Berrypicking

Berry-picking is a model of information seeking and retrieval that describes how people gather information in real-world tasks. Coined by Marcia Bates in 1989, it portrays searching as an iterative, non-linear process in which users collect small amounts of information from a variety of sources as their needs evolve, rather than following a single, linear path to a fixed answer.

Origin and concept: Bates' essay The Design of Browsing and Berry-Picking for Information Retrieval introduced the term and contrasted berry-picking with traditional linear search. The model emphasizes users as active, opportunistic performers who select, discard, and expand their information set as new gaps and questions appear.

Key features: Information needs are dynamic; search moves through diverse sources including reference works, databases, and the open web. Queries drift, results are revisited, and new questions emerge. Serendipity and exploration are common, and each discovery can change the user's goal.

Impact and applications: Berry-picking has influenced information retrieval research and interface design, supporting features such as browsing, iterative relevance feedback, and faceted or exploratory search interfaces. It underlines the value of context, user goals, and the iterative refinement of information.

Critique: The model recognizes realistic behavior but can be difficult to measure and generalize. Some tasks are more goal-directed and linear, but berry-picking remains a widely cited framework for understanding how people seek information in practice.