City of Sound is about cities, design, architecture, music, media, politics and more. Written by Dan Hill since 2001.

Trees, lattices, suburbs, and software

Written in

by

Brilliant post by Andrew Otwell joining the dots and connecting Clay Shirky’s report on the advantages of building ‘highly-localised’, or situated, software, Christopher Alexander’s seminal "A City Is Not A Tree" (which Joshua Kaufman had mailed me about and I had been meaning to follow up), and Abstract Dynamics’ post on navigating suburbs. I’m going to pull a hefty ol’ quote from Andy’s post because, well, it’s worth it:

"Navigating the suburbs is literally walking the DOM tree, an algorithmic march through a rigid structure, navigating by the neighboring nodes of a tree, the same stores off every highway exit … And what structure is conspicuously absent from the examples Shirky describes, yet is the catalyst for strip-mall scalability? The hierarchical tree, the technical and experiential structure of most sites on the web. Alexander makes the point clearly (stop and at least scroll through his illustrations on this page if you haven’t yet): "…this tree, and the idea of a single hierarchy of urban cores which is its parent, do not illuminate the relations between art and city life. They are merely born of the mania every simple-minded person has for putting things with the same name into the same basket."  Real cites, of course, have evolved a certain amount of overlap between functions, spaces, and people; more small pieces loosely joined … Alexander goes on to point out the particular difficulty of conceptualizing and visualizing semilattice structures. (His example of one, using an abstract painting, is wonderfully subtle.) There are some good reasons that urban planners, just like software designers, are unable to see past issues of scalability, and fall back on tree structures for everything."

Go read the whole thing though. Interesting stuff.

Heyblog: Structure & Situated Software

Tags

2 responses to “Trees, lattices, suburbs, and software”

  1. Tom Carden Avatar

    Great link, thanks for the pointer.
    I was thinking along the lines of “this data isn’t a tree” for a recent social network visualisation project, and discussion with one of the tutors on my course led me to re-read Tim Berners-Lee’s original proposal for the WWW at CERN.
    Berners-Lee talks of the non-tree nature of communications and collaborations transcending the tree-like organisational structure of CERN, and uses this as a motivation for storing and navigating data in the same way. It’s interesting to note how this vision has often been lost sight of in a bid to squish all our data into large hierarchies, but we can see the process being reversed within weblogs at least.
    Oh, if we want to read more stuff like the Alexander article, where do we look?

    Like

  2. Dan Avatar

    Cheers Tom.
    How about more Christopher Alexander? Take the plunge (financially as well as mentally, I’m afraid) into his classics:
    The Timeless Way Of Building
    A Pattern Language

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: