The information lens: an intelligent system for information sharing in organizations

This paper describes an intelligent system to help people share and filter information communicated by computer-based messaging systems. The system exploits concepts from artificial intelligence such as frames, production rules, and inheritance networks, but it avoids the unsolved problems of natural language understanding by providing users with a rich set of semi-structured message templates. A consistent set of “direct manipulation” editors simplifies the use of the system by individuals, and an incremental enhancement path simplifies the adoption of the system by groups.

One of the key problems that arises when any group of people cooperates to solve problems or make decisions is how to share information. Thus one of the central goals of designing good “organizational interfaces” (Malone, 1985) should be to help people share information in groups and organizations. In this paper, we will describe a prototype system, called the Information Lens, that focuses on one aspect of this problem: how to help people share the many diverse kinds of qualitative information that are communicated via electronic messaging systems.

It is already a common experience in mature computer-based messaging communities for people to feel flooded with large quantities of electronic “junk mail” (Denning, 1982; Palme, 1984; Wilson, 1984; Hiltz & Turoff, 1985), and the widespread availability of inexpensive communication capability has the potential to overwhelm people with even more messages that are of little or no value to them. At the same time, it is also a common experience for people to be ignorant of facts that would facilitate their work and that are known elsewhere in their organization. The system we will describe helps solve both these problems: it helps people filter, sort, and prioritize messages that are already addressed to them, and it also helps them find useful messages they would not otherwise have received.

The most common previous approach to structuring information sharing in electronic messaging environments is to let users implicitly specify their general areas of interest by associating themselves with centralized distribution lists or conference topics related to particular subjects (e.g., Hiltz & Turoff, 1978). Since these methods of disseminating information are often targeted for relatively large audiences, however, it is usually impossible for all the information distributed to be of interest to all recipients.

The Information Lens system uses much more detailed representations of message contents and receivers' interests to provide more sophisticated filtering possibilities. One of the key ideas behind this system is that many of the unsolved problems of natural language understanding can be avoided by using semi-structured templates (or frames) for different types of messages. These templates are used by the senders of messages to facilitate composing messages in the first place. Then, the same templates are used by the receivers of messages to facilitate constructing a set of rules to be used for filtering and categorizing messages of different types.