Saturday, July 05, 2008

Slow design (part 2)

(Continued from Part 1)

Perhaps surfacing rationale as a resource for fostering creativity in design, has to be thought of more as something to tap into. As, counter-intuitively, a generator of creativity. It's the very attempt to do it, to have the different kind of conversation, using different words, tools, and methods, that can bring about creative leaps.

However, as I write this I think of the thousands of meetings, conference calls, instant messages, and shared web-screen sessions that I have participated in (as have most people working in software design, I'd guess). In nearly all there was no explicit method or shared display, except, sometimes, a pre-written document or presentation as the ostensible focus. No way to check or to know whether individuals were actually looking at it or even paying attention, although that could come out in the degree to which people express agreement or disagreement with someone speaking. Usually there are one or a few people leading the discussion and doing most of the talking. Issues are discussed, alternatives posed, ideas voiced, realizations occur, forward movement almost always made, and a record of varying quality written down. If it's a face-to-face meeting, usually sketching on a whiteboard, sometimes extensive and a major focus of attention, sometimes just a few boxes and arrows.

It's not that this approach doesn't work. It usually does, and often very well, or it wouldn't be the dominant mode of design conversation. And the often substantial pre-work that occurs plays into it as well -- documents, diagrams, presentations, prototypes -- which can have a great deal of craft and care put into them. Earlier meetings and pre-conversations also inform any later conversation. All of these require and result in engagement and shaping, and often the quality of the artifacts and their usefulness in the larger conversation is a direct result of the engagement, participation, and skill of the participants who created or gave input to them. Each meeting is the sum of all these previous conversations, meetings, and artifacts -- a hive of memory and thought that is present as a resource, even if it has little or no explicit form.

So there is no question that the above isn't effective. Perhaps a better question to ask (one that was not voiced explicitly at the workshop), is "Why do anything else?" Why do anything other than what we normally do? Generally speaking, there may be no need -- the normal works fine. It's only when it doesn't, when it's felt that something else is needed, or that something is missing, that we need to try something different.

It seems to me that design rationale can best be expressed as questions: Why are we doing this? What does this mean? Have we captured all the alternatives? Do we understand why we're chasing this one? Are we going to need to remember this? Even if these questions come up in the normal flow, as they sometimes do, usually the answers and the deliberation are evanescent, gone with the wind except maybe as shreds in the memories of a few of the attendees.

Using tools and approaches like Compendium to make this kind of conversation explicit, to engage in it and record it, to actively shape and craft it, is the largely untapped resource that can make the benefits of this kind of conversation deeper and more lasting. It works best when it's engaged in as a careful, intentional exercise, confined to special times, like a kind of heightened speech. It can be difficult to make the switch into that mode, but that doesn't mean it's not worth doing. Plenty of similarly worthwhile activities feel like they go against the grain and are hard to get ourselves to do -- like meditation, or writing blog entries on topics like design rationale.

Setting aside a given amount of time -- say 45 minutes in a day of design conversation, even spread out in several 15 minute increments -- where we work in this different manner, carefully and explicitly forming questions, alternatives, pros and cons -- saying to each other that "we're not going to leave until we've done this" -- creating an intentional artifact that requires us to work, look, and talk differently for a little while -- can enrich both our designs and our thinking about design. At minimum we'll have the kind of record that normally doesn't exist, but we might also gain the hidden benefits that come from engagement in such collaborative shaping of meaning.


Erik Stolterman said...

Hi Al

I did have a chance to participate at the workshop unfortuantely, so I missed all the good discussions. I liked your posts! I have at several occasions tested a simple idea. If I call a meeting for a "design session" instead of a "meeting", it completly changes the process and outcome :-) And it is in most cases also more true, sinc emany meetings are about designing something...

Erik Stolterman

Al said...

Thanks, Erik, and sorry we missed you at the workshop.

Without disagreeing with what you're saying, here's what I was thinking about when I wrote this. I was mostly (though not only) picturing software design in the context of corporate/large IT organizations, especially when much of the work is carried about by distributed teams that rarely -- or never -- meet face to face. There are many sorts of meetings that are referred to as design sessions (e.g. "JADs") that differ very little from other sorts of meetings, regardless of what they're called. The form of the meeting itself, and its activities, follow the same template (document, conference call, discussion, etc.). These can (sometimes) be exciting and generative, but I still feel that often something deeper is called for, especially if "creativity" is a felt need.