Uses Netscape HTML extensions


LIBLABLIBLABLIBLAB

The Laboratory for Library and Information Science
Department of Computer and Information Science
Linköping University
Sweden


Menu Page | Current Research | Publications | Undergraduate and Graduate Courses | Lab Members

Originally published as:
LiU-LIBLAB-R-88:2
February 1988

Research Program for LIBLAB


Björklund, L.; Hjerppe, R.; Jägerfeld, M.; Olander, B.; Vainio-Larsson, A.

1. Introduction

The research program for LIBLAB that is presented below replaces that which was proposed in the report "A Program for LIBLAB, the Library Research Laboratory at Linköping University" (LiU- LIBLAB-R-1982:3). The direction of activities that have been established within LIBLAB, as well as developments in its environment, justify a new program. The research program has been deve loped jointly by LIBLAB's staff members.

LIBLAB, the Library and Information Science Laboratory, is one of ten research laboratories in the Department of Computer and Information Science, Linköping University. LIBLAB's activities have been funded since 1983 by the Delegationen för vetenskaplig och teknisk informationsför sörjning (DFI) (Delegation for Scientific and Technical Information). As a result of parliament's decision to abolish DFI as of 1 July 1988, the responsibility for funding library research has been transferred to Forskningsrådsnämnden (FRN) (Research Advisory Board).

In fall 1987 a review of LIBLAB and its activities to that point was carried out in cooperation with DFI. Among the recommendations made in this evaluation were the following:

2. Goals for LIBLAB's activities

The overarching goal for LIBLAB is to pursue long-term studies with particular emphasis on the possibilities of information technology and its effects on library and information work.

LIBLAB strives to build up a broad competence in library and information science, and the appli cation of computer science and information technology. This competence development has as its aim that LIBLAB should be able to actively influence and anticipate applications of new techno logy to the field of information provision, including libraries.

Exploratory system development and construction of theoretical models for bibliographic informa tion handling are important means for LIBLAB to develop knowledge.

With LIBLAB's connection to the Department of Computer and Information Science come unique possibilities to productively exploit the interaction between library and information science and computer science. The formal parts of computer science can for example contribute to theory buil ding for descriptions of documents and document collections as well as thesaurus construction. Re search in artificial intelligence contains problems and application areas that lie close to library operations and information provision.

Proximity to the Linköping University Library gives LIBLAB the possibility to follow and take part in debates and developments in the library world "from inside". Above all LIBLAB's interests are the new media, new forms of publication, and new methods of scientific communication and knowledge organization and their interaction with library functions.

3. Research Program for LIBLAB

The new research program is directed toward two main themes:

Both of these themes are focused and interwoven in the long-term project HYPERCATalog that aims to build and study the uses of a catalog that in many respects is radically different from present catalogs. The center of attention is the catalog, both as a tool for users and as a description and re presentation of the library's collections.

For each theme there are some subthemes within which projects are being pursued or are planned. These subthemes are instantiations of the overarching theme.

3.1 Theme 1: Users and information systems, especially bibliographic systems

This theme is directed toward the use and users of interactive information systems. In the center of LIBLAB's attention are information systems for bibliographic information. Information handling and systems per se are not of primary interest within this theme, but rather the relation to users is emphasized.

The research theme is formulated in general terms and thus is relevant across a broad spectrum of applications. The perspective is user-oriented. The user's central role as producer and consumer of information services is emphasized. Ideally implied in this is that the operational definitions and criteria which the system builder (designer) uses as a foundation must be consistent with the de mands that users put on the system. The system builder's task is to supply the technical realization of these demands. In practice this still means many times that incompatible demands must be ba lanced against each other and that the system builder as well as the user must find methods to for mulate, specify, and rank order these demands.

Design, use and evaluation of the system are three central research areas within this theme. These three areas together with the above-named user-centered perspective give three aspects from which systems should be studied:

The basis for the research theme is the fact that each information system (manual or computerized) is developed or has grown with plans for its functions (use), users and the interaction between these components. The actual use of a system often constitutes the largest investment in it. To utilize and take advantage of this investment therefore becomes an important condition for the adaptation of a system to its users.

These factors are fundamental to all system development as well as to the use of the system and recur even in an evaluation, irrespective of whether the goal is to control the system design process or whether it is a (presumptive) user who makes a judgment of a completed system. The methods for such an evaluation are also part of this theme.

3.1.1 Research areas within the theme "Users and information systems, ..."

Current projects include two doctoral dissertation projects in area 1.1.

Research Area 1.1: User participation and user behavior
Strategies for personal information management

One doctoral project is a study of how computer science researchers manage scholarly information for personal use. Through case studies the investigation tries to elucidate what information the re searchers think they need, how and from which sources they gather the information, how they or ganize their collections, and how they use the information. Scientific work demands and produces scholarly information, which ordinarily is paperbound.

The information services which researchers up to now were offered have in general provided an information specialist who searches for information for the client/researcher's review. The techni cal development within the computer area is moving increasingly towards end user systems, per sonal computers and personal workstations. Against this background it appears inevitable that information services must be adapted to the primary users, so that researchers themselves become information specialists. A survey of particular researchers' methods of managing "their" scholarly information is needed for such an adaptation.

The researchers in the study use electronic mail to a great extent for contacts with colleagues th roughout the world, and they are strongly dependent on computer power for producing scientific documents. The study will also try to elucidate what significance the medium has for scientific ac tivities.

Conceptual system models and user interface design

The design of a system's user interface forms an all important part in all system development. For many users the system's user interface constitutes the system itself. Modeling a system at a con ceptual level and then letting this model form a central and guiding basis for the continued design become quite necessary.

The project focuses on two central parts in such a design process:

Research Area 1.2: Orientation in the database: Maps and other tools

A new user is often perplexed within large and complex databases. A common reaction is: What is found in the database? How do I find what I think I want? More experienced users may have other questions: What have I done so far? Where have I been? What is left that I have not seen (but perhaps should see)?

In the discussions concerning the so-called hypermedia*, questions concerning orientation ability in abstract space and risks of getting lost have been brought up as potential problems. In connection with the HYPERCATalog project, it is therefore important to study more closely different possi bilities for graphically illustrating database structure and content. Generation and representation of maps and similar presentations, that can be updated dynamically in order to show both the "im mediate environment" and the global view, are therefore of interest. Classification systems and the sauri are descriptions of subject areas that can form a basis for illustrations. Co-citation graphs show the structures within particular literatures.

3.2 Theme 2: Document description and representation

This theme deals with descriptions and representations of documents and collections and their re lations at different levels. In this context "document" primarily means published texts, i.e., publi cations. The approach is broad, however, and description issues are common to all kinds of documents, such as notes, maps, sound and video recordings, computer files, etc.

Essentially, a document may be described from three different perspectives:

Each of these descriptions may vary in completeness, from a summary to one that can be used to reconstruct the document. The descriptions may also vary from being strictly formalized to ad-hoc ones. The connection between collection(s) and single documents are of vital importance because each individual document of interest can be part of several different collections and collections are built up from different documents.

The most interesting descriptions and representations are those used in catalogs of various kinds. Two central research areas have been identified for catalogs that are tools for knowledge organi zation and management:

For catalogs as well as collections, it is necessary to distinguish between public, internal, and pri vate views.

The focus of interest for LIBLAB is document descriptions and representations, handled by com puterized catalogs. An unbiased study of the possibilities and limitations of such a medium, and subsequent experimentation with different forms of such catalogs are central to this research the me.

Consequently, fundamental issues in description, representation, computer architecture, and data base structure as well as formats for exchange and processing of bibliographic records are included in this theme.

New media, document architectures, editors and other tools for creation and working with docu ments are also of interest in this theme, since new document forms are being managed by libraries, and since they provide examples of new description and representation forms. Above all they give clear suggestions of the possibilities of letting users use computerized catalogs and information systems in the same immediate way that they use literature and libraries, often unplanned and fol lowing interesting sidetracks. In the long term project HYPERCATalog the ambition is to work with applications for library and information services.

3.2.1 Research areas within the theme "Document description and representation"

Research Area 2.1: Description and representation of collections and elements

In this research area the fundamental description and representation questions are central. A uni form and adequate description method for collections is lacking today. Which description catego ries are necessary, sufficient, and relevant for a collection of documents depends partly on the collection's environment -- if it is isolated or forms part of other collections or is composed of sub collections -- and partly on how the document is described.

The description of a document should in part be made with thought of how collections can be de scribed. Up to now documents have been regarded as static and therefore considered to need a detailed description only once. Collections are on the contrary dynamic, they change, and the des cription should therefore be modified. In a HYPERCATalog perspective it is intended that the de scription of a particular document will be updated when new information has been added. How often this should be done, and in what way, is an example of the issues within this research area.

Research Area 2.2: The structures of collections, literatures, and subject areas

The point of departure for this research area is the distinction among three different kinds of struc tures in document collections: the structure of subject areas, as reflected, for example, in thesauri; the structure of literatures as can be generated, for example, from citation networks; and collection structures derived, for example, from the physical location of documents. The generation of struc tures and the connection between them is part of this research area, while the visualization of them is part of the research area on orientation in the database within the first theme.

Research Area 2.3: Hypertext and hypermedia

In this research area hypertext and hypermedia are of general interest. The characteristics of exis ting and proposed systems, the experience in different applications and the connection with docu ment architecture are some of the interests that this ressearch area includes. Research area 1.1 has as an important basis also the study of hypertext and hypermedia as a general metaphor and inte raction model for advanced interactive systems.

3.3 Project HYPERCATalog

HYPERCATalog is a vision, a metaphor for a complex yet flexible way of giving users access to the information and the structures that are more or less implicit in document collections.

Experiences and insights from the work within the two research themes are combined in the HYPERCATalog project. While the HYPERCATalog serves as a testbed for theories and ideas ge nerated within the research themes, it also functions as a generator of problems and research ques tions. The long-term goal is to build and implement a HYPERCATalog. Over the next few years, research efforts will be focused on building small prototypes, HYPERKITtens, in order to gain ex perience with various subproblems and to test different possibilities for solutions. For the most part existing software, such as GUIDE and HyperCard for the Macintosh and NoteCards and LINCKS for the Xerox 1186, will form the basis for construction of these HYPERKITtens.

*Hypertext is nonsequential text that branches and offers the reader and author alternatives. Hy pertext and its generalization hypermedia should be managed by an interactive system.


Originally published 1996-04-22, last modified 1996-04-22, by Roland Hjerppe, rolhj@ida.liu.se

Liblab, Dept. of Computer and Information Science, Linköping University, Sweden