The Internet radically impacts end user expectations of secondary and primary information sources

Gert Staal, Director, SilverLinker Business, SilverPlatter Information B. V., Amsterdam, Niederlande

Consequences and Lessons to be Learned

Abstract: The author has identified friction between the user’s expectations and the librarian’s perspective, as well as friction between user expectations and market realities to be some of the underlying causes for the relative poor penetration of electronic information on the desktop. Today the author will bring forward some ideas that are central to our company, and the way forward we see in unlocking scientific information resources on the Internet.

The User Expectation

Based on user’s experiences with Internet usage, the end user will have certain expectations when accessing primary and secondary scientific information resources. Let me add here by saying that I think there is no one typical user of Internet resources, but that each user will share, at least to some degree, the following requirements about information.

• information should be current
• information should be ‘free’, i.e.: if not ‘free’ in the sense of not-gratis, then at least freely accessible without transactional charges and/or password protection
• information should be accurate/reliable
• information should be integrated: the end user, provided the information is accurate and reliable, is not really interested in single information provider’s offerings, but in a combination of all possible resources.

Of course these expectations are fuelled by the current functionalities, or the conceived/extrapolated functionalities of Internet in the future, based on what is known today.

Why the User Perspective?

Users will, in our opinion, increasingly dominate the discussion on the requirements of information services. There are several very good reasons for that. In the first place, very obvious but forgotten mostly, because end users are the sole aim why information resources are there: essentially primary and secondary information is put on the Web for the end user, who uses the Web for a score of other purposes too!

Secondly, the average end user has used the Internet for exploring information resources for a longer time than publishers have put their resources there. This gives end users a vastly better position to comment on how information should be represented. Yet much of the discussions, even today, seem to focus on new functionalities of software packages, rather than address underlying end user and librarian’s behaviour.

End User Expectations and Market Realities

The end user will typically surf sites for practical information, and will view the Internet as an integrated ‘source of knowledge’. His/her preference, for scientific materials, would be to access a single system, and hope to find all their resources indexed there, in one fell swoop. Now that is a panacea that is still around the corner. I’ll explain why.

First of all, many print publishers have vested their money in developing one site for their electronic materials only, thereby making one of the greatest end-user centred functionalities, namely integration with other materials through hyperlinking, difficult on a large scale using traditional means.

Furthermore, there is no interoperability between the secondary and primary information business model. Typically these come from different sources altogether. Even if that is not the case, then mostly the combined secondary and primary resources offered by one publisher are normally just a subset of the total offerings in a given field.

Thirdly, the issue of scaleability is not addressed. Even though prices of hardware (in particular disk space and memory) are dropping fast, it is inconceivable that there will be a new Electronic Library of Alexandria soon, one that contains all and will contain all for centuries to come. Sheer growth of the scientific body of scientific literature alone would prevent that. In stead of building a library of Alexandria, perhaps we should concentrate on building its ‘wings’. A system that is truly scaleable therefore should be able to ‘point’ both to local information as well as remote holdings.

However, there may be a desire to maintain some diversity in terms of information systems used, also from a user perspective.

The Librarian’s Perspective

Of course, the librarian has recognised all of this, but there are more things to information than the user perspective. A librarian is better trained and equipped to judge the value of competing products. It is here that the disparity begins: the end users endorses a product by either using it or leaving it alone, but the librarian still has a final say in most transactions, whether advisory, endorsing, purchasing or otherwise. At least he or she will normally pay for services and be responsible for licensing in an academic environment, and to a large extent in the corporate environment. Therefore, the librarian is faced with a task that is sometimes at odds with user’s interests, namely

• to purchase (users want it all, librarian has a limited budget)
• to manage access rights to the user community (user wants full and unlimited access, librarian has to be a gatekeeper of access rights to respect license agreements). This surfaces e.g. in the licensing and purchase of electronic journals, as well as in the purchase and offering of online ordering to document delivery/ILL systems
• maintaining/safeguarding the system (user wants 24-hour access to a free system, librarian has to keep the system in the air financially and technically)

Balancing

In order to develop a successful product, it is absolutely vital to keep in mind that it will be necessary to meet both users’ demands and librarians’ needs, even though that may imply that almost contradictory demands will have to be married: electronic content access control is to be matched with openness/public availability of information, systems access control with interoperability of systems.

Many Internet primary/secondary sources have been moderately successful, in an overwhelming number of cases they at least were performing well below expectations. It is now time that we start, in a generic fashion, analysing factors that play a role in their success or failure.

The Internet as a delivery medium of full text information has not particularly been hassle-free so far. In fact, one could almost say it has a reputation for not being so.

Some factors are generic:

• there is the general issue of network (i.e. technical) speed: every user knows that in Europe the Internet has a main cardiac arrest after 2pm, especially when accessing northern American sites. The hunger for bandwidth is so great, that the growth of new bandwidth cannot keep up with the growth in demand. Unless some serious alternative and parallel networks scale up soon, the Internet will remain a questionable and somewhat unreliable delivery mechanism. The need for bandwidth is most obvious in a full-text environment. A typical scientific article in PDF format will vary between 0.5 and 1 MB. Now that in itself is not a large file, but it is when faced with a community of 20,000 students for a single university, not a small portion of which may be online simultaneously.
• furthermore, there is the issue of pre-print: not so much vanity in the name of science has driven many authors to sending their articles to pre-print servers, but the need to increase the publication speed of scientific work, relating to outside pressures on people. This seems to be particularly true in quickly developing areas of science: high-energy physics, astronomy, etc. Many services in this area however are still quite under-developed even if the potential is there. More: end users are aware of this unleashed potential which is proven by the mere fact that these start-up services exist parallel to publication programmes.
• as with publishers (see above) whether primary or secondary, have not been very successful in marketing format parallel with content. Many have tried- and erred, some have set up services the end users may not be waiting for.

Some factors are user-related:

• users have not been accustomed to purchasing information online. They haven’t been used to paying for information whatsoever. Therefore the economic models that were applied in some instances have failed.
• users have found that there is simply not enough critical mass to go on: approximately 2 to 3 percent of current scientific information is now produced in a reproducible, archivable electronic format.
• the complexity of the offerings: novice users find it extremely difficult to navigate to the right information, given their relative innocence in certain subject areas even in the printed environment

Some factors are librarian-related:

• librarians usually disliked being tied to individual publisher’s systems, and usually have had a favour for integrating access through third-party aggregators like e.g. Swets or EBSCO, but could only find part of the information really needed there
• there is the issue of evolving standards: XML versus HTML?, DOI or SICI? - battles that will last for some time. Battles also, that have not been fought yet. New campaigns are on the horizon.
• Librarians have not liked the economic models that were applied in some instances, bypassing the librarian completely, which has not stimulated their acceptance.
• Librarians have been cautious about direct and electronic access to ILL & document delivery services to certain groups of users, since no good management system existed to support it.
• librarians have had concerns about archiving materials, related to capacity, and to the stability of formats and platforms, as well as copyright

Partly due to a combination of causes mentioned above, varying from site to site, users have hardly any fulltext access (if available at all), whereas users are exploiting Internet for other purposes, as I refer to the use of Internet as a source of practical information, for email purposes, for ftp, shopping etc.

So we are faced with a paradoxical situation. Librarians and end users recognise that the introduction of real-life up-to-date, full content of scientific information on the Web is a milestone. Yet this information is to date only delivered in exceptional situations in the desired form!

The historical background should be taken into account. We all know that electronic access to full text promises the kind of paradigm shift we should compare to the invention of print. For instance, we all know that the copyright problems we now face equal the historical copyright problems that surfaced when the printing press was introduced. Drawing historical parallels, now that we are at it, reveals to me that electronic publishing is probably in its first stages of its evolution.

There is no binding factor, so it seems. No connection yet between user’s needs and what they are been served. No proper connections are yet made in many of the economic models, the business models, and the technology platforms, although this will change over time.

Centuries of experience in print only has a limited value in the electronic age. SilverPlatter has been in electronic publishing probably longer than most publishers have been, and we think we have some insight into how and what user’s needs are, and at least we have a view on what these models and platforms are. Although the issues are different for secondary and primary information in many respects we learned that what the market is waiting for are knowledgebases containing both. Not just one central monolithic repository that unites materials from all kinds of sources, in all sorts of formats, but diversified, de-centralised repositories. It is the links that matter, the integration that counts. With SilverLinker we will build a platform , that will balance the needs of the end-user and the librarian alike, a step on what we think is the right evolutionary path.