Thursday, November 02, 2006

Building (in) the Future @ Yale, Part 3

Written by Bob Yori.

Saturday’s sessions began with a series of speakers addressing Information Sharing. William Zahner, from A. Zahner fabricators, Hilary Sample of mos architects and Yale University, Kent Larson from MIT, and Chris Noble, a lawyer from Noble & Wickersham LLP. Cristiano Ceccato from Gehry Technologies provided the response to the presenter’s speeches and led the Q&A session.

Zahner’s company is highly regarded for its ability to build the skin for projects requiring sophisticated metal skins such as Morphosis’ Eugene Courthouse, Herzog & de Meuron’s deYoung Museum, and many of Frank Gehry’s works. He presented several, and talked about how using tools such as Pro Engineer and CATIA enable the models being made to be directly translated into built form.

The deYoung museum was a good example of the Architect and the Builder collaborating early in the project. The desired effect for the metal skin was to mimic the effect of light filtering through trees. Zahner and HdM worked together to develop a series of perforations varying in densities across the façade, enabling the light to filter through selectively.

More pragmatically, his collaboration with Frank O. Gehry Partners in the design of his ‘Pods’ project for the Ohr-O’Keefe Museum in Mississippi. Apparently, the cost of the steel structure combined with the cost of the metal skin almost cancelled the project. Mr. Zahner’s company worked with Gehry and “took the steel out of the project”, enabling the work to continue. Through their use of model-based collaboration and solutions, re-engineering the metal skin panel structure to also handle the buildings’ structural loads was possible. Unfortunately, the project has yet to be built, as Katrina ravaged the region before construction began.

Hilary Sample then spoke about what an “office” is. She reviewed today’s popular digital technology, and began to reinterpret the office in today’s YouTube/ Google/ Wikipedia/ Flickr world. Using tools such as these enables an ‘office’ to share information on such a scale and at such a speed that it’s truly in sync with today’s “interface culture”. She expands on the Flickr model of following a predefined set of trails (photos), to developing and posting your own via, and establishing your own connections via archinect.

She performed an analysis of SOM’s website and noted that there are very clear divisions between many of the disciplines and sub-disciplines that the firm markets to and contains. Her proposal is, from what I understood, to provide the antithesis of such a website. Interesting that her firm’s name (mos) is an anadrome of SOM. I’m sure it was her grand plan from day one. (disclosure: I work at SOM, in case you forgot)

Mos architects has developed a website that embraces this interactive environment. It encourages more transparent collaboration, and fosters the ability to work from a jobsite or client’s location. It provides access to both the client and contractor and enables communication via all the modern accoutrements of the web, such as image posting, tagging, chatting, and blogging. Project schedules are shared and coordinated, and the entire site is auto-archiving so that the entire process is as automated as possible.

I wish I could provide you with a link to the site to see it, but Archinect, Google, Yahoo, MSN, Altavista,, and Lycos turned up nothing. I find it a more than a bit ironic that Ms. Sample’s website of collaboration and interactivity doesn’t turn up as a result on major search engines and wasn’t immediately accessible. Perhaps this is by design? I wonder if, since the client’s projects are so easily accessible on the website, that it’s hidden from general view? It would make sense, but then one gets into the argument about the purpose of a firm’s website. SOM’s website, for example, is meant to be a portal to the casual browser as well as having a link to project specific information that’s protected via login. Wouldn’t that be the same thing as having a website that’s hidden from the general public? I would imagine her ideas about information sharing via the web would be best implemented in that ‘hidden’ mode, accessible by employees, owners, and contractors. It then begins to separate itself from the marketing side of the practice and becomes the ‘digital office’. Of course I could be wrong on all these counts, and the website may still be under construction.

Kent Larsen's comments focused on the pragmatics of making modular fabrication possible, as well as making the client/ owner more interactive in the design process. His presentation began with the statistic that Architects have no real role in 95% of the housing that is built in this country. While this is surely a debatable figure, I’m sure we could all agree that there’s a lot of housing being done out there without direct consultation of an Architect. So how do we fix this problem? (I assume, of course, that there are others besides architects that consider it to be a problem.) Kent argues that speculative development must no longer be a viable option for the builder, and that managers need to take control of the logistics of planning, whether it be urban, suburban, or rural.

Kent’s proposals are refreshing in that they seem so tangible, so easily implemented. He suggests that site information come in the form of digital point data, weather and climactic information, and other site specific data that enables a proper holistic approach to the parcel. In the case of urban environments, each unit would be delivered with a set of critical dimensions outlining location of services, critical dimensions, and the like. It would provide a consistent base of data for anyone to start planning their project.

A “Design Engine” would be crucial to involving the client in the earliest stages of designing their residence. He drew the parallel to the “build your own car” section of many automobile websites, giving Maserati as an example (an excellent choice if I do say so myself). The Design Engine allows the user to design/ configure their spaces. His slides showed what looked like a rigged-up light table that’s configured to respond to user input in the form of rearranging objects on the table’s surface.

Coupled with the Design Engine, there should be a ubiquitous search engine that allows the objects to be sourced immediately, and an iTune-like download process for accessing specifications and components. Builders are then simply assemblers of this “modular, disentangled plan”. Sounds good to me.

Going further, he proposed modular HVAC systems, and called for the need for a universal connection for services within these modular components that are to be snapped together on site. The construction equivalent of a USB port, this connector would allow disparate components from disparate manufacturers and sources to be combined easily. He completes his idea of the process by suggesting an eBay-like process for bidding on the assemblies, and gives Target as an example of a company who does something similar today.

Finally, Chris Noble spoke about the issues involving collaboration and intellectual property rights. He observed that today’s B141/ B151 agreements’ “Ownership of Documents” sections are very much rooted in a paper process in that they distinguish between originals and copies. This is all rooted in copyright law, and assumes that the Architect is the Author.

The fact is that architects are not the full authors of any of the “instruments of service”. And what about the specifications? Have any of you tried to get your spec writer to give you the Word files of their specs for a job? I’ve never had much luck with it. Between the specs, manufacturer data, and consultants contributions to those “instruments of service”, there are many authors. But it’s ignored.

The 1991 Architectural Works Copyright Act was, presumably, passed to protect the built work, not just the documents that define the work. For example, in principle, someone couldn’t duplicate the design of a building; under the Act it would be considered a copyright violation. But again, who actually is the “Owner” of the work? The building’s owner? The architect? The builder? All of this is unclear an open to interpretation/ litigation.

Mr. Noble sees the Federal Copyright Act to continue to be the basis for intellectual property (IP) rights. He also sees a continued dependence on contracts in the US, mainly because of the fragmented nature of the industry. There are many small firms (architects, engineers, consultants, builders) who aggregate themselves for a specific job. Because each of these organizations is separate from the other, what binds them together are the contracts. He also sees the need for tighter and more comprehensive contracts, because, in the future, we will shift from doing one-off projects to prototypes. Lastly, migrating collaboration to the internet will require legal language. The entire audience seemed to groan audibly when he suggested that when logging on to a site such as mos architects, one may have to click through an IP-like “licensing agreement” box, similar to when installing software on one’s computer today.

With that cheerful thought, Cristiano Ceccato presented his extremely brief response to the speakers. From his experience working in a digital environment, he’s seen the relatively new role of “3D Coordinator” rise to prominence, and argues for its centrality.

Diving right in to the Q&A session, his first question was about sharing models from disparate software platforms. With this workflow, there’s a need for translation from one file format to the other. While each player is responsible for the validity of his or her own data, who validates the accuracy of the translator software? Mr. Noble responded by describing the ‘defensive’ nature of the current transmittal structure, and hasn’t seen that the ‘validator’ really has a defined role yet. Mr. Ceccato answered his own question with the observation that Foster’s office shares raw numerical coordinate data and not models.

Adding another tough question, he asked about the vulnerability of data and the hackability of websites, and what could be done to address the potential? Kent Larson gave the most applicable answer, in that he recommended architects look to supply chain management software used in other industries as a model for security. He went on to posit that we don’t have “big players” in our industry, which inhibits rapid adoption of standards. He gave the example of residential window manufactures who haven’t agreed on something as simple as a common insertion point for their CAD blocks. This “big players” comment echoes Mr. Noble’s sentiments about why we’ll continue to depend on contracts.

It’s obvious that Larson has spent considerable time evaluating the logistics of standards implementation. He gives the example of initiatives in the UK and Australia, among others, that have successfully “flattened” the risk factor. I visited just today and am excited to see that standards are emerging, albeit not in the US.

Scott Marble (of Marble Fairbanks) was concerned that managing the 3D model would add extra cost to what is often a fee that’s too low to begin with. Ceccato responded with a brief “no”, explaining that he had worked on some very small 3D projects without any extra expense.

Finally, and I feel it’s important to note the dissenting voice, Bob Gutman (who later gave the symposium’s closing comments) pointed out that he’s seen more than his share of ideas designed to reduce coordination problems between architects and engineers, and so far, they’ve been of limited success. “Anything we can do to get people in the same room will fix the problems,” he said. His point? Technology is not the answer. Communication is.


  1. Nice synopsis..... great new addition to the writing staff!

  2. Now if we could get that Zoog guy to update a blog....

  3. hilary sample/ michael meredith website is