Monday, April 14, 2008

Iterations

So I was talking with my partner last night. She'd just read the most recent Vita Activa post (A Contentious Vocabulary) and she was still trying to figure out what I meant by "design" and "craft." So we sat and talked for half an hour or so, and I'm taking another crack at it today. This is the iterative process that designers always talk about, but anybody worth their beans does it all the time, regardless of profession or hobby.

When we're creating something for use (I'm still sticking to my differentiation between art and craft for this discussion), it seems to me that we're actually doing three things. To use the academic terms, we could call them design, craft, and research. But I think those are analogous (maybe synonymous...stay tuned) to the words plan, do, and test.

Here's an example from the profession I understand best: conducting ethnographic research. I spent about a year designing/planning my biggest research project before going off to do fieldwork. That entailed the conceptual literature review, the methods literature review, the coordination with the host site, the calendaring of research tasks, the writing of the proposal, and so on. The work of "research design" (a very common term in academic circles) was intended to both frame the problem to be resolved -- our misunderstanding of teenagers' spatiality -- and to create an appropriate plan for resolving it.

Then I went off and actually did several hundred questionnaires and dozens of formal interviews and a few thousand hours of direct observation and all the writing and puzzling and sorting and transcribing that went along with it. This stuff, the everyday doing, was the actual craft: taking the plan and making something real out of it, hopefully with care and attentiveness.

The final step was the evaluation of the work, testing whether it did indeed adequately address the original problem. This was (at the largest scale) the peer-review process whereby a fair number of experts, both known and anonymous, reviewed the work to understand whether it a) was competently conceived, b) was competently conducted, c) led to a new understanding of teenagers' use and conceptions of the physical world, and d) represented a productive change in our thinking. To use my earlier definition of "research," the question my reviewers had was to learn what was true about the specific circumstances (in this case, the book and the intellectual field it lived within).

Plan. Do. Test.

When I play pool, I have a much more constrained problem -- a tame problem, Rittel & Weber would say. I have to plan how to move the cue ball to accomplish three things: make the object ball I want to make, have the cue ball end up in a location that will allow me to sink another object ball on the next shot, and have the likelihood of being able to productively move the cue ball to another good spot for the subsequent shot. I'm thinking two shots ahead, usually; someday, I'll be good at this, but for now, two is what I've got. So I have to plan the angle, speed, and spin I need to impart in order to get these problems solved.

Once the design is in place, I chalk up the cue, lean over, go through my pre-shot routine, and stroke the ball. That's the craft. Once everything stops rolling, I assess whether I solved the problem appropriately, which in this case is easy to discern: I did or did not pot the ball I wanted to pot, and the cue ball did or did not stop where I wanted it to stop.

Plan. Do. Test.

Now, it's easy to understand all of this at the biggest scale. I spent a year planning my big research project. I spent a year and a half doing it. And I've spent the last eight years evaluating it. But the fact is that I was doing all three of those things on the small scale every few minutes. I'd be hanging out at school between classes, and I'd see something happen that seemed to me to be a useful thing to explore ("useful" meaning that it applied somehow to the bigger problem, which I was redefining every day as I learned more about conditions in the field). So I'd figure out how to spend the next few minutes asking about that event I'd seen: manufacturing questions on the fly, testing their utility by thinking about the answers I was getting, and creating new questions in response to my evaluation. It was plan, do, test at the finest grain, over and over and over again.

When the iterative work is that rapid and reflexive (and that sequence lies behind every good conversation you've ever had), it seems as though it's all one thing. But I think for our purposes, it's important to remember that there are three unique acts going on, because they have different ends and therefore different terms of critique.

So what about architecture? And more to the point, what about my partner's questions from last night? Design is the process of framing the question, making the question complex and difficult (and thus worth answering), and developing a plan for resolving the question. For instance, good design doesn't answer the question "How should I make this school?" It answers the questions "How should we educate these kids? What does it mean to be educated? What does it mean to be an educator? And how can a building (or perhaps not a building) help answer these questions better than we do now?" And lots of other questions as well, having to do with budget and energy and legality and safety and so on.

Once we have the plan, we embark on the crafting of the response. We draw and model and add and subtract and turn and separate and combine and select materials and change those materials. Eventually, we have the object for which we are responsible: a construction set to give to the contractor. And then we turn to research: does my product (the CDs) appropriately resolve the questions we raised? Does the ultimate building foster education in the ways I planned for it to? What do the kids and teachers and administrators and test scores say? Are there new questions to turn to for the next design? Or do I need to rethink what I did this time in light of its lack of success?

That's design-craft-research at the big scale. What does it look like at the small, day-to-day scale?

2 comments:

Alan Roberta said...
This comment has been removed by a blog administrator.
nora said...

So I will now "out" myself. I am Herb's partner, and I have a vested interest in responding to this post, since he has reported (accurately) our conversation. Hopefully you will forgive my jumping into THIS "conversation."

I think this is an interesting tripartite or quadra-partite (is that a word?) conceptual approach: Plan, do, test or design, craft, research with iteration thrown in to make us reassess the effectiveness of the other three components. It is an effective approach to examining the design process.

But I frankly think it is only one way to do this, and I believe it is a better representation of Herb's cognitive approach to problem solving and education than mine. For me, I think the process is different (for better or worse). I can not carve my problem solving into a 3-phase process. I think that is a reflection of a much "messier" thinking process than Herb's.

With a problem to solve, I am interested in a different kind of solution, but then I have been trained in Environmental Psychology not Architecture. I don't think in terms of objects / buildings as much as I think in terms of human behavior. My "design process" to use Herb's definition is one of seeing a socio-cultural problem that needs to be addressed-- community cohesion, affordable housing, the need for sustainability etc. When I do research I want to understand how people understand these problems - and how the problems interact e.g. how the lack of affordable housing stems from the building materials used, and the impact of that deficit-in-housing and materials-cost on community cohesion.

Herb wrote: "To use my earlier definition of "research," the question my reviewers had was to learn what was true about the specific circumstances (in this case, the book and the intellectual field it lived within)."

I think of research as responsible for addressing social change and therefore reviewers as responsible for assessing the impact of the "data" or "analysis" on CHANGING the community - not simply informing the intellectual field and the book itself. In classical experimental design, it seems that Herb is assessing the internal validity of the problem solution and I am committed to the external validity or generalizability. Can we translate what we know about teenagers from Herb's study to teenagers elsewhere? Can we extrapolate to other marginal populations and their use of space?

A mentor of mine once said that any good research study ends with more questions than it answers, and I believe that. Hopefully they are new questions or new ways of defining the problem. That's the iteration. And it is critical to the design of spaces in which we live, work, heal and play.

But as a messy thinker, and as a hand spinner who takes the fiber from the animal and cleans, cards, twists, knits it into something to wear (rather than as a pool player), I make a much less clear distinction between the design and the craft. At any moment, my fingers need to respond to a stray bit of chaff, a mat, a weakness in the fiber that causes it to break (because the animal was sick some months earlier). Is that design (as in something that I planned?) or craft? I think the product is constantly informed by the iterative decisions on a micro-second by micro-second basis that allow us to reframe the question in response to a changing context and to our own learning. It's like learning to stand. We can only stand if we can respond to the micro-second changes in balance needed to adapt to changing environment and body position.

For me, the three part distinction is an interesting IDEAL, even goal, but my reality is much messier.