Tag Archives: computer design

Architecture Computation Theory Writing

Defining “Building Information”

Students at Yale’s Architecture School are invited to propose topics for an annual journal named Perspecta. I co-authored a proposal with my classmate Raven Hardison intending to expand the theoretical background of building information modeling. I find that the discussion over its uses is largely focused on building process improvement, which is itself valuable, but only a narrow band of its potential. The nature of technologies as disruptive and immature as building simulation and information modeling is that we haven’t yet begun to imagine uses for it. See, for example, this week’s article about the Kinect hacking culture in the Times.

Since we were not selected, it seems in the spirit of things to publish the proposal in hopes that it can germinate other ideas in topics we ourselves were blind to. The original proposal included the names of proposed authors, which have been edited out.



A Failed Proposal for Perspecta 48:


Building information modeling has become fait accompli in the discipline of architecture. Sales of Revit subscriptions steadily rise, while technical facility becomes commonplace. Resignation to BIM’s inevitability has caused a critical blindness toward the nature, forms, and influence of those first two words,building information.”

For Perspecta 48, we propose to step back and investigate unresolved aspects concerning the use of knowledge in design and construction. As a metaphor for the discipline’s approach, consider the parable of the blind monks and an elephant.

A king brings a pachyderm to a few sightless monks as a test. He asks them to reach out and describe this “elephant” they had never before encountered. Unable to perceive the whole creature, each groped at different body parts and returned a correspondingly different answer, all rooted in something familiar. The tail became a whip, the body a granary, the tusk a plow, and one man declared the leg to be a column.

When the king revealed the elephant, he showed their wisdom to be burdened by two key epistemological errors: assuming one has the full scope of information on a subject and, secondly, relying on preconceptions to understanding new information. By approaching the issue from a business side and avoiding theory, the discipline has ceased examining information in a multifaceted way, as it once did.

Forty years ago, researchers like Charles Eastman and Nicholas Negroponte imagined software reminiscent of information modeling and parametricism. Over the next forty years they worked to realize these dreams. But at the same time, the scope of the theoretical aims narrowed from world order to interface to tool and finally to product. In order to produce viable software, designers had to cut back their ambitions to basic commercial goals. Speculation on the radical political uses of data or how digital tools transform our relationship to the physical have been left to coders at the margins.

To reorient the debate, Perspecta 48 looks at “building information” from five key angles. Taking a fundamental perspective, the authors in the first section will examine the epistemology of building. By this we mean what is considered constructional knowledge, and how the discipline organizes it. Another group will look at the ways simulation and dematerialization have transformed humans and their environment.

A third set explores the diverging ends of authority and participation on that building modeling is driving the profession to. Looking at the politicization of knowledge, one group will examine how information multiplies of power and raises new ethical dilemmas. Finally, one collection of essays will discuss the technical difficulties of realizing architecture with digital technologies, and even how to embrace them.

We aspire to reinvigorate the discourse about building information by revealing it to be not merely a technology, but rather as a fundamental element of architecture.


The proposal continues.

read more »

Architecture Theory

Drawing is Dead, Long Live Drawing

This past March, Yale hosted a conference on the role of drawing in architecture in an age where most design occurs in the head and on the computer. “Is Drawing Dead?” I don’t have too much to add, but if you have a few hours of thoughtless labor, listening to it can be surprisingly informative and you don’t really need to see what’s going on. The third session, “The Critical Act,” is much more oriented toward architects themselves.

The highlight of the series, for me, was (go straight to it) was Andrew Witt’s discussion of the much longer use of computer drawings than the architecture profession typically admits. Witt is director of research at Gehry Technologies, and spent a few years studying 19th-century mechanical tools to reliably draw the complex shapes desired by Beaux-arts architects but very challenging to obtain with the accuracy or precision needed to actually construct a building. So it’s a very interesting talk. Patrik Schumacher, on the other  hand, does nothing but embarrass himself and bloviate.

Here are links to all of the sessions, each three hours long, except for the keynote, #4.

  1. The Voice of Drawing: The history of and an apologia for hand drawing.
  2. Burning Bridges, Questioning Practice: New technologies and scientific developments in design.
  3. The Critical Act: What are we trying to do when we design?
  4. Real is Only Halfway There: Peter Cook on how architects draw for each other.

The answer to the question, BTW, was that drafting is dead, but sketching will be around as long as we have bodies. Seems simple enough, right?


ConX: Modular steel construction

ConXTech is a Bay Area company that has finally brought the steel frame into the 21st century.

Now, that phrase trite, but what they have mass-customized the design of steel beams and then greatly simplify the assembly. It reduces design time, reduces the amount of labor needed, reduces energy expenditures, and provides a sturdier and more flexible alternative to wooden frame structures. Those stick-built structures are the most common design of 1-5 story residential construction, however, ConXTech is looking to make 5-12 story structures affordable as well. To do so, they have approached the whole building process in a slightly different way.

With ConX, the architect must involve the manufacturer in the design process, as the company custom builds the structure offsite in a factory. ConXTech needs to schedule the amount of work and materials necessary, while the architect needs access to their computer components at the schematic design phase. Still, it’s really not unlike the conventional process in which plans are sent to a structural engineer for engineering work, just that the engineer is the manufacturer too. Integrating the manufacturer reduces overhead, while entering the design development phase with a working model of the structure keeps the building lean and reduces the number of design changes that result from unexpected structural revisions.

So, once the project does reach construction, the CNC robots cut out pre-designed structure members to specification, as each piece is needed on the job site, delivering the product only as necessary. The system consists of vertical and horizontal beams, joining elements, wall panels, whole flights of stairs, and various other parts for specialized situations. Because the factory environment allows for meticulous control and use of accurate cutting, the frame and all other elements are extremely precise, resulting in better quality and speedy assembly. The frame itself is put together using snap-in-place fittings that a steelworker then bolts into place to meet code.

It’s a persistent myth, perpetuated by traditionalists, that steel is not a sustainable construction material, due to its embodied energy, which is around eight times that of wood. However, embodied energy, that is, the amount of energy required to make the materials and then assemble them on-site, rarely exceeds 20% of the total energy required for a building torn down after only 50 years. Steel frames overwhelmingly outlast that timeframe (steel-framed buildings from the 1880s are still around, and we have no idea how long they will ultimately last), so the percent of energy the frame requires diminishes in relation to heating, cooling, and interior renovation costs. They also use much less material, allow for greater density, and when they are torn down they can be wholly recycled, but I’m digressing here.