About Michael Batty

I chair CASA at UCL which I set up in 1995. I am Bartlett Professor In UCL.

Against the Smart City


Absolutely the best thing to read on the corporate hype and innuendos from the big computer companies pedalling the idea of ‘the smart city’. Adam Greenfield’s new book – that you can only get on Kindle and which was my first Kindle purchase that I read on my iPad (a success I must say) – is a wonderful and eloquent essay on the extreme hype surrounding the top down new town-like smart cities of Songdo (in South Korea), Masdar (in the UAE), PlanIT Valley (near Paredes in Portugal). He also comments on Singapore, Rio de Janeiro and some of the other established cities who are injecting automation into their urban services and other functions from the top down. His message is that most of the smart cities hype associated with IBM, Cisco, and Siemens amongst others which he recounts in detail is based on the most simplistic of notions as to what a city actually is. There is a nice summary on the Urban Omnibus Blog

Readers of this blog will know that I espouse the notion that cities have almost infinite complexity and that their planning is entirely contingent on the people and places for which we might identify problems to be solved. Plans throughout the 20th century have failed entirely to anticipate this complexity and the diversity that characterises our cities has been trampled over and often destroyed in the name of planning. The modern movement in architecture but also the scientism that accompanied the early arrogance in public planning, particularly housing and transportation, are examples. The prospect that the same is happening again with the ‘smart city’ is what Adam Greenfield is so concerned about and rightly so. The image of the smart city which comes from the corporate world betrays a level of ignorance about how cities function that is woeful and dangerous while the notion that all the routine functions of cities can be nearly ordered and automated is simply fanciful. Adam does a great job in describing this world and this is by far the best, in fact perhaps the only, really detailed and incisive critique of what are in fact the ‘smart city new towns’.

In fact smart technologies – I still have problems with the term ‘smart’ in this context as it is so un-English English – are being employed quietly and less obtrusively in many contexts. Transport for London are a major agency in a world city that is at the forefront of new technologies with their smart ticketing, APIs telling you where trains are located, and their focus on trying to minimise disruption. But so are many places which you have never heard about. I went to Zurich last Monday and as ever was amazed by the calm, considered integration of the different modes on the transport system that enabled me to use bus and train to get to ETH from the airport with virtually no wait and with me being informed at every stage of the journey (in English) of where I was and how long it would take me to get to my destination. New information technologies lies at the basis of all of this. In fact it would now appear in many cities that smart systems technologies are being introduced faster and more efficiently from the bottom up in the classic and necessary incremental, evolutionary fashion that all complex systems build upon. Adam makes all these points and more in his essay which might be summed up in the simple notion that the smart cities corporate world sees cities as being ‘complicated’ rather ‘complex’. Complex is of course the emerging wisdom of how we need to treat cities so that we can avoid being entrapped in the mesh of wicked problems that emerge when life is treated too simplistically, when life is considered to be based in immediate order effects, not on order effects that can persist and grow in time often the further away you are from the initial source. A key message of complexity theory.

But Against the Smart City comes with a warning. We need to temper the corporate response to automating the city if only to avoid the kind of things that have plagued public intervention many times before when new technologies are imported into political contexts. The experience with urban models half a century ago in New York City and other places around the world must not be repeated. I wrote about this in my last but one blog post – two below this one – where I directed the reader to the commentaries that myself and colleagues wrote on the notable article by Douglas B. Lee published 40 years ago in 1973 called ‘Requiem for Large Scale Models’. To avoid a ‘Requiem for Smart Cities’, we need to take Adam Greenfield’s message to heart and these should be essential reading for all of us who are involved in the smart cities movement.

Rediscovering the Old Days


I first met Leo Kadanoff in Cambridge, England in the summer of 1974, I think. He would not remember me but he had been invited to give a talk on urban dynamics at the second Cambridge Land Use and Built Form Studies Conference on Urban Development Models as he had published an article in Simulation (From simulation model to public policy: An examination of Forrester’s ‘Urban Dynamics’) in 1971 which speculated on how to improve Jay Forrester’s model which was an example of how to apply Systems Dynamics to Cities. Forrester’s work attracted a lot of attention in those days as he had produced a simulation of a city which had no space – it was the central city but – lots of time – dynamics – and it did not manifestly allude, by his own admission, to anything that had been produced in urban theory and models so far.

Kadanoff gave an inspiring talk, I remember, but I did not know his background other than the fact that he was a physicist and assumed his interest in cities was a one off. It was not until many years later once I had encountered ideas about fractals and scaling, that I realised that it was the same Leo Kadanoff who had been one of the architects of renormalisation theory in statistical physics. And his YouTube clip puts all this in perspective.

More recently, I came across another article by him and his colleagues (Computer Display and Analysis of Urban information Through Time and Space) published in the second volume of Technological Forecasting and Social Change published in 1970 one year earlier than his Simulation article. What is amazing about this first paper is how prescient it is. Not only is it about early computer graphics in our field and those who follow my complexcity.info blog will know that I spent a few years in the 1980s reskilling in computer graphics but it is also about the limits to modelling. On my blog you will find a potted history of computer graphics in our field but this article is one of the first to show the power of interactive visual display. Not only is planning support and interactive visualisation anticipated, but the notion of interactive design of urban futures is hinted at.

You may say what is remarkable about all this. Many people anticipated this and yes they did. Well, it is very early work of course and fills in a bit more of the picture. It happened in parallel to Alan Schmidt’s East Lansing movie and the genesis of GIS at the Harvard Computer Graphics Lab. And it also built on the first interactive devices used in the Chicago Area Transportation Study in the late 1950s where Illinois Institute for Technology was pioneering the Cartographatron (There is a picture in my New Science of Cities book). All of this is on my blog but the Kadanoff paper is well worth looking at and you can get a copy if you click the link here. The article which is online by the journal is an awfully bad scan and the one I have made here is better.

Last but not least, what all this implies is that we are able to rediscover the past as virtually everything gets scanned and published on the web. I would not have found this but for the web. I never would have gone back to look at this journal. I don’t think it was even in the Reading University library back in 1971 when it appeared when I was a lecturer there. I would have had to get it on interlibrary loan and my guess is that it simply didn’t appear on my radar. There are many others things now that we are discovering about the past which help fill in the picture of how our field has evolved and is evolving. My last post about large scale models also linked to literature from past times that we are only becoming aware of. This to me is as important as new work for it establishes a perspective on our field.



The Requiem Revisited


40 years ago, Douglass B Lee published his notorious article “Requiem for Large Scale Models” in the Journal of the American Institute of Planners on the demise of the first generation of urban computer models in the United States. In it he identified seven ‘deadly’ sins of modelling: defining these relative to our understanding of cities at that time and the technologies used to implement the models as: Hypercomprehensiveness, Grossness, Hungriness, Wrong-headedness, Complicatedness, Mechanicalness, and Expensiveness.

20 years ago, a number of us – myself, Britton Harris and Michael Wegener gathered under the auspices of Dick Klosterman in Chicago at the APA conference to review Lee’s findings. This lead to a special issue of the Journal of the American Planning Association (renamed from the JAIP) where we all recounted our reactions to the intervening 20 years with respect to large-scale models. In fact the field was just about to upturn at that point. Large scale models we need – indeed desperately so – to examine and structure impacts analysis –and a year later in Dallas Texas federal highways sponsored a meeting that we to assess the field and provide momentum for the continued upturn. In fact the special issue was rather disappointing. It was written at a turning point for sure and since then the field has revived in some respects and certainly there are now many more and different types of urban models. But in way we are still grappling with the problems of dynamics, equilibria,  predictions from such models and of course they way we might use them in supporting planning. A recent theme issue of Environment and Planning B reviews progress with LUTI models

20 years on again and we face new issues. Large-scale models almost seem like history now with new approaches to cities coming from every quarter. In particular the smart cities movement is generating enormous momentum and big data is shortening the time horizons over which we might consider planning to be effective. In some respects the new found interest in smart cities is attracting a very different set of people who are intent on computerising our approach to their management and understanding and I wonder just what this holds for applications. Are we about to repeat the mistakes of the past? Does new technology imply that we will always overreach ourselves? What are dilemmas and pitfalls of big data, new forms of management and control using computer systems and so on?

In these two commentaries in the current issue of Environment and Planning B, we seek to unpack these notions. Marco te Brömmelstroet, Peter Pelzer, and Stan Geertman first review Lee’s article and then your truly comments on these issues in terms of the smart cities movement.  I will not elaborate this message further but simply refer you to these commentaries and urge you to form your own views of what is happening now and how it relates to what wee happening then. It’s a fascinating dilemma.

The picture above is taken from http://www.toysperiod.com/blog/scale-models/toysperiods-model-railroader-review-2004-2010/