This position paper addresses research topic five:

    "The characterization of CSDM processes, including but not limited to 
     the specification of task models in various domains such as 
     environmental, transportation, natural resource, economic
     development, emergency management, and other high priority subject
     domains; and investigations which elucidate the use of CSDM 
     technology in various CSDM subject domains." 



Urban design, by its very nature, is highly collaborative and involves complex spatial relationships that exist in three dimensions. Decisions are made by multiple entities and effect large and diverse constituencies. Decision-makers must concern themselves with the form of the environment being created, its impact on what exists now, where the money will come from to build it, how it will affect the city's finances, and how the local citizenry will respond. The collaboration is hindered by the parochialism of the various professions involved, the differing analytical frameworks employed by those same professions, and a lack of a common vocabulary. While the microcomputer is employed in the process and has, to an extent, democratized access to these tools (Klosterman, 1992; Brail, 1987), most of these tools have remained largely inaccessible for use in collaborative situations. Much of this is due to the fact that most microcomputers were designed for individual interaction (Shiffer, 1995).

Traditional Methods

Urban design collaboration has take many forms, both active and passive. Suppose that a new development project has been proposed. Collaboration between two professionals for such a project, say architects and urban designers, might involve different design philosophies, but they are drawn from a shared knowledge domain. For input, they rely heavily upon maps, plans, sketches, hard-line drawings (both manually and electronically produced), scale models, photographs, and site visits. Output consists of much of the same. Generally speaking, the collaborative aspect is manifest during meetings in which designs are presented, discussed, modified through quick sketches or on a tracing paper overlay, and agreed upon. The solution is then drawn up with an increasing amount of detail as the project progresses. The process is a dynamic and iterative one and must often conform itself to constraints imposed by conditions at the site, the client's particular concerns, or by limitations of budget.

Real estate professionals/developers working on the same project draw upon a different kind of shared knowledge domain, one involving economic models and market research. For input, they rely on large amounts of socioeconomic data, survey results, various economic indicators, sources of available financing, and the local political climate. Internal rates of return and the like take the form of spreadsheets, tables, descriptive text, photographs, and a number of other specialized tools that support these kinds of analyses. However, unlike an architectural rendering, this kind is output is difficult to depict graphically and explain to a lay person. Like the architects, this is a dynamic and iterative process, although the redesign that takes place is involves massaging the numbers-studying the effects of the financial restructuring that is necessary to make the project economically viable.

To the citizenry impacted by such a project, concerns arise not just relative to the design of a project, or its economic viability, but the effect of the project on their "community" as well. Input here takes on a completely different dimension because the domain shared is viscerally rather than knowledge based. The difficulty in defining "community" itself is indicative of the value-ladened nature of this information. A neighborhood, or town, or city, can mean many things to many people-yet there are strands of shared values that can be woven together and which can constitute an agreed upon "character." The cognitive blueprints, or mental maps, which people use to organize in daily life or for the sense of place which they ascribe to an area, has challenged researchers for years. Yet this is one of the inputs which individuals use in their decision-making process. Output might include a letter to the editor of the local newspaper expressing support or dissatisfaction with the project, involvement with a community organization or attending one of their public meetings to express views and concerns, or even going all the way to city hall.

A resurgence of interest in the creation of human-scale communities has created a renewed awareness of the role and the importance of urban design. Further, it has been demonstrated that the kind of suburban development which typically takes place in this country has created enormous social, environmental and economic costs, which until now have been hidden, ignored, or quietly borne by society. Businesses suffer from higher costs, a loss in worker productivity, and underutilized investments in older communities.

New Urbanism as a Collaborative Process

An example of that interest which I've chosen to pursue in my research is known as Traditional Neighborhood Development and, as practiced by Andres Duany and Elizabeth Plater-Zyberk, involves a week-long "charrette"[1] which provides a setting in which all constituents-from municipal officials to interested citizens-can participate in the planning process. The charrette helps to educate the participants to different approaches, incorporates their contributions, verifies decisions and diminish the difficulties of the ensuing permitting process.

The charrette establishes a full working office of five to 20 people on site, staffed with a small core of experienced Duany and Plater- Zyberk designers, working with local architects, landscape architects, historians, engineers, ecologists, and financial and marketing consultants. The charrette begins with a day of visits to the site and to nearby towns which might serve as models, and a presentation to the community of the principles of town planning. During the following days, the team, including the client, works day and night, meeting often with the local officials and advocacy groups, designing everything from the master plan to typical buildings, the codes, and specific landscapes.

While it is of significant interest that this integration of urban design into the planning process facilitates more workable design concepts and the accompanying regulatory frameworks, more enticing is the charrette process employed-it is example of the kind of productive, collaborative process which people are able to participate in, if given the opportunity. Most of the process is based on the assumption that a greater degree of access to relevant information (with the ability to present that information in a variety of ways) will lead to the consideration of a greater number of alternative scenarios. Furthermore, the consideration of a greater number of alternative scenarios will lead to better informed public debate (Shiffer, 1995).

Under the banner of "new urbanism" (because they base their work on principles known to the planners who, earlier in this century, created such communities as Scarsdale, New York, Mariemont, Ohio, and Lake Forest, Illinois), Duany/Plater-Zyberk have promulgated some surprisingly simple and obvious rules for building better suburbs, redesigning existing suburbs, and redeveloping existing urban environments. They can be summarized in these five principles which are the foundation of Traditional Neighborhood Development (TND)[2]:

1) the neighborhood has a center and an edge;
2) the optimal size of a neighborhood is a quarter mile from center and an edge;
3) the neighborhood has a balanced mix of activities--dwelling, shopping, working, schooling, worshipping and recreating;
4) the neighborhood structures building sites on a fine network of interconnecting streets;
5) the neighborhood gives priority to public space and to the appropriate location of civic buildings (Duany & Plater- Zyberk, 1991).

What's Needed

The challenge, as I see it, is to actualize the theoretical goals of interface design for collaborative spatial decision-making within a real environment which generates new forms of physically defined communities in contemporary society. The interface should support and supplement, not supplant, these working processes. All to often the automation of a task leads to a redesign of the task itself. What the TND charrette process offers is an incremental design process that respects the complexity and diversity of urban life. The challenge is to combine electronic processes with traditional methods that will model, facilitate, and possibly extend this dynamic decision-making environment.

Communication, in this context of the visualization of information, is about delivering symbols to an audience. A technology that can blend symbols from many different knowledge domains into a shared environment portends a potentially revolutionary change in the means of designing and distributing documents. The current state of interface design doesn't make it very easy for a novice to act intelligently, or accommodate varying levels of expertise, or express a multiplicity of content--to produce any value for collaborative decision-making.

Urban design is a highly collaborative process that relies heavily on spatial decision-making and spatial relationships. The use of GIS in this process allows the recording of new types of information, the linking of data from disparate sources, the provision of more timely information than ever before, and opens new possibilities for understanding environments and communities that planners and researchers have only begun to see. An interface created to support urban design, and in particular Traditional Neighborhood Development and the charrette process, should make it possible to pose informal queries about the existing and future conditions of a particular location and quickly visualize and quantify impact. It should employ GIS and other analytical tools to facilitate syncretic solutions. After focusing primarily on the structure and development of the tools themselves, the GIS research community must now address the use of those tools. Structuring the use of the tools themselves should allow users to exploit the power of GIS and related information systems. A successful urban design interface would allow the users address the task in the substantive concepts associated with the problems they're facing, and at the same time facilitate the communication needs of transdisciplinary users with their attendant conceptual frameworks and skill levels.


[1] A charrette is a word that describes an intensive, continuous 
    design session leading up to a presentation. It originally was used to 
    describe the cart which architecture students at the Ecole des 
    Beaux-Arts used to carry all their materials through the streets of 
    Paris on the way to their reviews.
[2] Traditional Neighborhood Development is a set of urbanism 
    principles created by the architecture and planning firm of Andreas 
    Duany and Elizabeth Plater-Zyberk (DPZ) which were derived from 
    DPZ s insight that our cities are segmented, our land uses 
    segregated, our transport mechanized, and our public spaces 
    fragmented, not because of economics or planning philosophies, but 
    because our planning tools, especially the zoning ordinances, 
    mandate it. Change the codes and you change the built 
    environment, is the deceptively simple message of DPZ. To that 
    end, DPZ s codes concentrate commercial activity, including 
    shopping and working, in town centers. 


Brail, R.K. 1987. Microcomputers in Urban Planning. New Brunswick, NJ: Center for Urban Policy Research.

Duany, A., Plater-Zyberk, E. 1991. Towns and Town-making Principles. Cambridge, MA: Harvard Graduate School of Design.

Forester, J. 1982. "Planning in the Face of Power," Journal of the American Planning Association, 48, 67-80.

Shiffer, M. 1995. "Environmental Review with Hypermedia Systems," Environment and Planning B: Planning and Design,(forthcoming).


Tom Pederson received his Master of Science degrees in Historic Preservation and Real Estate Development from Columbia University. He first began studying urban applications for GIS in 1988 while employed at the architecture, planning, and urban design firm of Michael Kwartler and Associates in New York City. Currently, he is a doctoral student in the City and Regional Planning Department at the University of Pennsylvania under the supervision of Dr. C. Dana Tomlin. The working title of his dissertation proposal is "User Centered Task-Driven Interface Design for GIS."

He is also currently a consultant to the City of Philadelphia's Office of Housing and Community Development assisting in the implementation of GIS in their Community Development Corporation's Strategic Neighborhood Planning program. This work involves the introduction of GIS concepts and analysis, principles of cartography and map design, the issues surrounding the quantification and mapping of place-based qualitative attributes, public access to GIS data via the World Wide Web, and strategies for the integration of neighborhood generated data into the structure of the city wide GIS.