BCS-FACS Evening Seminar Series

The Three Amigos: or why 'There is more to specifying systems that connect to the physical world' than they admit

                  Professor Tom Maibaum

                 McMaster University, Canada

                       7 June 2006
                          6pm start

                  BCS London Offices
                  First Floor,
                  The Davidson Building
                  5 Southampton Street
                  London WC2E 7HA

In Cliff Jones's BCS FACS talk on 24 April on joint work with Ian Hayes and Michael jackson
[see http://www.bcs-facs.org/events/EveningSeminars/cliff_jones_06.html], some claims were made
about the efficacy of what was being proposed and the possibly mistaken view of the so called
'4 variable model', about there being no need for a model of the environment internalised in the
software, and so on. The work is also based on proposals of Michael Jackson that require an
ab initio distinction between environment and software. I will argue that these assumptions are
wrong, outline a different set of assumptions and adduce evidence for these assumptions. One
important assumption will be that the ab initio distinction referred to above cannot and should
not be made. Further, I will argue that internalising an appropriate model of the environment
in the software is a necessity in order to control it and that internalising in the software a
distinction between normal and abnormal behaviour is a necessary condition for defining
fault tolerance and self management mechanisms.

Refreshments will be served from 5.30pm

The seminar is free of charge and open to everyone. If you would like to attend, please email
Paul Boca [Paul.Boca@virgin.net] your name by 4 June 2006. Pre-registration is required, as
security at the BCS Offices is tight.

Location of the venue: http://www.bcs.org/upload/img/londonsscolour.jpg

FACS Evening Seminars: http://www.bcs-facs.org/events/EveningSeminars