Cabinet of Curiosities

I’m of the first generation that grew up with computers. This doesn’t mean that I was born with an innate ability to use them, but being around something fosters a sense of acceptance and curiosity towards that thing. To see what I mean, give a two year old an iPhone and see how quickly she […]

I’m of the first generation that grew up with computers. This doesn’t mean that I was born with an innate ability to use them, but being around something fosters a sense of acceptance and curiosity towards that thing. To see what I mean, give a two year old an iPhone and see how quickly she manages to unlock it and then delete all your apps. At first her actions are accidental, but eventually she recognizes a pattern of cause and effect. I grew up learning how to learn software.  My father had a range of CAD software installed on our home computer and I constantly played around with these tools for no reason other than my own amusement. Ashlar Vellum (the 80s version) was my prepubescent gateway drug to digital design and the computer simply became part of how I experience the world.

During a final review in graduate school, a well-known ‘environmentally-responsible’ architect told me that my project would suffer from inadequate daylight levels due to the strategy I had employed. This came as a surprise to me because the ‘problematic’ spaces were rather small with large amounts of north facing glass, but I assumed that the critic must’ve been right, given his collection of completed projects and, after all, I was just a grad student whose projects were small enough to fit in a jump drive. I later studied the project in a lighting simulation course and found that the ‘problematic’ spaces received plenty of daylight and, if anything, they were better than the ‘unproblematic’ spaces because of the more evenly distributed northern light.

The point isn’t that the critic was wrong, though he likely was, but rather that the success of a design is based upon decisions made in response to collections of information. In some cases, we don’t have all the necessary information or we don’t fully understand the inherent relationships so we fill in the blanks by throwing darts with our intuition. As I start my career in architecture, I’m left wondering what influences the accuracy of my intuition: is it possible for me to develop it faster with better accuracy, and broader relevance?

Intuition was first defined for me as an undergraduate psychology major and was introduced through the theories of Carl Gustav Jung who thought of intuition as perceiving by way of the unconscious. Jung thought of the unconscious as being split into parts: the personal unconscious and the collective unconscious – both can help us quickly form solutions to novel situations. The personal unconscious relates to our memories and experiences, which are not immediately available to us though they can be accessed with some effort. The collective unconscious represents what Jung referred to as archetypes, or what we have come to think of as genetic predispositions. Biomedical engineering may someday allow us to alter or affect our collective unconscious, but for now we are only capable of interacting with and building upon our own past experiences.

What, then, are the appropriate exercises and vitamins for a stronger intuition? I think the answer partially relies on vitamin I (iterations) and lots of it, and a large collection of spectacles for viewing the world in different ways. Now here you have to keep in mind that I only have a BA in psychology, which means I know just enough to think most people are crazy. The other takeaway is that I was forced to take statistics so I can assure you that a sample set of one is not enough for drawing conclusions, but I will share my experiences anyway.

After finishing graduate school, I was hired by LMN and helped start LMNts. My time is split between R&D and projects, allowing me to explore new means of design in the context of real world situations and conditions. Some of these endeavors might improve the efficiency of how we work; others might open new design potentials, and some we haven’t figured out what to do with yet. In a way, the task is to assemble a cabinet of curiosities. Historically, these cabinets were collections of specimens gathered by explorers and they often served as a means for finding relationships between disparate objects. That speculation can be important in helping to develop a stronger intuition by broadening one’s set of experiences through exposure to novel situations and pairings. The work I’m involved with serves a similar purpose. We collect strange tools and processes, and then share these discoveries with others in the firm.

Past generations of architectural designers were introduced to CAD as the digital substitute to putting ink on multiple layers of Mylar, so their use of the computer tended to be limited to the media and its constraints. Our goal is to get designers to see how new tools can improve how things are done, and also to demonstrate completely new abilities that designers didn’t have access to before. The problems designers grapple with are increasingly more complex and trying to squeeze every last drop of efficiency out of old solutions won’t always be enough. We strive to look at these problems in new ways with the hope that alternative perspectives will help reveal new potentials. Does analyzing a problem using evolutionary principles lead to different solutions than the typical best of three?

A lecture I constantly think about from my undergraduate days was on sensation and perception where the professor began with a discussion about our five senses and how each sense represented a window through which we could look out onto the world around us. The trouble was, he said, that we were stuck sitting in the middle of that circular room which contained only those five windows. Were the windows big enough to see everything around us or were there large gaps in what we were seeing? I tend to imagine that LMNts’ cabinet of curiosities serves the purpose of reshaping or increasing the number of windows in that room of senses. In the digital realm, we can isolate a given window, play with the qualities of the perspective it offers and then reinsert it into our overall way of looking out onto the world of design. We also have the possibility of creating or adopting entirely new perspectives. Our cabinet has the potential for unlimited expansion. Over time we’ve adopted tools that have been created by others or created our own tools, and with each addition we gain yet another way of looking at a problem.

New and fresh perspectives are great but it can’t stop there. With each new perspective comes the need to understand what exactly you’re looking at. This is where I believe iterations become essential to the process of design and development of good intuition. The goal isn’t to just generate hundreds of iterations for the sake of showing how fast we can fill a hard drive, instead, the goal is to establish a process which contains a feedback loop where each iteration is judged against an evolving set of criteria.


My first experience with photography was using a digital camera. Before I ever took a class, I learned how to use the camera by switching it to Manual Mode and then incrementally adjusting the aperture or the shutter while trying different compositions until I got the image I was looking for. Along the way I filled the memory card with my mistakes. The beauty of that process was that I could learn from my mistakes immediately and either delete them to save memory or keep them because something unexpected had happened which proved to be inspirational. I developed a deep interest in this iterative process of setting the controls, seeing what happened, and adjusting as necessary.

Photography is still a hobby of mine and that iterative process of learning has stuck with me as I’ve entered the field of architecture. In the realm of architectural design, the photographic subject becomes the building type and context, and the design process serves as the architect’s camera. With a camera, you can switch lenses, add filters and change settings. In architecture, the equivalents become the tools, the consultant relationships, and the parameters we set within the design process. The task becomes capturing the best exposure and composition of the subject or designing the best building. If intuition is based on past experiences, then it would seem that the more experience of looking at a problem, the better.

For me, the static modeling process of tools like SketchUp or Rhino is akin to taking photographs with film. Effort is poured into creating a single exposure or iteration and then the model is assessed. There are points in that process where changes can be made but the scope of change is generally small and can require a significant amount of effort. This type of modeling is still capable of generating beautiful and well thought out designs but the feedback loop is relatively long, making it difficult to quickly develop a comprehensive understanding of a new problem.

My experience with parametric modeling is more aligned with my use of a digital camera, where a greater number of iterations through the feedback loop can be experienced in a relatively short period of time. What I enjoy about working in this way is the ability to adjust the model in real time and see how the relationships between parameters are affected. Complexity tends to be more manageable because of our ability to see the rippling effects of a change. If a designer is working on a solution to provide sufficient daylight, he must also consider the implications of visual and thermal comfort, aesthetics, thermal performance of mechanical systems, coordination with structure, and cost effectiveness. Few problems are influenced by only one dimension, so it is critical that designers have tools that enable them to explore the interaction between both objective and subjective variables of design.

While in graduate school I focused a fair bit of my time on learning simulation analysis and parametric modeling. The parametric modeling makes it possible to explore a vast number of iterations while the simulation tools offer an additional way of judging those iterations. With those possibilities, consciously designing the design process becomes as important as designing the actual building.


The experimental design process isn’t new. Antoni Gaudi’s design process used hanging chain models to shape the structure based upon the path of forces. Frei Otto used soap film to simulate tensile structures. What has changed is that architects are increasingly able to do the same simulations using computers and instead of running a few options, they can run hundreds. If you can now run hundreds, then what is important to study?  What should change?  What remains constant?  The answers to these questions will vary from one problem to the next and likewise the design process will also need to vary. As Dan mentioned in his post , experimental design is a crucial aspect of addressing goals of sustainability. There is a new generation of designers who are beginning to engage this challenging problem and the tools they have access to are capable of providing useful insight. But do they have the experience to know the right questions?

As part of that generation, I feel comfortable saying that we don’t know all the right questions. While there are certain types of experience gathering that can be accelerated—for instance, a basic understanding of daylighting—there are also experiences that must be gained through the construction of buildings, assessment of their performance, and relationships with collaborators. The acquisition of digital skills can take years, and school is a great environment for this experimentation and skill acquisition. The digital generation is now bringing these skills to the workplace and, in collaboration with the pre-digital generation, tailoring them to the design problems of the new millennium. And in so doing we get to teach old dogs new tricks through what has become known as reverse mentoring.

While the pre-digitals should not be expected to know how to operate the newest tools, it is critical that they understand what the new tools are capable of and engage with the younger generation by asking questions, proposing applications, and helping define new design processes that integrate these computational tools. I mostly taught myself how to use parametric modeling software and what I couldn’t figure out, I learned virtually through tutorials, looking at others’ files, and browsing discussion forums.  But skill acquisition is only part of the challenge because once you have the skills you have to figure out what to do with them. I was fortunate to have professors who, even though they might have lacked knowledge about how to use the software, were still willing to discuss possible applications. Not everyone was always this open and at times it seemed that the usefulness of the tool was denied due to ignorance of the possibilities.

Fortunately, LMN leadership sees the benefit in exploring new technologies and is directly engaged with our exploration of the new tools. George doesn’t know how to model parametrically, but he is interested in how these tools will change practice and, I’ll admit, he comes up with some interesting applications. In the future, I imagine I’ll be in a similar position where I don’t know how to use all the tools of the next generation, but I’m excited about trying to discover their potential. I look forward to that day, but hopefully it doesn’t happen too quickly.

Scott Crawford is a design technologist and architect at LMN Architects and a founding member of LMNts.

This post is the 3rd in a series on the Seattle architecture firm’s response to the changing landscape of Design Computing.  For the background story, see George Shaw’s Re-Upping on Design Technology, and Dan Belcher’s Field Notes: Confessions of a Design Technology Evangelical.

Recent Viewpoints