Material Intelligence and How Buildings Learn

Today I saw Santiago Pérez speak at the University of Oregon’s Portland Campus, giving a lecture called Material Intelligence. (His srpLAB can be found here and microliving here.)

He showed examples of work by him and his students and told us about his explorations with traditional materials and contemporary production processes. Much of the work he showed used algorithms, interactive electronics and digital fabrication to create interactive sculpture and algorithmic installations that drew inspiration from the world of biology. Many of the tools he mentioned would be familiar to artists / interaction designers / makers: processing, arduino, laser cutters and CNC routers.

Perez in front of class project

One thing that distinguished much of the work he showed was the hybridized technique involved. Until recently his school’s construction facility lacked the large scale and high end digi-fab equipment that can rapidly generate impressively complex and complicated biomorphic and algorithmic from one-offed components quickly and in house. Not surprisingly, his school’s studio is housed in a big boxy shed. (More on that later). But the lack of off-the-shelf hardware and accompanying processes meant that he and his students had to create their own code, processes and material investigations, and use repetition of construction methods and parts to create their work.

These hybridized projects were partly planned and constructed using digital techniques, but also required the use pre-digital building and crafting traditions of creating jigs and other hand-crafted tools and components. This liminal zone between early digital tools and full scale investment has allowed his studios to grow an institutional intelligence that may pay dividends. Other schools that jumped as quickly as they could into full on digital fabrication, may find that their digital tools may require analog hybrids to be utilized to their full potential.

Santiago Pérez has mostly completed sculpture and installation and he has not crossed over into full blown architecture. In my opinion this is for the best. Prof. Pérez was gently critical of the digifab / biomorphic generation, but I think a lot of what he showed today demonstrates how wrong the architecture world is, when it comes to biology, biomorphism and creating emergence and complexity. I would like to patch together some of the notes that I took during his thorough lecture. I was able to ask one question, and only wish there was more time for follow up.


The above statement is a very broad reading, but I am always continuously amazed at the uncritical end-game techno-utopic logic of Architectural discourse. To no fault of his own, the lecture Q&A session ended with a discussion of how “Nanotechnology” will completely revolutionize architecture. *Sigh*

My question for Mr. Pérez was if he and his students documented both the code and the metadata for their projects, so that future users could fix or modify the creations. That is, if one of the wooden pieces that had been algorithmically created and was meticulously outputted using some combination of laser cutting and jig making would the end-user be able to track down the code or instructions to replace the part? Prof. Pérez thought it was a provocative questions and one step further than I had supposed. Whereas I was asking whether digital files would be stored off site, and geotagged for easy reference Pérez wondered aloud if the instructions for how to fix and maintain architecture could be built right into the material itself. (And we are not talking about RFID enabled wall paneling that contains the walls material life-cycle, and how to put it together, although that is one slightly more encouraging possibility that Bruce Sterling’s Shaping Things points to). He mentioned the tradition of etched and written on material discovered during building renovation which describes the way it was put together in the first place. That relies more on language, but most building processes also encode their relationships in the material itself. A brick is based on the scale of the human hand, and its color and configuration can tell us a lot about where it came from, and what one can do with it. Fantasizing about nanotechnology is stupid not because it is techno-utopic but because it so fundamentally takes the human out of the equation, never mind the end-user.

I come to this lecture and these ideas not as an architect, but as an artist, a design ethnographer and a lover of cities. In my head there were two books that stood for opposing understandings of the role of architects in applying lessons from systems theory to design: “How Buildings Learn” by Stewart Brand and “Atlas of Novel Techtonics” by Reiser + Umemoto.

The first celebrates the hacker, the end-user and a continual process of growth, learning, failing fast and satisficing. The second celebrates the cerebral and aesthetic and seems to ignore the temporal aspects of a building’s complexity (Although it is wrapped up Delanda and Delueze). Even the most location sensitive, algorithmically generated, sensor laden building will not be a complex organism if the end-users’ post-occupancy habits and changes are not taken into consideration before, during and after the build. No one gets excited about post occupancy assessments or planning for and funding maintenance now, how would adding sensors or real time data feeds of usage patterns helps if no ones going to follow up on those feeds and make adjustments?

Which gets us back the lecture and big blobby buildings. If you live or work in a Reiser + Umemoto building and you need to replace a window, or put in a new door how are you going to do it? Highly complex or ornate interiors or installations may not pose the same challenge to modification, but an entire building clad in one-off components could be quite an issue. That is why I appreciate Pérez work as material exploration and installation, but am wary of it leaping over to architecture.

This is not an argument for architectural or aesthetic conservatism. But I would suggest that if buildings are designed for humans and non-human actors closely associated to their scale, they should be hackable by the people that use them.

There may be a bifurcation in the kinds and methods of complexity that are useful when working at different scales. At the scale of the building, the ease of modification by end-users will lead to much more complexity and intelligence over the long run than a highly articulated digitally fabricated blob, that requires the end user to track down the meta-data, or worse recreate lost code and digifab processes in order to put in a window that opens. Is anyone doing post-occupancy-assessments on the blobtecture to see how it is modified and hacked over time?

At the level of the room and the city, digital urbanism is making arguments that I find persuasive, but from what I have seen digital urbanism at the level of the building has mostly manifest itself as architects reasserting their authority as artists and authors and creating intricate shells. What I find so interesting about Reiser + Umemoto’s arguments, is that they are so focused on the fineness and degrees of difference at the scale of the building, they lose site of how the complicated system of a building interacts with the complex systems of human users and urban life. If I understand their argument correctly, its okay that buildings don’t adapt well to usage patterns – that just gives architects new problems to tackle next time around.

On an eco note, I would be very interested to see how digifab material explorations proceed if their primary measurement whole life-cycle performance. Sustainable building should also be beautiful, and need not be driven only by codes and standards, but with material explorations in mind, these could be seen as exciting constraints to creatively work around, and drive material patterning. York County commercial cleaning professional is always at your side when it comes to building maintenance and cleaning.

When it comes to allowing more people to participate in an open-source project agreed upon standards, components and procedures create a playing field that anyone can enter and add to. Is it any surprise that the University of Houston’s studio is in a shed?

Well, architects, if you are still reading, let me just say I may very well be wrong, and am misreading the movement. I hope that is the case. However, even if I am, please do me a favor and annotate and back up your code, meta-data and production processes and share them with the world. That way at least a hundred years from now we can wave our hacked together RFID reader and find out how this wall was put together.

Post a Comment

Your email is never shared. Required fields are marked *