Saturday, December 09, 2006

Testing theories of architectural intelligibility

Architecture involves designing spaces that are intelligible: discerning the purpose of a structure, being able to find the entrance, knowing how to get around. Courses in architecture aim to impart the theory and practice of designing intelligible spaces.

3D simulation software is commonly used to construct building models, and can be used to test whether people moving through them can, in fact, make sense of them.

Rather than just test the building for intelligibility, one can test the underlying theories by creating virtual spaces that instantiate them, with a “volume control”: an experimenter can adjust the degree to which a rule is implemented to find the point at which a user can now longer make sense of a building.

Think of it as usability testing meets architectural theory (feng shui, Christopher Alexander, New Urbanism, and on and on).

By the “if I can think of it, someone’s already built it” rule, it’s certain that this has already been done. If you know of examples, Dear Reader, please let me know.

The expensive way would be to use commercial architecture design packages; a quick and dirty approach could use Second Life. The challenges include (1) extracting variable-based rules from architectural design principles, and (2) building the volume control functionality.


Bret Battey said...

Hi Pierre, thanks for the thoughts on intelligibility and on wicked problems. I'm glad you have time to read this stuff and point me in the direction of yet more great books I wish I had more time to read...

This and the previous post set me to thinking about intelligibility, the ability to measure it, and how this might relate to art-making.

As a teacher of composition, including algorithmic techniques and avant-garde aesthetics, I'm often ruminating with students about this borderline between the intelligible and unintelligible. In the design of reactive-art, there seems to be a sweet spot between trivial directness of control and incoherent unpredictability. How to find it? And as a composer, (presuming traditional expressive goals), one has to be consciously aware that what constitutes intelligibility for oneself does not necessarily constitute intelligibility for an audience. How to step back and be objective about one's own work? (Or should one?)

So thinking about testing intelligiblity through an algorithmically-variable system... I wonder though, almost like wicked problems, if one can only base such systems in the case of known (and hence in some sense trivial) aesthetic bounds. I'm thinking of a quote from composer Herbert Brün once paraphrased to me as: "I write pieces that I haven't learned how to like yet." If we build an algorithmically variable architecture assessment system, the choice of variables will be based on presumptions about what variables are meaningful. So if the system becomes strongly used, it would preclude architectural solutions/innovations that require breaking of the assumptions.

Hmmm... I'm not usually one to groove on the (often self-pleased) rhetoric of creatives who declare their work as necessary to break closed systems, but I realize I just argued something to that effect!

Riding on the line between intelligibility and unintelligibility, -=Bret

Pierre de Vries said...

Thanks for stimulating response, Bret. Your point that an algorithmically-variable system would only delineate known, and thus trivial, aesthetic bounds is a usefully provocative one.

One response is that a system with sufficiently many variables is likely to find unexpected boundaries. With only a handful of variables, and smooth variation, practitioners probably wouldn’t be surprised by where the edges lie. However, the transitions between clarity and ambiguity in a high-dimensional space are likely to unexpected.

Even with relatively few variables, one is likely to find that the boundaries vary wildly for different people. While it’s a commonplace that something trivial for the creator could be incoherent to many in the audience, one might find configurations where the creator is surprised by how intelligible a variation is to the plebs.

Finally, one might be able to explore useful new ground even with a few variables if they’re recombinant in the manner of genetic algorithms. Imagine that the model is driven by meta-variables that are combinations of sub-variables. Changing the combinations will the way the model behaves. One might even find “intelligibility vectors” in the aesthetic space be recombining meta-variables until they give the accepted answer for a given cultural milieu.

P.S. The Herbert Brün quote reminds me of what I apparently used to tell my mother at bed-time: “Tell me a story you don’t know.”

P.P.S. Sorry about the tiny text size of the comments. It’s a bug in the template, not an assertion of authorial dominance over commenters!