What is complexity and why do we need to think about it? After reading Aaron's last post I realised that it had been quite some time since I'd looked at complexity and that my knowledge of what it is and why it's important to keep in mind had started to fade. Luckily, the next day someone suggested I read the UK government 'do tank' the Design Council's Transformation Design paper. The paper proposes using a human-centred design approach, usually employed for commercial products, in producing solutions for social and economic problems. Existing solutions, such as the division of government into departmental 'silos', are based on the failing assumption that the world is complicated, that is that problems can be solved through analysis, decomposition into the pertinent variables and conditions and finally implementing a logical solution. In the Transformation Design paper, they have neatly summarised an emergent problem solving approach based on the assumption that the world is complex, that it is extremely difficult (impossible for a lot of systems) to decompose a problem into its constituent parts due to the extent of interdependence between the parts of the system and the multiplicity of viewpoints. They put forward that the best way to solve problems in a complex world is through an approach that employs collaboration between disciplines, includes regular people as part of the design team and builds into a solution the ability to adapt to change.
I hope this sheds a little bit more light on the topic. Chris might be able to offer an explanation of some modelling-based approaches to dealing with complexity.
Subscribe to:
Post Comments (Atom)
1 comment:
I like the differentiation between 'complicated' and 'complex' (I'd been pondering that distinction myself recently too...
"Complicated Systems Theory" doesn't quite have the ring to it, does it?).
I was lucky enough to be invited to day-long workshop with a whole lot of complex-systems scientists today at Melb Uni. There were some pretty brainy folk there talking about network epidemiology, simulations, modelling and other stuff. Seems like it is still very much a budding field, lots of big questions and speculations and lots of space for new thinking. Most of my thinking to date has been trying to understand what complexity is, how it manifests in physical, biological and human systems, its implications for our understanding of our place in the cosmos, and our notions of (predictive) science (see my response to Aaron's initial post on 24/02 for more of the story). But I see that this kind of approach, however useful, is actually a bit un-reflexive. People like those I saw talk today are still doing complex-systems science in a regular science way. I.e. via analysis into component parts, using linear-type problem solving methods. Matt's identified the next step, which seems to have been taken in that Transformation Design paper, where we self-consciously allow the process of discovery mirror the world it is trying to transform - i.e. the research itself becomes an emergent process, collaborative and interconnected, etc. Given that this is, in the end, how human problems get solved anyway, it will be interesting to see if the becoming-self-conscious of that strategy changes how things work.
As far as potential inspiration goes (still of the former, rather than the latter, kind though), these links may have done the rounds before but they are still awesome: Visual Complexity Maps
And especially: Big Bizzo Networks.
Post a Comment