Suchergebnisse

E-Infrastructures recognise that controlling information security is crucial for providing continuous and trustworthy services for their user communities. Such infrastructures, including grids and clouds, are subject to many of the same threats and vulnerabilities as each other because of the use of common software and technologies. Users who take part in more than one infrastructure are potential vectors that can spread infection from one infrastructure to another. All infrastructures can benefit from working together and sharing information on security issues. Security for Collaborating Infrastructures (SCI) is a collaborative activity within the WISE trust community. The aim of the SCI trust framework is to manage cross infrastructure operational security risks. It builds trust between Infrastructures by defining policy standards for collaboration. The SCI group published version 1 of its trust framework in 2013. Two derivative frameworks have also been published; SIRTFI in 2015, and SNCTFI in 2017. WISE/SCI has more recently produced version 2 of the SCI trust framework, to reflect changes in technology, culture and to cover a broader range of infrastructures. The framework contains numbered requirements in five areas (operational security, incident response, traceability, participant responsibilities and data protection) that each Infrastructure should address as part of promoting trust between Infrastructures. SCI's updated version 2 was officially endorsed during the TNC 2017 conference by representatives of EGI, EUDAT, GÉANT, GridPP, HBP, PRACE, SURF, WLCG and the USA's XSEDE e-infrastructure. The webinar will present the SCI Trust Framework together with current work on a new baseline AUP and a Policy Development Kit. Possible future activities will also be presented. Speaker Bio: David Kelsey is head of the particle physics computing group at STFC, UK and has been leading Grid Security activities in many projects. He founded the Joint (WLCG/EGEE) Security Policy Group in 2004. He is currently the Chair of the WISE steering committee and was founder of the SCI activity. He has a Masters degree in Physics (Trinity College, Cambridge) and a PhD in Physics (University of Birmingham). ; NSF Grant #1547272; European Union Grant #777536 ; Open

We assume that a decision‐maker has expected utility preferences over a large space which includes some variables not observable by the theorist. These will induce preferences over observable variables, which typically will not have the expected utility form. This paper focuses on implications for multi‐period decisions. We show that such preferences are not vulnerable to 'Dutch books'. In particular, we consider preferences arising from non‐additive subjective probabilities and show how they can arise as induced preferences.