摘要：This study examines changes over time in the functionality of social tagging systems. Social tagging systems are conceptualized as a special kind of communication system that enables users to organize and discover information resources using tags as signals. Functionality is defined as the extent to which the system minimizes both encoding and decoding efforts, following Zipf’s principle of least effort. The Yule-Simon model is adopted as the growth mechanism of the system. Two hypotheses are proposed: The first predicts decline, increase, and stabilization of the tag-frequency distribution; the second predicts an oscillation in the functionality leading to a dampened stability. The empirical data support the first hypothesis and partially support the second hypothesis. The article discusses the implications for collective intelligence.
