The ACADIA 2010 Conference was held at the Cooper Union and the Pratt Institute in New York/Brooklyn last week. The conference was organized as a series of thematically grouped presentations of peer-reviewed work with invited keynote speakers in between. I have never been a fan of concurrent sessions at conferences – especially the small ones like ACADIA – because you inevitably end up having to be choosy. Most of the peer-reviewed sessions were split up between The Great Hall Auditorium on Cooper Square and the new Morphosis building across the street. Two very cool venues, but you simply can’t see everything. However, one does gather up the threads of the other sessions from conversation over drinks, dinner, and during walks through the city. Though the overall theme of this year’s ACADIA was contemporary architecture’s stance between the biological and the technical, I can safely say that this year’s conference was dipped in a theoretical stew.   I don’t want to focus exclusively on the theoretical aspects of the conference, but I think it would be misrepresenting ACADIA 2010 – considering the successes of the last two years’ events – if I simply passed over in silence. Let me explain…

If the patron saints of last year’s ACADIA were Gordon Pask, Norbert Wiener, and J.J. Gibson, then this year was dominated by the worship of Gilbert Simondon, Bernard Cache, and Gilles Deleuze. Don’t get me wrong, I can stomach quite a bit of theory, even of the continental variety.  When I come to ACADIA, I do expect a certain amount of the use of the terms performativity, milieu, topology, etc.  Georges Teyssot actually apologized (in a way) for “bringing us back to the fold,” for bringing us back to the plexus of Deleuze.  The fold was covered adequately by Teyssot, but we returned to it over and over again, sometimes topically, sometimes not.  It was not the subject matter that bothered me, but the way the theory was used.  There was a tendency for some presenters to simply unpack one piece of theoretical furniture after another, not really arranging them or inviting the audience to sit down.  If you are going to engage in theoretical discourse, please ground it firmly in the topics of design and computation.  This is ACADIA;  I did not come to ACADIA for a hagiography.  And please, get your facts straight.  If you say: “In the two hundred years since Nietzsche…” (Nietzsche died in 1900), or “regarding the topo-morphological restraining mechanisms circumscribed in the 16th century work of ____”  my mind starts to wander off, sometimes wondering: How did this person get tenure?  Occasionally, I would snap out my stupor to hear the speaker say something like:  “…and then there is the question of foam…but I cannot go deeply into this subject because I am already out of time.”  Tant pis. I have never been terribly sympathetic to the Sokal affair, but I can see where his frustrations stemmed from.  Productive conversations were often derailed by spherical and hyperbolic speculation.  Tant pis, encore.

The highfalutin’ discourse wasn’t all bad. There were moments when the academic and the practical crossed in productive ways.  My favorite of these moments was during a round-table exchange between Antione Picon and George L. Legendre. Legendre had just finished presenting his stunning work on the Henderson waves bridge project in Singapore.  Legendre prefers using mathematical expressions instead of scripts or code to design complex structures.  It looked to me like he works in Mathematica or MATLab. (Wouldn’t it be cool if you could just e-mail your structural engineer a formula and call it good?).  After Legendre’s presentation, Picon asked (and I’m paraphrasing): “I detect a sort of niche-market Ruskin-esque refusal of the technical in your adherence to formal expressions instead of opaque tools. Is that accurate?” Legendre thought about it for a moment, smiled, and said: “Yes.” (Of course he went on to explain why, but it was a nice moment of clarity).

Toward the end of the conference, during the closing sessions, there was a frank and polite push-back (from Mark Cabrinha, Rob Woodbury and others) against the domination of  the theoretical over the digital.  The message was clear: theory has a critical place (no pun intended), but a balance should be struck.  Agreed.

Ok, now that I’ve gotten that out of my system, on to the goods (in no particular order)…

Rhett Russo presented some beautiful work from his recent courses using industrial CNC embroidery machines and Processing.  Russo’s students used Processing to “draw with code” within the budget constraints and the fabrication processes of CNC embroidery.  The mappings between Processing code and the embroidered patterns were striking, where the overlapping threads gave a depth to the surface and contracted the fabric in unpredictable ways.  Unlike a CNC router – in which the router head moves across a fixed surface – with CNC embroidery, the platform that the cloth rests on moves and the heads remain fixed – much like a sewing machine.  The machine embroiders from both sides of the fabric.  The resulting isometrics between code and material topology were very interesting.  The embroidery itself was beautiful.   During the Q&A, Rhett asked: “have we exhausted the limits of information as embedded in a material?”  I think this question was meant as a provocation or a challenge, but I’m not entirely certain after Achim Menges’ presentation (see below).  I wonder if any of his students decided to mimic the double-sided sewing behavior in their Processing code by using an additional z-buffer that inverts the standard top-left 0,0 of Processing’s draw() routine…it might have yielded fun results.

I really enjoyed the work of Christian Derix of Aedas R&D.  His excitement was contagious.  Derix’s work is deeply grounded in theory, yet free of jargon.  Theory becomes “subordinate to crafting.”  There is a bottom up approach that speaks fluently all the way up the design communication chain as possible: from designers vision to client understanding.  An example is the Vita Shelving System, a unitized generative shelving system that is based on Conway’s Game of Life.  The end-consumer becomes part of the design process by selecting certain constraints and unit types (using a custom web-based UI) to start generating a shelving system based on cellular automata.  Derix and AEDAS pay close attention to the interfaces that drive communication with clients and stakeholders, making their computational design approaches as inclusive as possible.  Design tools that “your grandmother could use.”

Vlad Tenu‘s work on Minimal Surfaces as Self-organizing Systems is well worth checking out.  Tenu was another person who engaged the audience through his verbal presentation and use of graphics, something that was lacking with other presenters who decided to read their entire paper.  He presented his method of using Processing to iteratively refine the mesh of mathematically defined surfaces and showed a number of beautiful physical models that he built of these explorations.  It’ll be interesting to see what he comes back with next year.

Flora Dilys Salim presented her work on UbiMash which allows for coordination of many different software environments, specifically looking at the connection between parametric modeling and physical computing.  After giving an overview of what UbiMash is, Flora showed a number of four day projects that were completed during the SmartGeometry conference in Barcelona.  Very impressive stuff given the compressed time schedule.

Nathan Miller of NBBJ presented on the subject of information exchange and collaborative design work-flows, an extension of the work he presented at last year’s ACADIA: an overview of the information flow during the design of the Hangzhou Sports Park.   He mapped out the paths of interoperability, translation, and conversion that were necessary across multiple platforms, offices, disciplines and cultures.  It was nice to see a practitioner step back and explain how things worked (or sometimes did not work).  As a result, NBBJ is developing on a method of tracking parametric state information for different design options (they call it a Feedback Cloud).  Parametric and performance-based work-flows are powerful, but they create data management challenges as options branch continually, generating more and more information and representations.  This is analogous to a SVN or a CVS in the world of software development.  NBBJ’s efforts on this front are similar to those of …

Kermin Chok and Mark Donofrio of Halvorson and Partners.  Set amongst their many other progressive approaches to structural design, Halvorson has developed a custom database that tracks critical variables and states (parametric, geometric, etc) of different design iterations.  Halvorson uses this database in their structural optimization processes to manage the large amount of performance data generated by working “at the velocity of architecture.”  To them, this database is indispensable.  Though data management may seem unglamorous, it is critical to avoiding the information overload associated with iterative parametric modeling and performance-based design.

Jenny Sabin presented her recent digital ceramics work.  She is trained in ceramics, and it’s interesting to see her work constantly returning to the craft.  She has been taking complex and dynamic data-sets and translating them into ceramic structures using 3D printers and clay.  Jenny was also a constructive voice on a number of round-table panel discussions that would have otherwise spun out of control.

If you have not done so already, check out the work of Achim Menges on Performative Wood.  Achim presented the progression over the past year (he presented at ACADIA 2009) showing the Research Pavilion at Institute for Computation Design in Stuttgart.  There is far too much of his work to unpack here, so I encourage  anyone to read his paper.  Menges and his collaborators are pushing our understanding of one of the oldest construction materials to an entirely new level.

In sum…
This year’s ACADIA certainly privileged the critical over the digital…and perhaps that’s inevitable considering how far the pendulum may swing back next year. I really like the ACADIA Conference, and I can only hope that this year’s missed opportunities and discursive distortions were an anomaly. Next year’s conference will be held in Calgary and Banff, Canada, with the theme: Integration Through Computation.




  • Brian Lockyear

    Thanks Dan for the Good, Bad and Ugly. I was unable to attend this year’s conference but am happy to have the short list of reading materials! Will particularly enjoy following up on the work of Mark Donofrio and Jenny Sabin. By the way, Donofrio recently joined UOregon. It is exciting to have him here in Oregon.

    Looking forward to Banff!

  • dbelcher


    Good to hear from you. I assure you, reading through the above listed works will be a pleasure. I wish I could have been in every session, so I probably unintentionally left some great work out of the list.

    I was happy to hear Mark Donofrio has moved to the Pacific Northwest.

    It will be good to have ACADIA (somewhat) closer to home next year.