Regularity-Based Trust in Cyberspace

One can distinguish between two kinds of trust that may be placed in a given entity e (a person or a thing), which we call: familiarity-based trust and regularity-based trust. A familiarity-based trust in e is a trust based on personal familiarity with e, or on testimonial by somebody who is familiar with e, directly or indirectly, or even on some measure of the general reputation of e. A regularity-based trust is based on the recognition that e belongs to a class, or a community, that is known to exhibits a certain regularity—that is, it is known that all members of this class satisfy a certain property, or that their behavior conforms to a certain law. These two types of trust play important, and complementary, roles in the physical world. But, as we shall see, the role of regularity-based trust in cyberspace has been limited so far, because of difficulties in establishing it in this context. It is this latter kind of trust, which is the focus of this paper. We will describe a mechanism for establishing a wide range of regularity-based trusts, and will demonstrate the effectiveness of this mechanism, by showing how it can enhance the trustworthiness of a certain type of commercial client-server interactions over the internet. ∗Work supported in part by NSF grant No. CCR-98-03698