Design: what it is, and how to teach and learn it

XV.2 March + April 2008
Page: 57
Digital Citation

On modelingThe analysis-systhesis bridge model


Authors:
Hugh Dubberly, Shelley Evenson

The simplest way to describe the design process is to divide it into two phases: analysis and synthesis. Or preparation and inspiration. But those descriptions miss a crucial element—the connection between the two, the active move from one state to another, the transition or transformation that is at the heart of designing. How do designers move from analysis to synthesis? From problem to solution? From current situation to preferred future? From research to concept? From constituent needs to proposed response? From context to form?

How do designers bridge the gap?

The bridge model illustrates one way of thinking about the path from analysis to synthesis—the way in which the use of models to frame research results acts as a basis for framing possible futures. It says something more than “then the other thing happens.” It shows how designers and researchers move up through a level of analysis in order to move forward through time to the next desired state. And models act as the vehicle for that move.

The bridge model is organized as a two-by-two matrix. The left column represents analysis (the problem, current situation, research, constituent needs, context). The right column represents synthesis (the solution, preferred future, concept, proposed response, form). The bottom row represents the concrete world we inhabit or could inhabit. The top row represents abstractions, models of what is or what could be, which we imagine and share with others.

Ideally, the design process begins in the lower-left quadrant with observation and investigation—an inventory (or description) of the current situation. As the process moves forward, it moves to the upper-left quadrant. We make sense of research by analysis, filtering data we collect to highlight points we decide are important or using tools we’re comfortable with to sort, prioritize, and order. We frame the current situation, but move out of the strictly concrete. We define the problem. We interpret. Analysis begins as thoughtful reflection on the present and continues as conversation with the possible. Crucial for progress is documenting and visualizing our analysis, making it possible for us to come back to it, making it possible to imagine alternatives, making it possible ultimately to discuss and agree with others on our framing and definition. We might write down a list of findings or a statement defining the problem. Better still is writing a story. A story describes actors and actions; it suggests relationships, which we may represent in visual form. A story of what happens suggests a model of what is—an interpretation of our research. The process of coming to a shared representation externalizes individual thinking and helps build trust across disciplines and stakeholders.

Having agreed on a model of what is (framed the current situation, defined the problem) then the other side of the coin (the preferred future, the solution) is implied. An interpretation provides “a description of the everyday in such a way as to see how it might be different, better, or new [1].” We can devise stories about what could happen. We can model alternatives in relation to our first model. In doing so, we’ve moved to the upper-right quadrant, to the use and development of models of what could be. It is in the realm of abstraction—by thinking with models—that we bridge the gap between analysis and synthesis.

These models are hypotheses, speculations, imagined alternatives to the concrete we started with, but they are still abstract themselves. It is easy to “play” with models at this point, to test and explore. But design requires that the work return to the concrete, that we make things real, realize our models as prototypes or even finished form. This is the lower-right quadrant.

Of course, results improve with iteration. Submitting the new prototype to testing, further observation and investigation, continuing around the quadrants, we learn and refine our work.

The bridge model has several antecedents and variations.

The bridge model grew out of personal discussions over the past few years. Rick Robinson (one of this article’s co-authors) has written about “the space in between” research and concept. He has described anthropologist Clifford Geertz’s essay, “Deep Play: Notes on the Balinese Cockfight,” as an example of abstracting a model from research, and one that parallels strongly the moves that other forms of research and design make in moving from description through interpretation to application. “[The construct of] Deep Play becomes a lens through which Geertz can show what’s important about the Balinese cockfight, and his colleagues can understand important underlying factors in something like fan riots at soccer matches [1].”

Writing about the relationship of science to management, Stafford Beer presented a more elaborate model of the move from cases to consensus, from particular to general. He points out that several levels of models are involved [2].

At the beginning of his career, Christopher Alexander described a six-part model. It differs from the bridge model in two important respects. First, Alexander explicitly separates the mental picture (model) from a formal picture of the mental picture (a representation of the model). Second, his notion of a model (at that time at least) was highly mathematical [3].

Vijay Kumar has proposed a model of the innovation process [4]. He frames it as a two-by-two matrix moving from research, to “Framing Insights,” “Exploring Concepts,” and “Making Plans.” He notes, “‘Framing Insights’ are primarily about descriptive modeling, creating abstract mental pictures about the patterns that we recognize about reality. ‘Exploring Concepts’ and ‘Making Plans’ are about prescriptive modeling.” Where the bridge model forefronts the role of models, Kumar’s model forefronts steps that make use of modeling. He recently published a wonderful poster that maps the steps in the “innovation process” to a series of methods.

During the process of writing this article, interactions co-editor Richard Anderson pointed out the Kaiser/IDEO model of the innovation process. Christi Zuber reports that Kaiser Permanente’s Innovation Center (working with IDEO) developed this model in 2004 as part of an innovation toolkit created for use inside Kaiser. This model is similar to Kumar’s model, but the Kaiser model emphasizes storytelling and brainstorming as key methods.

Responding to questions about the origin of the Kaiser/IDEO model, Jane Fulton Suri supplied a recent model of the process of moving from synthesis to strategy. It shares the same basic structure as the Robinson model, though here synthesis (depicted as the right column in other models) is depicted as the left column. The framing of models as a link between patterns and principles is a useful addition [5].

While practitioners and educators increasingly make use of models, few forefront the role of modeling in public summaries of their work processes. Glossing over modeling can limit design to the world of form-making and misses an opportunity to push toward interaction and experience. We see modeling becoming an integral part of practice, especially in designing software, services, and other complex systems.

The bridge model makes explicit the role of modeling in the design process. Explicit modeling is useful in at least two ways. First, it accelerates the design process by encouraging team members to understand and agree on the elements of a system and how those elements interact with each other and their environment. Second, by making the elements and their interactions visible, it reduces the likelihood of overlooking differences in point of view, which might otherwise eventually derail a project.

Explicit modeling also helps scale the design process. It enables designers to develop larger and more complex systems and makes the process of working with larger and more complex organizations easier. Discussing the role of modeling in design also invites comparison and interaction with other disciplines that use models. Ideally, practitioners that use models may, over time, be able to see patterns across their models that will advance the practice of design.

References

1. Robinson, R. “Locating the Work: The Spaces Between”, in Everyday Matters, unpublished manuscript. 2005.

2. Beer, S. Decision and Control: The Meaning of Operational Research and Management Cybyernetics. New York: John Wily & Sons, 1966.

3. Alexander, C. Notes on the Synthesis of Form. Cambridge, MA: Harvard University Press, 1964.

4. Kumar, V. “Design Innovation Process.” Presentation at the About, With and For Conference, Illinois Institute of Technology/Institute of Design, Chicago, 2003.

5. Fulton Suri, J. and Gibbs Howard, S. “Going Deeper, Seeing Further.” Advertising: What’s Next? Conference, San Francisco, December 2006.

Authors

Hugh Dubberly
Dubberly Design Office
hugh@dubberly.com

Shelley Evenson
Carnegie Mellon University
evenson@andrew.cmu.edu

Rick Robinson
Design Continuum
rrobinson@dcontinuum.com

About the Authors

Hugh Dubberly manages a consultancy focused on making services and software easier to use through interaction design and information design. As vice president he was responsible for design and production of Netscape’s Web services. He was at Apple for 10 years, where he managed graphic design and corporate identity and co-created the Knowledge Navigator series of videos. Dubberly also founded an interactive media department at Art Center and has taught at San Jose State, IIT/ID, and Stanford.

Rick Robinson is vice president for practice innovation at Design Continuum in Boston. As chief experience officer (or CXO) at Sapient, Rick oversaw the development of innovative research approaches for understanding human interaction with products, environments, communications, services, and technologies. Before joining Sapient, he founded E-Lab, a user research laboratory. An interdisciplinary social scientist, Robinson received his Ph.D. from the Committee of Human Development at the University of Chicago.

Shelley Evenson is an associate professor in the School of Design at Carnegie Mellon University, where teaches graduate and undergraduate courses in interaction design. She is a voting faculty member in the Human-Computer Interaction Institute (HCII). She is the director of graduate studies in design and for the joint master’s program in HCI between Carnegie Mellon and the University of Madeira. Shelley brings more than 20 years of experience in multidisciplinary consulting practices to the school. She is a frequent speaker at design conferences and conducts design strategy workshops with large and small corporations. Shelley also works with graduate students from the school of design to host the international Emergence Conference in service design. Her current interests include design languages and strategy, organizational interfaces, what lies beyond human-centered design, and design for service.

EDITOR
Hugh Dubberly
hugh@dubberly.com

Footnotes

DOI: http://doi.acm.org/10.1145/1340961.1340976

Figures

UF1Figure. Analysis-synthesis bridge model

UF2Figure. Robinson model

UF3Figure. Beer model (Reproduced with permission)

UF4Figure. Alexander model (Reprinted by permission of the publisher)

UF5Figure. Kumar innovation model

UF6Figure. Kaiser/IDEO model

UF7Figure. Suri/IDEO second model

©2008 ACM  1072-5220/08/0300  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2008 ACM, Inc.

 

Post Comment


No Comments Found