I had an interesting discussion this morning with Nicco Mele, Visiting Edward R Murrow Lecturer at the Kennedy School of Government, comparing thoughts about the network of the city and the network of cyberspace – each a network of things. I introduced him to Space Syntax and shared a few thoughts I have had since reading “The Cathedral and the Bazaar”, one of the readings on his course: DPI-659 Media, Politics & Power in the Digital Age.
Nicco had earlier sent me a link to a post on Dan Hill’s blog: “The City of Sound“, which added further food for thought about relationships between different kinds networks: digital and spatial.
This post is part note of our discussion and part reflections since.
The City of Sound presents a compelling description of the street as platform. Nicco showed me an iPhone app tyat brought up local information as an overlay on the camera view – thus telling the user what they are looking at and guiding them to whe they might be searching for.
The development of such technology lends itself to further enhancements in the ability of humans to “see” the worked around them. But what about the street as not only platform but network too – creating intended and unexpected connections not only between people and technology but especially between people and other people?
The unexpected encounter system between people in cities is an essential component of sustainable urbanity. Gated communities and segregated land uses remove this essential component.
Describing the urban network
Space Syntax analysis defines spatial networks in terms of linear elements of space: street segments between street intersections.
Mathematical algorithms then calculate various network properties of these street segments using concepts of mathematical “depth”. Key among these properties are measures of “near-ness” and “between-ness”, each of which can be measured at different scales eg at a local scale (useful when thinking about pedestrian journeys), an intermediate scale (eg cycling) and at the larger scale (eg driving).
Urban network properties
Urban systems throughout the world are typically characterised by having a small number of long, continuously unfolding chains of street segments (that typically pass from edge to centre of the city) lines and a large number of short lines that make up street grids between these lines.
Most historical urban systems are grids, not “trees”. Grids allow route choice, browsing, the making of minor navigational errors without catastrophic consequences. Trees limit choice, prevent browsing and one wrong move can spell trip disaster. Trees prevail in recent urban planning – they are easier to conceive and traffic on them is easier to control.
Trees remove through movement – a seemingly essential component of safe and successful urbanity, if a counterintuitive one to generations raised on “stranger equals danger”.
But 80% of London retail locates on the 20% most integrated lines. Isn’t this the head outperforming the tail?
Defining the city
Bill Hillier offers a spatial network definition of a city as being a foreground network of linked centres set against a background tapestry of residential streets. Is the web anything like this ie with with priority given to a small number of clusters of domains?
Space Syntax spatial network analysis reveals that each centre has its own profile. Within this profile, a balance of integration at different spatial scales seems critical – too local or too global is risky. In urban centres it seems healthy to bring different scales as well as different modes of movement into a relationship with each other. This is one of the roles of a centre. Not every scale needs coexist in every centre – just more than one scale. This has implications for different modes of transport and their presence or absence on urban streets. Likewise, there are implications for traffic speed.
As Ben Hamilton-Bailie (Loeb Fellow ’00) has observed, slow traffic speeds within urban centres are advantageous since this allows for a better quality of human interaction – slower speed traffic also flows more freely. It is safer too. High speed traffic within urban centres may be globally efficient (although congestion is a factor) but it is locally divisive eg the highway that cuts off home from park. Between centres, however, high speed movement seems right. Are there similarities on the web?
Planned versus unplanned encounters and phenomena
The Cathedral and the Bazaar – it seems counterintuitive how a more effective product emerges from a seemingly unbounded, unstructured network of individual efforts; more effective than the prefect of a bound, hierarchical entity. Open source v closed source.
There are clear parallels in urban planning. Many planned places have failed to create the very community that was intended. Urban places have a habit of defying intentions. Great effort goes in and poor results come out. By contrast, little gems emerge in unexpected places. This leads to people doubting the effectiveness of planning processes, preferring instead the unexpected manifestations of serendipity. This is a threat to conventional, professional governance structures.
Further amplifying this threat are the strongly socially bonded environments to be found in many unplanned settlements. These places, formed without formal planning process, foster communities and local economies. Not always, but often enough to suggest a counter model for human development. Especially so if the infrastructure for water supply and waste handling can be incorporated into the informal planning an design process.
The future for urban planning must be to learn from such places and for professionals to engage with local community members. Putting technology in the hands of non-professional users is the subject of a future post but the essence of that argument is that technology, most often in the form of maps and models, provides a common platform for discussion, especially when it can be visualised through the medium of augmented reality.