[Paper Review] Toward Collaborative Information Seeking (CIS)

Chirag Shah|ArXiv.org|Aug 5, 2009

Personal Information Management and User BehaviorDecision Sciences15 references19 citations

TL;DR

This paper proposes a theoretical model for Collaborative Information Seeking (CIS) that distinguishes collaboration from coordination and cooperation, emphasizing shared problem-solving beyond individual contributions. It extends a traditional information seeking model to include collaborative layers across communication, shared interfaces, and collective result management, advocating for user-centric CIS systems over mere extensions of single-user IR tools.

ABSTRACT

It is natural for humans to collaborate while dealing with complex problems. In this article I consider this process of collaboration in the context of information seeking. The study and discussion presented here are driven by two dissatisfactions: (1) the majority of IR systems today do not facilitate collaboration directly, and (2) the concept of collaboration itself is not well-understood. I begin by probing the notion of collaboration and propose a model that helps us understand the requirements for a successful collaboration. A model of a Collaborative Information Seeking (CIS) environment is then rendered based on an extended model of information seeking.

Motivation & Objective

• To formalize the concept of collaboration in information seeking, distinguishing it from coordination and cooperation.
• To address the lack of direct collaboration support in existing information retrieval (IR) systems.
• To develop a conceptual model of Collaborative Information Seeking (CIS) that supports group problem-solving with shared goals and dynamic roles.
• To guide the design and evaluation of CIS systems based on user-centered principles rather than mechanical extensions of single-user IR.

Proposed method

• Proposes a five-layer model of collaboration: communication, contribution, coordination, cooperation, and collaboration, with collaboration as the highest and most integrative level.
• Extends a standard information seeking model (with layers: information, tools, user, results) to include collaborative interactions across users.
• Introduces CIS-specific components such as shared interfaces, shared result spaces, and collaborative result organization mechanisms.
• Emphasizes user autonomy and freedom in collaborative processes, rejecting systems that enforce collaboration invisibly.
• Uses conceptual analysis and comparative definitions from literature (e.g., Denning, Taylor-Powell, Gray) to ground the model in established theory.
• Applies the model to real-world scenarios like political campaigns to illustrate how collaboration leads to solutions beyond individual capabilities.

Experimental results

Research questions

• RQ1 How can collaboration in information seeking be formally distinguished from coordination and cooperation in a way that supports system design?
• RQ2 What structural and functional components are required in a system to enable effective collaborative information seeking?
• RQ3 How can CIS systems be designed to preserve user autonomy while supporting shared information goals?
• RQ4 In what ways does collaboration in information seeking produce outcomes that exceed individual contributions?
• RQ5 What are the limitations of collaboration in information seeking, and when might it be ineffective?

Key findings

• Collaboration in information seeking involves creating solutions that are more than the sum of individual contributions, with shared authority and responsibility.
• True collaboration requires more than shared queries or documents; it demands shared interfaces, communication channels, and collective result management.
• Existing IR systems often fail to support collaboration directly, instead treating it as a secondary or mechanical extension of single-user features.
• The proposed CIS model supports collaboration at three key stages: information request formulation, result retrieval, and result organization.
• Collaboration is not universally beneficial—its effectiveness depends on task type and group dynamics, and not all problems benefit from group involvement.
• The model provides a foundation for evaluating and designing CIS systems with a user-centric approach, avoiding invisible or forced collaboration mechanisms.