Issues in Science and Technology Librarianship	Winter 1998

The Wannabee Culture: Why No-One Does What They Used to Do

Anne Dixon
Electronic Publisher
Institute of Physics Publishing
anne.dixon@ioppublishing.co.uk
http://www.iop.org

Abstract: Electronic publishing has been an agent for change in not just how one publishes but in what one publishes. There is increasing fluidity between what used to be ring-fenced sectors. This paper describes one publisher's attempt to move into another publishing category, and why.

Introduction

In Scientific, Technical and Medical (STM) publishing we see the "wannabee culture" all around us. Authors want to be publishers, secondary publishers want to offer primary material, intermediaries want to offer both, librarians want to expand their role as intermediaries and, in our case, a primary publisher wants to offer secondary, or abstracting and indexing, information.

Cynics would argue that all this activity is a frantic scrabbling to "follow the money" or gain competitive high ground; however, many of us feel that it can be a genuine opportunity to improve the lot of the scientist by linking or improving hitherto disparate resources. In this article I'll describe what Institute of Physics Publishing, where I am the Electronic Publisher, is doing in this context, and why.

Background

Firstly I should describe a little more about my organization, its electronic publishing history, and our reading of this turbulent but exciting business. Institute of Physics Publishing (IoPP) is a not-for-profit Learned Society publisher which is wholly owned by the Institute of Physics. The Institute has some 22,000 members world-wide, primarily research physicists. Our mission is to disseminate information in physics. Physicists are "power-users". They have access to high levels of computing power generally, and are used to working in geographically distributed groups. They have been using electronic communication methods for 20 years or more, and, of course, they invented the world wide web. Publishers serving physicists play a more reactive role than those serving other disciplines: we do not need to persuade end users of the benefits of electronic publishing; indeed they demand that we push the technology envelope.

Accordingly IoPP launched the first ever electronic journal (EJ) in physics in 1994: Classical and Quantum Gravity. We were also the first sizable publisher to put all of our serials output online in January 1996 (see http://www.iop.org). We provided this at no extra cost to the end user or customer. Any institution which subscribes to our print journals gets all electronic manifestations at no extra cost (including 5 years' worth of archive). And this year we introduced HyperCite^(tm). This is a seamless linking system from all references in the 33 IoPP EJs to the INSPEC database, the Los Alamos pre-print database, and back into our own full text. This system works "backwards and forwards in time" so one can link back to the author's stated references, but also go forward in time to see what effect an author's work has had on subsequent literature. As we expand our archive and link to and from more and more publishers, this becomes a hugely valuable tool. And, as is always our policy, it is available at no extra cost to end users or customers.

Behind the HyperCite^(tm) technology lies our concept of "Distributed Publishing". In essence this is a system where any user can find items of interest, irrespective of the publisher. It is based on seamless publisher linking and is independent of one's starting point on a particular "information trail". We endorse this concept because no content provider can hope to be fully comprehensive in terms of coverage, or be able to anticipate the vagaries of an individual's information trail. End users need breadth of content and have no interest in who publishes what, or, largely, how the information gets to them.

The INSPEC Database

HyperCite^(tm) was a joint project with the Institution of Electrical Engineers (IEE), who own INSPEC. It was only with IEE's tremendous support that we were able to create such a service. As part of the HyperCite^(tm) project we had to learn how to link into and out of the INSPEC database. This database, the physical manifestation of a substantial abstracting and indexing operation, had, at the time of loading, some 5.6M records and it covers most serials output in physics, engineering and computer science. The database covers the period from 1969 to the present day and has some 69 fields within a record, many of which are themselves subdivided. Although IoPP's content in INSPEC is relatively small, we had to make sure all the references would link - a formidable undertaking, as anyone who has studied article references will know. It was this, ultimately positive, experience which started us thinking that we could be capable of offering the INSPEC database itself to the world.

The Research Phase

However convinced we may have been of our technical abilities, we still had to establish whether there was a market need. Accordingly in Spring 1997 we sent an e-mail questionnaire to all of the then-registered EJ customers - some 1200 at the time. These customers are almost exclusively librarians. They are not wholly representative in that, because they have registered for EJs, they are presumably already well-disposed towards things electronic. A creditable 386 (32%) responded. The vast majority of responders - 89% - already had access to the INSPEC database in some shape or form. Two big surprises arrived with their responses. The first was that a whopping 81% thought that it would be of benefit to them to have full access to the INSPEC database via our web site (however nearly half of these had reservations about price, speed of access and quality of search interface and search engines). The second was that, within the 83 densely packed pages of unsolicited comments, there was a strong demand for links from the INSPEC database to our and other publishers' full text. This was very encouraging. But we had two further concerns: firstly that we couldn't offer cross-database searching a la FirstSearch (as long as we only offered the INSPEC database); and secondly that we would probably have to build our own interface and command language (due to potential "passing-off" legal problems if we were to emulate existing systems). This, of course, would mean that the librarian and end user would have to learn yet another interface. So, to establish whether these concerns were a barrier to success we asked our European and North American Library Advisory Councils questions on these two subjects. Our Library Advisory Councils are comprised of senior librarians who are knowledgeable in matters of electronic publishing. Answers to the first question, about cross-database searching, were mixed. There are only a few services which actually offer this right now and responses, understandably, tended to be based on whether the responder did or didn't have this type of service. The overall theme seemed to be that it was desirable, but not essential. And of course IoPP could, theoretically, offer some of the other databases which pertain to physics in the future. On the second question, about a new interface, we were again surprised. The following comment sums up the responses: "We have to learn new interfaces all the time. Just make sure yours is as good or better than the rest!"

What to Offer and How Much?

By now we were getting to crunch-time. We wanted to do this, and our customers seemed to want it. But what exactly to offer and how much should we charge for it? Purchasers of the INSPEC database will know that there are many permutations on offer related to format, number of users, depth of years covered, and other factors. Our decision to start with an offer of an unlimited site license for academic sites was based on three main criteria:

Our mission is to disseminate information in physics. We want as many people as possible to have access to physics information. Thus with our EJs, for example, we place no restrictions whatever on usage within a subscribing site, and have no enforced barriers, such as user IDs and passwords, to usage.
For INSPEC to link into our own full text and to the HyperCite system (and, hopefully, to other publishers' full text), which is what the customers have indicated they want, it is most suitable if the offering is the whole of the database, not just a subsection of it. Thus there can be the maximum number of links possible. And we know, from previous research, that breadth of content is important to end users and customers.
Most of our existing customers are academic sites, so it makes sense to start with them. However there are comparable packages for government and industrial sites; due to the INSPEC pricing mechanism they are, however, a little more complex, being based on number of users.

The flip side of all this is that, by offering the widest possible usage, we are obliged to pay INSPEC a substantial price - $43,500 per site per year. All INSPEC prices are available from their web site (http://www.iee.org/). [Note: Link moved; URL changed 3/28/02 by ald]

We therefore decided to work with very low margins so that our pricing would be in the region of 12-16% less than any comparable offering. Accordingly our price for 1998 is $45,000. We also decided to publicize our prices to the world. Remarkably, in 1998 we may well be the only re-seller of the INSPEC database to do so. We hear from acquisition librarians that it is increasingly common for content providers or intermediaries to use contract law to persuade purchasers of services to keep terms of purchase or licensing confidential. This does not seem to be in the spirit of what is best for the community.

Finally, we wanted the offering to be simple. Our experience with EJs has shown that this, along with brief and succinct contracts or registration forms, is much appreciated.

What Next?

Our current task is to refine the interfaces we are building to the INSPEC database. Accordingly we are compiling a list of volunteer beta test sites who will help us to get it right. Once we are content that this is the case we shall be offering 60-day trials of the service to all interested customers. We appreciate that changing vendor, changing type of offering bought, or deciding to take on a new database is a very big step and that the customer must feel justified in their decision. Furthermore the situation may be complicated by the decision to do so within an existing or new consortium. If you would like to become one of our planning partners, and take part in beta tests or free trials, do please let me know.

We welcome your comments about this article. Please fill out this form for possible inclusion in a future issue.