Note: I assume that the "alternative outputs" won't be published on time, so we should use the main ones given.

Output 1

M. Plath and M. D. Ryan. Feature Integration using a Feature Construct. Science of Computer Programming. 41(1), pp.53-84, 2001. Code to support the paper.

Narrative. A feature is a post-hoc addition made to a software system, in order to add new functionality. The paper is the first to make formal and rigorous the idea of a feature construct for programming and specification languages, and it enabled industrially-relevant verification of featured telephony systems to take place in conjunction with BT. The research was sponsored by BT and EPSRC (GR/R02214 evaluated as outstanding). The methods developed in the paper previously won first prize in the open competition at International Workshop in Feature Interaction, 2000, and led to Ryan's 2005 co-chairship of the International Conference on Feature Interaction.

((Not part of commentary: together with background conference papers, the work has had 105 citations (Google scholar).))

Output 2

Dimitar P. Guelev, Mark D. Ryan and Pierre-Yves Schobbens. Model-checking Access Control PoliciesSeventh Information Security Conference (ISC'04). Lecture Notes in Computer Science volume 3225, Springer-Verlag, 2004. 16 pages.

Narrative. Access control systems in web-based workflow systems can become very complicated, and it can be hard to determine whether a configuration really denies or permits access in the intended way. This is the first paper to bring automatic model checking techniques to the verification of access control policies. The paper is supported by freely downloadable tool, and we have used it to find flaws in access control policies for web-based workflow systems.  The work has led to a fruitful collaboration with IBM and Ryan has received a prestigious "IBM Faculty Award" of GBP 18,000 to carry it forward. 

((Not part of commentary: 25 citations in Google scholar.))

Alternative output 2 (if published in time)

Nan Zhang, Dimitar P. Guelev, and Mark Ryan. Synthesising Verified Access Control Systems through Model Checking. Journal of Computer Security (in print), 2007. 58 pages.

Narrative. Access control systems in web-based workflow systems can become very complicated, and it can be hard to determine whether a configuration really denies or permits access in the intended way. An earlier conference version was the first paper to bring automatic model checking techniques to the verification of access control policies. The paper is supported by freely downloadable tool, and we have used it to find flaws in access control policies for web-based workflow systems.  Work has led to a fruitful collaboration with IBM and Ryan has received a prestigious "IBM Faculty Award" of GBP 18,000 to carry it forward. 

((Not part of commentary: Conference papers of which this paper is the culmination have 45 citations.))


Output 3

Steve Kremer and Mark D. Ryan. Analysis of an Electronic Voting Protocol in the Applied Pi Calculus. In Proceedings of the European Symposium on Programming (ESOP'05),  Lecture Notes in Computer Science volume 3444, pages 186-200. Springer Verlag, 2005.

Narrative. Governments the world over are exploring systems for electronic voting. This is the first paper to apply formal verification techniques to verification of electronic voting systems. The paper analysed a classical and established electronic voting protocol [FOO'02], and proved several properties such as vote privacy, voter eligibility, and fairness. The paper led to the successful EPSRC proposal Verifying properties in Electronic Voting Protocols (EP/E029833) to further develop this seminal work, to Ryan's invitation to the Dagstuhl seminar Frontiers of Electronic Voting and membership of the programme committee of Workshop on Trustworthy Elections 2006, 2007.



Output 4


Aybek Mukhamedov and Mark Ryan.  Resolve-Impossibility for a Contract-Signing Protocol. In 19th Computer Security Foundations Workshop (CSFW), IEEE Comp. Soc. Press, 2006.

Narrative.Multi-party contract signing protocols provide a method for n participants to send messages to each other in order to be simultaneously bound by a contract, so as to guarantee fairness (that is, no-one is bound unless everyone is bound). Protocols for achieving this are important for e-commerce and p2p systems, but have proved notoriously difficult to obtain. We show that a 1999 protocol revised in 2004 is still flawed. In a follow-up paper, we propose a new protocol and prove it correct in a rigorous mathematical framework. Our work  resolved a question which was open for 8 years.


Alternative output 4 (if published soon enough)


A. Mukhamedov and M. D. Ryan. Improved multi-party contract signing. Proceedings of the 11th Financial Cryptography, LNCS 4535, Springer Verlag, expected date of publication: September 2007.

Narrative. Multi-party contract signing protocols provide a method for n participants to send messages to each other in order to be simultaneously bound by a contract, so as to guarantee fairness (that is, no-one is bound unless everyone is bound). Protocols for achieving this are important for e-commerce and p2p systems, but have proved notoriously difficult to obtain. We showed that a 1999 protocol revised in 2004 is flawed. We propose a new protocol, and prove it correct in a rigorous mathematical framework. Our work  resolved a question which was open for 8 years.


Mark D Ryan
Last modified: Fri Jul 27 10:41:41 BST 2007