A reference model for designing effective reputation information systems

Reputation is a socially constructed, distributed knowledge phenomenon that creates powerful incentives for good behavior. The idea of using an information system to make reputation explicit and measurable is promising and not new. Well-known examples include Amazon’s seller ranking system and eBay’s feedback forum. The thesis of this article is that future reputation information systems can be more effective and useful than existing systems if their designs are informed by multi-disciplinary research on reputation. The contribution of this article is a reference model, or meta-specification, that describes the essential functionality and behavior that embedded information systems need to provide to be effective and useful mechanisms for making reputation explicit and measurable. The reference model specified in the industry-standard Unified Modeling Language (UML) is based on a multi-disciplinary understanding of reputation. The model addresses systems that focus on personal reputation. Extensions for organizational reputation are discussed along with other challenges for future research.