Copy the page URI to the clipboard
Classen, Andreas; Laney, Robin; Tun, Thein Than; Heymans, Patrick and Hubaux, Arnaud
(2008).
DOI: https://doi.org/10.1145/1370811.1370826
URL: http://mcs.open.ac.uk/ttt23/publications/AClassen-...
Abstract
Central to the problem frames approach is the distinction of three different descriptions: requirements R, domain assumptions W and specifications S, tied together with the so-called 'frame concern', a proof obligation that has to hold between them if a problem diagram is to be correct: S, W |- R. The form this proof should take is not fixed a priori. It might, however, be desirable to automate it in order to allow for an efficient analysis of large diagrams. To make this possible, we follow some earlier suggestions to use the Event Calculus as a suitable formalism for these descriptions. The main contribution of the present paper is a set of consistency rules as well as guidelines for passing from a problem diagram to its formal description.