Christopher Alexander on Pattern Languages

In this presentation, architect Christopher Alexander spends quality time with an industry well outside his own at the time — software. The conference captures Alexander and his audience somewhere in the early-to-mid 90s, when the globe began digitally stringing itself together, adding a new layer of complexity to increasingly wicked problems. Pattern Languages were a matured design theory, benefitting from two decades of being put to the test and finding fresh life in an industry that was quickly building whole new worlds. Alexander can't see this at the time of the talk, and he struggles to understand where the interest and intent lies from this crowd of practitioners. He notes that their intent possibly lacks a moral foundation (Under what circumstances is an computer program “good”?), which he claims were foundational to pattern languages. Rather, this technical crowd was using the “format” of a pattern — context, evidence, problem definition, plus a few possible solutions — as a way to start a dialogue of ideas. It's interesting to observe how Alexander brushes off the effectiveness of format, calling it a neat idea, in light of the current era of design thinking in which one could argue that format is the very thing which the larger arena of business finds so unique and invigorating.

As he continues his talk, he wonders how far the crowd has pushed the idea of patterns. Have they taken their ideas and work to the level of actually making human life better? How would they measure the impact of design? At the time, the software profession was scratching the surface. Now, we're seeing the affects.

He also notes that a part of the process of “writing” a pattern language was to consider to what extent the language would produce a coherent whole, an entity. Could a community use the language to design and build structures that felt good and were “more whole in themselves”? (See The Oregon Experiment.) They didn't always succeed.

Learning From John Rheinfrank