Workshop Organisers

¹ Computing Department, Lancaster University, Lancaster, LA1 4YR, UK

² Sociology Department, Lancaster University, Lancaster, LA1 4YR, UK

³ Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign, IL, USA

All correspondence to be addressed to the first author (mbt@comp.lancs.ac.uk)

Aims

• To consider the details of actual and potential kinds of collaboration in digital libraries
• To assess and compare studies of collaborative activity in information searching
• To attempt to spot emerging themes from the different studies
• To consider recommendations for improving collaboration in digital libraries

Intended audience

• People who have performed studies which reveal information about collaboration (or the lack of it, or barriers to it) in information searching
• People who have developed systems that provide functionality for collaboration in digital libraries
• Developers of digital libraries wishing to provide functionality to support such collaboration

Estimate of numbers

We would hope for an audience of approx. 25

Details of the Workshop

We wish to encourage a more detailed consideration of the importance of collaboration in a digital library. We believe that it will be a powerful means of improving the effectiveness of digital libraries for a wide range of users. As librarians have noted for some time, physical libraries are places that have a distinct social dimension.
Examples include:

• Individual users interacting with individual librarians. Interactions include short conversations like those that occur at a help or information desk as well as the much studied reference interview.
• Library staff collaborating; with each other; co-working on a common task, having an awareness of the activities of colleagues and contributing solutions; sharing hints, tips and expertise in tackling a tricky search request of a client; teaching and learning from each other, particularly as new databases and information systems flood in, and the interfaces to existing systems are continually updated.
• Users interacting with each other. Libraries are public spaces. People arrange to meet there, bump into acquaintances and colleagues and can learn from each other by observing how others use the library's resources, or by asking them for help. Users may choose to learn how to use systems by asking friends and colleagues for help instead of approaching librarians. With the increasing trend towards group projects as part of learning, teams may meet and collectively search for and discuss findings in the library, or remotely.

Some new library buildings (including those of Limerick University, the University of Southern California Leavey Library and the Grainger Engineering Library at the University of Illinois) have explicitly acknowledged the need to more actively support collaborative use of library resources by providing meeting rooms and generous space around terminals to make collaborative working using both physical and electronic resources easier.

Our concern is that much of the development of digital library systems will proceed with the implicit assumption that the typical user will be the classic stereotype of the lonely scholar, resenting the need to share resources with other users and dreaming of the pristine virtual library where all other users are as far as possible rendered invisible. Implicit assumptions about "typical" users can have unfortunate consequences for usefulness, usability and acceptability when designing systems for a broad range of people.

We want to emphasise the advantages that can accrue from collaboration in helping users efficiently to find out what they want. It should be noted that many of the collaboration examples from a physical library outlined earlier occurred despite rather than because of the underlying systems. It would seem likely that even more effective collaborations could occur if the systems acknowledged the existence of collaboration and even actively supported it.

The danger is that in a digital library, where users may be accessing information remotely from their homes and offices, the collaborative interactions that occur in a physical library will be lost. This is unfortunate as the digital library offers the opportunity not just to maintain such useful collaborations but also to support quite new kinds of collaboration, where participants are unconstrained by factors of distance and time, or indeed even the need to know who they are collaborating with.

The workshop will consider some of the following issues, as well as others contributed by people attending:

• Supporting collaborative learning using the digital library
• Supporting collaborative learning about how to use the digital library
• Information filtering: taking advantage of the recommendations of others
• Taking advantage of the usage behaviour of others
• Designing systems that support collaborative interactions
• Supporting help giving about and using electronic systems, both remotely and face to face
• Supporting the incremental learning of more advanced search skills
• Determining a research agenda for collaboration in the digital library.

This involves gaining a better understanding of:

• How collaboration operates both in existing digital libraries and in physical libraries.
• How people find information in their offices and workplaces, away from both physical and digital libraries, and how people move between working non-collaboratively and collaboratively, including using their colleagues and co-workers as primary information sources.
• How collaboration operates in different academic contexts (for established researchers, for graduate students, for undergraduates and for schoolchildren)
• How collaboration operates in information searching in other contexts (commerce, professional development, the law, consumer purchasing, health, etc.)

From an improved understanding we can look at ways in which computer systems can be built to improve collaborative activity. This can involve both collaborations that currently exist and those (such as remote help-giving) that are impossible or impractical in traditional contexts.

Proposed activities

Prior to the workshop

Participants are invited to submit a short position paper outlining their particular interests within the theme of collaboration and digital libraries.

Those participants who have undertaken studies of activity that involve collaborations are requested to include references to this work, and/or drafts of, say, internal reports or protocols that they are willing to share with the other participants.

All participants are asked to submit a wild generalisation, rule of thumb, or abstraction about collaboration and how it is or might be applied in the Digital Library. These generalisations will be then tested in the workshop against the findings of the studies undertaken by participants and the published results of other studies. An example of such a wild generalisation might be: "Users never go on formal courses, they only ever learn from their friends". You don't necessarily have to believe in your generalisation: this is a hypothesis-testing exercise and provocative extreme statements can serve as a better starting point for refining our understanding.

At the workshop

Brief presentations of studies including but not limited to ethnographic studies which reveal aspects of collaboration

Testing of the wild generalisations across studies. An attempt to spot similarities and account for differences in findings. An attempt to refine the generalisation if possible to fit all the available evidence. An attempt to come up with alternative competing hypotheses that also fit the evidence.

An attempt to spot emerging themes. We wish to see whether equivalent patterns of behaviour will emerge from the results of different studies by different researcher using different methods in different contexts. If so, we can be much more confident about the general significance of such behaviour patterns. An example would be similarities between the interactions at a help desk in a library and those in retail settings. We also wish to consider exceptions to such general behaviours.

A discussion of how an understanding of these behaviour patterns can be used in the design of more effective functionality that can actively support or enhance their effect or widen their applicability.

Brief presentations from people who have written systems that support collaboration in information searching.

Discussion of where to focus future studies to address the most important gaps in our knowledge as identified by the workshop.

Organiser backgrounds

More details of our work can be obtained from the Ariadne project web page:

http://www.comp.lancs.ac.uk/computing/research/cseg/projects/ariadne/

Michael Twidale is a lecturer in the Computing Department at Lancaster University, and a member of the CSCW Research Centre. He is currently working on Digital Libraries, particularly their collaborative aspects for both working and learning and the implications for designing appropriate functionality and interfaces. He has presented this work at CAL95 (UK computers and learning conference), Computer Supported Cooperative Learning CSCL95, ELVIRA 95 (the UK Digital Libraries conference) and the last two Allerton institutes on digital libraries run by the Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign.

David Nichols is a post-doctoral research associate in the Computing Department at Lancaster University and a member of the CSCW Research Centre. His recent research has been in the area of collaborative browsing - examining social interaction in Digital Libraries. He is currently working on the British Library Research and Innovation Centre project Implications of collaborative browsing for library systems design. His work has been presented at CAL96, CSCL95 and ELVIRA95.

Jon O'Brien is also a member of the CSCW Research Centre at Lancaster University. He has worked on a number of projects which have involved the use of ethnographic studies to inform the systems design process. A recent project is a British Library funded ethnographic study of library use, currently focusing on interactions at the help desk. His work has been presented at CSCW96, CHI96, RE95 and IRIS19

Bob Sandusky is a Ph.D. student in Library and Information Science at the University of Illinois at Urbana-Champaign. His research focuses on the collaborative nature of the work of managing large distributed information systems like Digital Libraries and telecommunications networks. This includes the integration of digital libraries with existing libraries and communities of practice, and issues of user education, including applying MOO features to support learning and getting help.

Last revision: 7th March 1997