Fidelis: a policy-driven trust management framework

  • Authors:
  • Walt Teh-Ming Yao

  • Affiliations:
  • Computer Laboratory, University of Cambridge, Cambridge

  • Venue:
  • iTrust'03 Proceedings of the 1st international conference on Trust management
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We describe Fidelis, a policy-driven trust management framework, designed for highly decentralized distributed applications, with many interoperating, collaborative but potentially distrusting principals. To address the trust management needs for such applications, Fidelis is designed to support the principle of separation of policies and credentials, and the notion of full domain autonomy. Based on these, credentials are considered simply as static data structures, much like membership cards in real life. Policies, which are autonomously specified, administered and managed, interpret and provide the semantics for these credentials. In this paper, we describe the Fidelis policy framework which serves as the abstract, conceptual foundation. We also describe a specific implementation of the policy framework, in the form of the Fidelis policy language. Both the syntax and the semantics of the language are described. A discussion is given to show that the Fidelis approach is attractive for many applications.