In July 2003 Ole Jensen and I finished a Technical Report (TR570 of the Cambridge Computer Laboratory) on bigraphical reactive systems; we revised it (TR580) in 2004, with the title Bigraphs and mobile processes (revised). It describes a topographical model of distributed agents that can manipulate their own linkages and (nested) locations. It aims to model both man-made and natural systems, and draws inspiration from the pi calculus and the calculus of mobile ambients. The main technical advance is to generalise the behavioural theory of both these calculi. Here are the slides for a lecture course on bigraphs, based on TR580. I gave it first in September 2003 at the Advanced Course on Petri Nets, at Eichstaett (Germany).
In a subsequent paper, Bigraphs for Petri nets , I applied bigraphs to outline a simple behavioural theory for condition/event Petri nets.
A third Tech Report (TR581), Axioms for bigraphical structure, axiomatises the static structure of bigraphs (i.e. their structural congruence).
In August 2004 Jamey Leifer and I extracted, from his PhD Dissertation and from TR580, our original theory of s-categories on which the bigraph model is based, and applied it in the simpler setting of link graphs -- essentially bigraphs without nested locations. As an example, we developed the application to Petri nets in fuller technical detail. This appears as another Tech Report (TR598), Transitions, link graphs and Petri nets .
In September 2004 I wrote a short Tech Report (TR603) called Bigraphs whose names have multiple locality . This is a smooth generalisation of the binding bigraphs introduced in TR580; I hope it will ease the presentation of some dynamical systems, such as the lambda calculus.
In June 2003 I gave a keynote lecture entitled Computing in space to the annual Conference on Computer-aided Radiology and Surgery. Here also are the slides to go with the talk.
In January 2003 I wrote a short essay called What's in a name?, to be published in a volume of papers written in honour of Roger Needham. It relates to his well-known paper on "Naming", which highlights the notion of a pure name. The paper explains what the pi-calculus allows you to do with pure names. It also considers whether these are all the things you should be able to do with them. As an illustration I use an example on directory lookup from Needham's paper.
At the IFIP World Computer Congress in Toulouse, August 2004, I gave a plenary lecture A scientific horizon for computing . After an account of the <a href="http://www.nesc.ac.uk/esi/events/Grand_Challenges/> UK Grand Challenges Exercise, I discussed two of the seven current Grand Challenge proposals. These are closely linked by the theme of the Global Ubiquitous Computer; one aims to develop scalable engineering design principles, the other to formulate concepts and theories to specify and analyse the designs.