been published, even if mistakes are discovered, or if new data would render an interpretation obsolete. Such changes would be dealt with, as with print publication, by subsequent addenda and new editions. The contents of the journal are archived with the Archaeology Data Service (http://ads.ahds.ac.uk), whose remit is the long-term preservation of digital research materials. Thus we have taken all reasonable steps to make sure that our content will be as durable as a hard copy volume. This is not to underestimate the problem. Migration of on-line content offers the best solution but who can predict what browser-dependent plug-ins will still be available in 10 or 100 years time to play some of our multimedia features? Nonetheless these are widespread concerns not limited to archaeology and increasing activity at the level of major national institutions, such as the British Library and National Preservation Office, gives us increasing confidence. Ironically, it is the CD-ROMs, often regarded as preservation formats in themselves, that will probably have the shortest life-span.

Revenue models

We believe, therefore, that the Internet will provide longevity for our content, but in choosing to follow an Internet publication model rather than CD-ROM we sacrificed the ability to follow a traditional marketing model for our journal, whereby we had a physical product which could be exchanged for hard cash in bookshops. For this reason traditional publishers and distributors have tended to favour a CD-ROM model for electronic publication as there is a tangible product which they can sell to customers who at least need to keep it long enough to install it on their computer in order to use it. It is our belief that ultimately CD-ROM publication will be short-lived. The medium is hardware dependent and in academic publishing this raises serious worries over long term preservation. It is also unpopular with academic libraries as it causes major problems for borrowing and access.

However, there are major problems for an academic institution trying to raise revenue from selling Internet content which we are currently having to address. Although the success of the Internet Archaeology project led to renewed funding from the eLib programme, this funding declines until July 2001, when it disappears altogether. We are now, to use the jargon, embarking upon our "exit strategy". Our business plan envisages three sources of income, none of which will be sufficient to support our full running costs on its own.

The first source is publication subventions from traditional researcher sponsors. We already receive publication grants for several papers and as the popularity of electronic delivery increases we anticipate this will grow. However, we still face the problem faced by traditional journal publishers: the subventions generally cover the publication of primary data - the site reports. The synthetic articles which are often of greater interest to an international readership find it more difficult to attract funding.

Advertising revenue can be used to top up the income raised from subventions, but again does not offer the complete solution. As an academic site our use of the national academic computer infrastructure means that we are subject to certain rules. These prohibit us from hosting advertising content, but do allow us to carry sponsorship on a web page. However, we also recognise that sites with large advertising content are unpopular with users and our feeling is that any advertising would need to be discrete and targeted towards our readership - adverts for archaeological books and conferences and so forth. Such revenue from a fairly limited field is unlikely to pay the salary of the Editor on its own.

The third source is that adopted by all print journals, that of charging readers, through subscriptions. For e-journals without print equivalents which can be bundled with the soft-copy subscription this is dangerous and uncharted territory. Our substantial reader-base suggests it is an obvious solution and that with such a large readership we could charge a very low annual fee per user and still cover our costs. However, our research suggests that this is not the case. Supporting a subscription base of 20,000 readers would itself be extremely costly, and could eat up most of the revenue it was designed to capture. Instead it is more cost-effective to aim for a much smaller number of subscribers, each paying a higher fee. Furthermore, our market research suggests that individuals are still not used to paying for Internet content and would be reluctant to take out a subscription (http://intarch.ac.uk/news/evaluation/index.html). If we assume that the 189 readers who responded to our survey questions are representative then only 5% would be prepared to pay an annual licence of £30 and 20% would pay £10. As an academic journal trying to increase the availability of archaeological research results we are also faced by conflicting motives. We do not want to restrict access to our journal. Indeed, many of those bodies who we hope would be providing publication subventions would wish to see the research they sponsored being made available free of charge.

After long deliberation we have adopted a compromise solution for the time being, of charging institutions, but remaining free to individuals. In effect we are means-testing access to the journal, and charging those who we believe can afford to pay. At first site this might seem a rather bizarre approach, but if so we are certainly not alone in adopting it, as it is also a model adopted by a sister eLib-funded publication, Sociological Research Online. It does have a certain logic as so far we have received top-sliced funding from the UK HE sector, and now we are passing the cost on to the institutions who use us most. The solution is also partly borne of expediency. It is much more cost-effective to collect 200 subscriptions at £100 than 2000 at £10. This became particularly apparent as we investigated methods of online subscription collection via credit cards and discovered the high overheads involved and the technical difficulties. We were also under some pressure to provide a means of access which would be popular in UK academic libraries. These were currently favouring a single national database, ATHENS, of user-names and passwords to control access to the range of national electronic resources. Such deals rest upon local site licences and would have run counter to individual user-names imposed by us.

Technically this solution is relatively straightforward to implement. Individual home access will continue to be controlled by readers' user-names and password, registered free of charge on our server. Users from institutions which have not subscribed will be barred from access by their institutional IP address. Those which have taken out a subscription will be allowed in via IP control and/or ATHENS user-names. One of the concerns raised by libraries concerning access to e-journals is the issue of what happens when they cease subscribing. Do they lose access to all the back-issues that they have paid for? This can be characterised as the difference between a licence model (whereby access is licenced for a fixed period of time) and a subscription model (whereby access is bought into perpetuity for a fixed amount of content). In the end we have settled for the approach preferred by the librarians rather than the publishers. If a library ceases to subscribe they will not be given access to new issues, but they will retain access to the back numbers. Of course, this is more complex than the print situation whereby the back-runs are still available on the shelves. Our authentication system must be sensitive to which issues a user is permitted access to, but again this has not proved too difficult to implement.

Conclusion

At the time of writing we are about to embark on a major subscription collection exercise across UK academic and national institutions. If all goes according to schedule from the end of March UK institutional access will be barred, although individuals will still be able to access content via home connections provided by commercial ISPs. Depending on how this phase goes later in 2000 we intend to extend this operation to .edu addresses, with a US price of $180. By the time of the SAA meeting I will provide an update on progress and will welcome discussion of these issues.

 HOME   ABOUT   FOR AUTHORS