1. An Introduction to Decentralized Trust Management Sandro Etalle University of Twente thanks to William H. Winsborough – University of Texas S. Antonio. The DTM team of the UT (Ha, Marcin, Jeroen Jerry)

2. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 2 Overview Reputation-based trust management Rule-based trust management Problems & Challenges (rule-based systems)  scalability & chain discovery  trust negotiation  integrity constraints Conclusions

3. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 3 Reputation-based TM concrete community of cooks (200 people) need to interact with someone you don’t know,  to extablish trust: you ask your friends  and friends of friends ... some recommendations are better than other you check the record (if any)  after success trust increases reputation-based TM – rule-based TM – problems & challenges - conclusions

4. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 4 Reputation-based TM virtual p2p community of hackers (2000 people)  exchange programs & scripts need to interact with someone you don’t know, ... difference with concrete community:  larger, faster trust establishment has to be to some extent automatic reputation-based TM – rule-based TM – problems & challenges - conclusions

5. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 5 for instance reputation-based TM – rule-based TM – problems & challenges - conclusions

6. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 6 challenges trust metrics  how to model and compute trust  evaluating initial trust value  combining evidences, recommendations, reputation management of reputation data  secure & efficient retrieval of reputation data automating trust based decision closing the circle: using experience as feedback reputation-based TM – rule-based TM – problems & challenges - conclusions

7. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 7 Reputation-based TM: salient features open system (different security domains) trust is a measure & changes in time risk-based recommendation based (NOT identity-based) peers are not continuously available Some systems:  PGP,  EigenTrust Algorithm (Stanford) reputation-based TM – rule-based TM – problems & challenges - conclusions

8. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 8 rule-based TM: concrete example reputation-based TM – rule-based TM – problems & challenges - conclusions

9. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 9 rule-based tm, virtual scalability reputation-based TM – rule-based TM – problems & challenges - conclusions

10. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 10 RT: a language for rule-based tm family of languages [Li, Mitchell, Winsborough] four types of credentials EPub.discount  Alice EPub.discount  UTwente.student EPub.discount  FAB.accredited.student EPub.discount  UTwente.student  UTwente.student principal role name principal.rolename = Role trusting principaltrusted principal (somewhere else: delegation) reputation-based TM – rule-based TM – problems & challenges - conclusions attribute-based delegation

11. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 11 some language requirements [Bertino] Monotonicity Constraints (omitted) Credential combination Sensitive Policies reputation-based TM – rule-based TM – problems & challenges - conclusions

12. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 12 Reputation vs rule based TM open system (different security domains) trust is a measure & changes in time risk-based recommendation based (NOT identity-based) peers are not continuously available Some systems: PGP TBD open system (different security domains) trust is boolean & less time- dependent no risk rule (credential) based (NOT identity-based) peers are not continuously available Some systems: keynote, Trust-X reputation-based TM – rule-based TM – problems & challenges - conclusions

13. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 13 Problem 1: scalability attribute-based delegation:  accepting student ID from any university EPub.discount  FAB.accred.student FAB.accredited  UnivTwente UnivTwente.student  Alice Credential chain proves authorization. Scalability problem reputation-based TM – rule-based TM – problems & challenges - conclusions

14. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 14 Problem 2: trust negotiations credentials can be confidential credential disclosure is a matter of... trust three strategies [Seamons]  Naive  Reasonable  Informed additional problem: what do you do with the info in a credential after it has been disclosed reputation-based TM – rule-based TM – problems & challenges - conclusions

15. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 15 Problem 3: control Policies change in time: P  P1 ...  Pn A principal controls only a portion of the policy Delegating trust implies an understanding between principals,  Trusted principals need assistance  Who could get access to what? (Safety)  Who could be denied? (Availability) “No-one should ever be both a buyer and an accountant”  Mutual Exclusion reputation-based TM – rule-based TM – problems & challenges - conclusions

16. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 16 Conclusions Context:  2 or more parties in an open system.  parties are not in the same security domain. Goal  establish trust between parties to exchange information and services (access control) Constraint  access control decision is made NOT according to the party identity BUT according to the credentials it has reputation-based TM – rule-based TM – problems & challenges - conclusions

17. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 17 Open problems Analysis  safety analysis we are now working with Spin in RT0, for RTC (with constraints) nothing is available  of negotiations protocols w.r.t. the TM goals. Integration with other systems  e.g. privacy protection location-dependent policies  ambient calculi? DRM Semantics is not correct when considering:  chain discovery  negotiations is not modular  certainly possible to improve this using previous work on omega-semantics. Types

18. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 18 Integrity Constraints: General Form General: L.l ⊒ R.r  Formally, L.l ⊒ R.r holds in P ( P ⊢ L.l ⊒ R.r) iff [[L.l]] P  [[R.r]] P  sets and intersections are allowed Special cases  Membership: A.r ⊒ { D 1, …, D n }  Boundedness: { D 1, …, D n } ⊒ A.r  expressiveness is limited (it is a universal formula) but we can express all safety properties of [LWM03]  counterexample: at least a manager should have access to the DB

19. IPA Herfstdagen SecurityEtalle: Decentralized Trust Management. 19 Examples buyers and accountants should be disjoint   ⊒ A.buyer  A.accountant every employee should have access to the WLAN network  WLAN.access ⊒ UT.employee welders of BOVAG-accredited workshops should be fellows of the British Institute of Welding Bovag.welder  Bovag.accr.welder Bovag.accr  PietersWorkshop PietersWorkshop.welder  Pieter  BIW.fellow ⊒ Bovag.welder