In parallel with the issues identified in Section 2 (the information suppliers' viewpoint), the issue of how people go about finding and using information must also be considered (the consumer viewpoint).
Before the widespread use of the Internet and Web, researchers were required, after possibly searching paper abstracts and indexes, to formally request information from repositories such as academic and commercial libraries. The information tended to be specialised, well-organised and relatively straightforward to locate e.g. patents, journal articles. However, unless they were subscribers, users had to wait (for a relatively long interval by today's standards) between the time that a piece of information was requested and the time it was generally available.
The information age, of course, has radically changing this approach and now masses of information is loaded daily onto web servers that is usually instantly available to mostly anyone who wants it. To understand today's requirements and those of the near future, we need to consider the different quality of service levels experienced by today's users and how they can be enhanced in the future.
With these points in mind, the following sections look in more depth at the ways users access or have information delivered.
Users search for information because it is thought to exist, e.g. 'hunting and gathering'. This demands a degree of knowledge of the task, the information that exists, and skills in searching or using search tools.
Information is often automatically delivered - not necessarily on-demand because presumably someone thinks a user needs it. These fall into three types of service:
• subscription services - a user purposefully subscribes to a one-off (e.g. pay per view);
• or periodic service (e.g. journal subscription, list server, Net-minderfn 2);
• channel services where the user selects from a number of continuous streams of information that the user selects also called 'push technology').
This is often termed the 'serendipity effect' since it involves discovering interesting information perhaps while browsing for something else.
TAP deliverables are special information types in that they are the results of EC funded project work in particular application areas. Traditionally they include reports, tutorials, pilots, demonstrators, publicity information etc. but increasingly will include other media types, professional indexing as they grow in size and become more interactive. See for example the ESPRIT Showcase and The Handle SystemÒ sites [9, 10].
It seems intuitive that few people have a systematic approach to finding information, and consequently any dissemination method that relies solely on people searching out information for themselves will not be wholly effective.
In addition to this, the ongoing digitisation of all types of media and the easier access to information sites are creating new problems for end-users in terms of navigation and the difficulty of keeping pace with information supply [3].
As can be seen from previous examples, novel ways of presenting information in an interactive way are beginning to be seen as the norm in Web navigation. As a result providers need to be aware of the most effective ways of disseminating information simply to keep up with the expectations of users.
The most important observation is that as more information is provided on the Web, so in turn it becomes more necessary, and correspondingly difficult, for providers to be able to offer a 'different' product. For example, in the case of the PROSOMA showcase [9], the idea of adding interviews with project experts provides the all important 'edge' feature that distinguishes the site from other similar competitor sites and leverages one of the key resource of the programme.