Do we need a Pattern Language for Smarter Cities?

(Photo of the Athens Olympic Sports Complex from Space by the NASA Goddard Space Flight Center)

The UK Department of Business, Innovation and Skills held a workshop recently to determine how to create guidance for cities considering their approach to Smarter Cities.

A robust part of the debate centred on the challenge of providing “delivery guidance” for cities embarking on Smarter Cities initiatives: whilst there are many visions for smart and future cities; and many examples of projects that have been carried out; there is little prescriptive guidance to assist cities in defining and delivering their own strategy (although I’ve provided my own humble contribution in “Six steps to a smarter city” on this blog; an article which organises a broad set of resources into an admittedly very high level framework).

In setting out a transformative smarter city vision and then taking the steps to achieve it, a great deal of change is involved. Large, formal organisations tend to approach change with prescriptive , process-driven techniques – for all that the objective of change might be defined disruptively by individual insight and leadership or through the application of techniques such as “design thinking“; the execution of the changes required to achieve that objective is usually driven by a controlled process with well defined roles, scope, milestones, risks and performance indicators.

My own employer, IBM, is a vast organisation with over 400,000 employees; a similar number of people to the population of a city of modest size. It was the subject of one of the most famous transformations in corporate history when Lou Gerstner saved it from near-failure in the 1990s. The transformation was achieved by brilliant personal leadership; trial and error; and a variety of techniques and ideas from different sources – there was no “off-the-shelf” process to follow at this scale of organisational change.

But transforming a city is not the same thing as changing an organisation, however big. A city is a complex system of systems, and we have comparatively little knowledge about how to drive change in such an environment. Arguably,we should not even think about “driving change” in city ecosystems, but rather consider how to influence the speed and direction of the changes that will emerge from them anyway.

Some very different approaches to process-driven change have emerged from thinking in policy, economics, planning and architecture: the Collective Research Initiatives Trust‘s study of Mumbai, “Being Nicely Messy“; Colin Rowe and Fred Koetter’s “Collage City“; Manu Fernandez’s “Human Scale Cities” project; the “Massive / Small” concept and associated “Urban Operating System” from Kelvin Campbell and Urban Initiatives; and CHORA’s Taiwan Strait Atlas, for example have all suggested an approach that involves a “toolkit” of ideas for individuals and organisations to apply in their local context.

(In this light, it’s interesting to observe that in order to steer the ongoing growth of IBM following the transformation led by Lou Gerstner, his successor as CEO, Sam Palmisano, took the organic approach of seeking to inspire a consistent evolution of business behaviour across all 400,000 individual IBMers by co-creating and adopting a common and explicit set of “values”).

(Stories of Mumbai: an exploration of Mumbai’s history of urban development, and its prospects for the future, using storytelling and puppetshows, by the BMW Guggenheim Lab)

In “Resilience: Why Things Bounce Back“, Andrew Zolli and Ann Marie Healy, give a fascinating description of the incredible impact such approaches can achieve through the example of the response to the earthquake near Port-au-Prince in Haiti on January 10, 2010 that was led by Patrick Meier, the Ushahidi information crowd-sourcing platform and the Tufts Fletcher School of Law and Diplomacy in Massachusetts. Meier catalysed an incredible multi-national response to the earthquake that included the resources of organisations such as Thomson Reuters, Digicel (the largest mobile phone company in Haiti), and MedicMobile; and just as importantly hundreds of individuals literally spread across the world, with nothing more in common than a desire to do what they could to contribute:

“I told people, ‘We’re going to let this be emergent,’” Meier explained. “There are so many things that need to happen every single hour and so many things that need to keep evolving in such a short amount of time. I have to just let it flourish and deal with what happens when it starts getting inefficient.” The open nature of the platform – both the code that powers Ushahidi and the collaborative nature of the mapping – meant that people could easily be recruited to perform discrete, useful tasks with a minimum of formal authority.”

(Patrick Meier, quoted in “Resilience: Why Things Bounce Back“, p179, by Andrew Zolli and Ann Marie Healy)

In my own work, I’ve tried to follow a similar course, inspired first by the Knight Foundation’s report on the Information Needs of Communities. The Knight Foundation counsel a process of engagement and understanding between institutions and communities, in order to identify the specific information and resources that can be most usefully made available by city institutions to individual citizens, businesses and social organisations. As I described in “The Amazing Heart of a Smarter City: the Innovation Boundary“, the resulting portfolio provides a toolkit customised to the needs of a city, and that can be used to shape a collective case for investment in the development of that city.

The idea of a toolkit recognises both that no one approach, philosophy or framework is applicable to every city, or to every context within a single city; and that an idea that works in one place might work in many others.

For example, in the UK, the regions around the cities of Birmingham and Manchester are of similar size in terms of population and economic activity; but they are very different in the structure of their political administrations and economies. The approach that one of these cities adopts as its Smarter City strategy will not necessarily transfer to the other.

In contrast, however, specific ideas concerning economic development and the attraction of talented young people that I’ve found useful in Sunderland in the UK have been inspired by past experience in Wuxi, China and New York State; and in turn have informed initiatives in Spain, Singapore and Nairobi; in other words they have transcended contexts of vastly different size, culture and economics.

A tool that emerged from town planning in the 1970s and that was then adopted across the information technology industry in the 1980s and 1990s might just provide the approach we need to harness this information. And it’s perhaps not surprising that a tool with such provenance should become relevant at at time when the architects of information technology systems, buildings and cities are finding that they are working within a common context.

That tool is the “Design Pattern”.

A Pattern Language for Smarter Cities

(A pattern language for social software features, image by Amber Case)

The town planner Christopher Alexander invented “design patterns” in the 1970s. He addressed the challenge that many problems in planning were (and are) too large and complex for one person to consider them in their entirety at one time; and that it is hence necessary to break them down into sub-problems.

The difficulty is that it is not at all straightforward to break a problem into sub-problems that can be solved effectively in isolation from each other.

Consider city transport systems: in many cases, road management, bus operations and the rail network are the responsibility of different organisations. It “makes sense” to break up transport systems in this way because each is different; and so different organisations are better at running them effectively.

But from the perspective of the users of transport systems, it doesn’t make sense to do this. Bus and rail timetables don’t work together; cars, buses, freight vehicles, bicycles and pedestrians have conflicting requirements of road space; and the overall system does not behave as though it is designed to serve travellers consistently.

In “Notes on the Synthesis of Form” in 1969, Alexander described a mathematical technique that could be used to manage the complexity of large problems and to break them down into sub-problems in a way that accommodated interdependencies between them. As a result, those sub-problems could be solved separately from each other, then integrated to form an overall solution.

This process of decomposition, solution and integration is fundamental to process-driven approaches to the design and delivery of complex solutions. It is not possible, for example, to assign responsibilities to individuals and teams without going through it. Many projects that fail do so because the  problem that they are addressing is not decomposed effectively so that individual teams find that they have overlapping areas of responsibility and therefore experience duplication and conflict.

However, in developing his technique for decomposing problems, Alexander concluded that it was overly complex, rigid and impractical; and he recommended that it should never be used. Instead, he suggested that it was more useful to focus not on how we deal with problems; but on how we re-use successful solutions.

By identifying and characterising the components of solutions that have been proven to work, we enable them to be reused elsewhere. Christopher Alexander’s particular insight was to recognise that to do so successfully, it is vitally important to precisely describe the context in which a solution is applicable. He called the resulting description of reusable solutions a “design pattern”; and a collection of such descriptions, a “pattern language“.

Design patterns and pattern languages offer a useful combination of formal and informal approaches. They are formal in that each pattern is described in a consistent way, using a structured framework of characteristics. And they are informal in that the description isn’t constrained to that framework of characteristics; and because design patterns do not assert that they should be used: they are simply there to be used by anyone who chooses to do so.

Christopher Alexander’s patterns for town planning and architecture can be found in his books, or online at the “Pattern Language” community; in information technology, Martin Fowler’s “Enterprise Application Architecture Patterns” provide a similar example.

To my knowledge, no-one is yet curating a similar set of Smarter Cities patterns; I believe that there would be great value in doing so; and that in order to do so skills and expertise across domains such as planning, architecture, technology, social science and many others would be required.

In the final part of this article, I’d like to suggest some examples of Smarter City initiatives and ideas that I think can be usefully described as patterns; and to give one example of such a description. Please do share your views on whether this approach is useful by commenting on this blog, or through one of the Linked-In discussion groups where I’ve posted links to this article.

Design Patterns for Smarter Cities

Here are just a few of the ideas I’ve seen applied successfully in more than one place, either as part of a Smarter City strategy, or simply as valuable initiatives in their own right. It is certainly not an exhaustive list – a quick survey of Linked-In discussion Groups such as “Smart Cities and City 2.0“, “Smarter Cities” and “Smart Urbanism” will reveal many other examples that could be described in this way.

• Information Partnerships – collaborations between city institutions, communities, service providers and research institutions to share and exploit city data in a socially and financially sustainable system. (I’ve provided a more detailed description of this example below).

• Incubation Clouds – the use of Cloud Computing platforms and hybrid public/private commercial models to enable co-operative investment in technology capabilities that can lower the barriers to successful innovations in city services. Examples: Sunderland’s “City Cloud” and the Wuxi iPark.

• Community Energy Initiatives – the formation of local energy companies to exploit “smart grid” technology, local energy generation (such as solar panels, wind power, wave power, geo-thermal power and bio-energy) and collaborative energy consumption to reduce carbon emissions and reliance on external energy sources. Examples: Eco-island and Birmingham Energy Savers.

• Social Enterprises – a collective term for models of business that audit themselves against social and environmental outcomes, as well as financial sustainability and returns. Examples: co-operatives, credit unions and organisations using “triple-bottom-line” accounting.

• Local Currencies – local trading systems using paper or electronic currencies that are issued and accepted within a particular place or region; and that have the effect of influencing people and businesses to spend the money that they earn locally, thereby promoting regional economic synergies. Examples: the Swiss alternative currency, the “Wir”; the Brixton Pound and Bristol Pound; and the Droplet electronic currency in Birmingham.

(The components of a Smart City architecture I described in “The new architecture of Smart Cities“)

In order to describe these concepts more completely as re-usable patterns; and in a way that allows them to be compared, selected in comparison to each other, or used together; it is important that they are described consistently, and in a way that accurately identifies the context in which they are applicable.

To do so requires that we describe the same aspects of each pattern; and that we describe each aspect using a common language. For example:

• The city systems, communities and infrastructures affected; using a framework such as the “The new architecture of Smart Cities” that I described last year, shown in the diagram above.

• The commercial operating model that makes the pattern financially sustainable.

• The soft infrastructures, hard infrastructures and assets required to implement the pattern, perhaps described using the categorisation I suggested in “Pens, paper and conversations. And the other technologies that will make cities Smarter“.

• The driving forces that make the pattern applicable, such as traffic congestion; persistent localised economic inactivity; the availability of local energy sources; or the need to reduce public sector spending.

• The benefits of using the pattern; including financial, social, environmental and long-term economic benefits.

• The implications and risks of implementing the pattern – such as the risk that consumers will not chose to change their behaviour to adopt more sustainable modes of transport; or the increasing long-term costs of healthcare implied by initiatives that raise life-expectancy by creating a healthier environment.

• The alternatives and variations that describe how the pattern can be adapted to particular local contexts.

• Examples of where the pattern has been applied; what was involved in making it work; and the outcomes that were achieved as a result.

• Sources of information that provide further explanation, examples of use and guidance for implementation.

I’ll finish this article by given an example of a Smarter City pattern described in that way – the “City Information Partnership”.

(Coders at work exploiting city information at the Birmingham “Smart Hack”, photographed by Sebastian Lenton)

An Example Pattern: City Information Partnership

(Note: the following description is not intended to be written in the fluent style that I usually hope to achieve in my blog articles; instead, it is meant to illustrate the value in bringing together a set of concisely expressed ideas in a structured format).

Summary of the pattern: a collaboration between city institutions, communities, service providers and research institutions to share and exploit city data in a socially and financially sustainable system.

City systems, communities and infrastructures affected:

(This description is based on the elements of Smarter City ecosystems presented in “The new Architecture of Smart Cities“).

• Goals: Any.
• People: Citizens; innovators.
• Ecosystem: All.
• Soft infrastructures: Innovation forums; networks and community organisations.
• City systems: Any.
• Hard infrastructures: Information and communications technology.

Commercial operating model:

City information partnerships are often incorporated as “Special Purpose Vehicles” (SPVs) jointly owned by city institutions such as local authorities; universities; other public sector organisations such as schools, healthcare providers and emergency services; services providers such as transportation authorities and utilities; asset owners and operators such as property developers and facility managers; local employers; and private sector providers such as technology companies.

A shared initial investment in technology infrastructure is often required; and in order to address legal issues such as intellectual property rights and liability agreements.

Long-term financial sustainability is dependent on the generation of commercial revenues by licensing the use of data by commercial operations. In cases where such initiatives have been supported only by public sector or research funding, that funding has eventually been reduced or terminated leading to the stagnation or cessation of the initiative.

Soft infrastructures, hard infrastructures and assets required:

Information partnerships only succeed where they are a component of a co-creative dialogue between individuals and organisations in city institutions such as entrepreneurs, community associations, local authorities and social enterprises.

Institutional support is required to provide the models of legal liability and intellectual property ownership that create a trusted and transparent context for collaborative innovation.

Technologies such as Cloud Computing platforms; information management; security; analytics, reporting; visualisation; and data catalogues are required to manage city information and make it available and useful to end users.

Information partnerships require the participation of organisations which between them own and are prepared to make available a sufficiently broad and rich collection of datasets.

Driving forces:

Information is transforming the world’s economy; it provides new insight to support business model creation and operation; makes new products and services possible; and creates new markets.

At the same time global and local demographic trends mean that the cost-base and resource usage of city systems must change.

Information partnerships expose city information to public, private, social and academic research and innovation to discover, create and operate new models for city services; with the potential for resale elsewhere; leading in turn to economic and social growth.

(A visualisation created by Daniel X O Neil of data from Chicago’s open data portal showing the activities of paid political lobbyists and their customers in the city)

Benefits:

Community hacktivism can usually be engaged by information partnerships to create useful community “apps” such as local transport information and accessibility advice.

The creation of new information-based businesses creates local employment opportunities, and economic export potential.

Information partnerships can provide information resources for technology education in schools, colleges and universities.

New city services developed as a result of the information partnership may provide lower-carbon alternatives to existing city systems such as transportation.

Implications and risks:

If participating organisations such as local authorities include the requirement to contribute data to the information partnership in procurement criteria, then tendering organisations will include any associated costs in their proposals.

For information partnerships to be sustainable, the operating entity needs to be able to accrue and reinvest profits from licenses to exploit data commercially.

The financial returns and economic growth created by information partnerships can take time to develop.

Genuinely constructive partnerships rely on effective engagement between city institutions, businesses and communities.

Existing contracts between local authorities and service providers are unlikely to require that data is contributed to the partnership; and the costs associated with making the data associated with those services available will need to be negotiated.

Alternatives and variations:

Some organisations have provided single-party open data platforms. These can be effective – for example, the APIs offered by e-Bay and Amazon; but individual organisations within cities will rarely have a critical mass of valuable data; or the resources required to operate effective and sustained programmes of engagement with the local community.

Many advocates of open data argue that such data should be freely available. However, the majority of platforms that have made data available freely have struggled to make data available in a form that is usable; to expand the data available; to offer data at a reliable level of service; or to sustain their operations over time. Making good quality data available reliably requires effort, and that effort needs to be paid for.

Examples:

• Dublin – http://www.dublinked.ie/
• San Franciso – http://radar.oreilly.com/2013/01/san-francisco-open-data-appallicious.html  
• Chicago – https://data.cityofchicago.org/

Sources of information:

The UK Open Data Institute is championing open data in the UK – http://www.theodi.org/

O’Reilly Media have published many informative articles on their “Radar” website – http://search.oreilly.com/?q=open+data&x=0&y=0&tmpl=radar

The report “Information Marketplaces: The new economics of cities” published by Arup, The Climate Group, Accenture and Horizon, University of Nottingham – http://www.arup.com/Publications/Information_Marketplaces_the_new_economics_of_cities.aspx

Finally, I have written a series of articles on this blog that explore the benefits and challenges associated with the collaborative exploitation of city information:

• Why open city data is the brownfield regeneration challenge of the information age
• Open urbanism: why the information economy will lead to sustainable cities
• How cities can exploit the information revolution
• Open data isn’t free data

What next?

It has been an interesting exercise for me to write this article. Many of the ideas and examples that I have included will not be new to regular readers of this blog. But in describing the idea of an “Information Partnership” as a formal design pattern I have brought them together in a particularly focussed and organised manner. There are many, many more ideas and examples of initiatives within the Smarter Cities domain that could be described in this way; and I personally believe that it would be valuable to do so.

But my opinion on that subject is less valuable than yours. I would really appreciate your thoughts on whether the “Smarter City Design Patterns” I’ve suggested and explored in this article would be a valuable contribution to our collective knowledge.

I look forward to hearing from you.