Issues of Collaborative Spatial Decision-Support in City Planning Contexts

Michael J. Shiffer, Ph.D.
Department of Urban Studies & Planning
Massachusetts Institute of Technology
77 Massachusetts Ave. Room 9-514
Cambridge, MA 02139
Phone: (617) 253-0782
Fax: (617) 253-3625


This position paper is intended to illustrate a particular approach to a study of Collaborative Spatial Decision-Making (CSDM) that focuses on city and regional planning contexts. While this approach is related to many of the topics mentioned in the Open Call for Participation, it is most closely related to topic 2 ( and implementation of methods to improve decision-makers' interaction with spatial analysis tools...), and topic 5 (...the characterization of CSDM processes...). Rather than provide an exhaustive review of the relevant literature, this paper will describe a group decision support context from the perspective of personal experience with research in this area.

City Planning Settings

The focus of this research involves face-to-face meetings of people involved in discussions with a spatial emphasis in city planning settings. The use of information in these situations frequently involves recollections about the past, descriptions of the present, and speculation about the future. This is most often conveyed using the cognitive information provided by the meeting participants and may be augmented using various media and tools. Spatial representations such as maps, typically provide a central organizing metaphor.

For example, in many planning situations people will gather in a room, spread a map on a table, point to areas, and verbally recollect, describe or speculate. At times, these discussions will be augmented with documentation or imagery. In other cases, they will be augmented with analytic tools.

Recollection of the Past

In recollecting the past history of a given site or planning issue, conversations may revolve around what was said, what was done, or what a place was like among other things. For example, members of a group may try to recall the impact of past interventions on an urban landscape to better understand what may lie ahead given similar circumstances.

Where the recollection is about fairly structured recent activities, (such as past planning meetings), the conversations can be augmented with records and systematic documentation of past interactions. However, access to this information is frequently not random and can be dependent on a specialized information recording and retrieval "system" such as a meeting secretary or stenographer. Furthermore, such methods of recollection rarely incorporate spatial referencing (using such things as historical maps). Where a systematic documentation is lacking, as with recollection of the past environmental conditions of an area, the high degree of dependence on human memory can lead to problems based in the inconsistency of individual memories. This is exacerbated by the fact that personal points of view tend to be subjective. For example, someone may recall traffic on a particular street to have been particularly heavy, while another may think of the same stretch of roadway as lightly traveled.

Where there is a lack of documentation or data to support these recollections, arguments related to inconsistent memories are likely to persist. These arguments can dominate a discussion and shift the focus of a meeting from the matters at hand. GIS support of recollection in collaborative contexts has been somewhat limited due to a number of factors including fairly weak historical references. While GIS (in combination with historical data) can indeed be used to facilitate recollection about characteristics such as demographic trends, property values, and other generalizable information, it is less adept at conveying the past character of an area. Furthermore, using this technology to assist recollection depends on systematic archival of spatial "snapshots".

Descriptions of the Present

Descriptions of present conditions are generally used to familiarize participants in a collaborative situation with the area being discussed so that everyone can work from a common base of knowledge. These descriptions frequently involve some sort of spatial referencing. While many of these references are verbal, (i.e.: "over by the river", "on the site of the former factory"), such references become increasingly inappropriate as the level of familiarity with the site on the part of meeting participants is lacking. For example, the term "on the site of the former factory" is completely meaningless to meeting participants who are unfamiliar with the area being discussed.

The lack of familiarity with a given site in the context of a description of existing conditions can be aided through the use of an up-to-date map that is used as a central reference point. Individuals describe present conditions verbally and augment this by gesturing to a map spread on a table or tacked to a wall. Such descriptions may be further augmented using thematic data and visual imagery. The thematic data may be provided in the form of land-use maps, or demographic conditions of an area. Visual imagery may include photos or video tape of selected sites. The juxtaposition of the above media can strengthen a collective understanding of the various characteristics of a given site. Until recently, GIS has been unable to effectively support descriptions of existing conditions due to issues of speed, human interface and integration with other forms of media. While many of these issues are being addressed using modern GIS tools that take advantage of inter-operability and component software, the techniques for the effective juxtaposition of this information for retrieval in collaborative contexts need to be further developed.

Speculation About the Future

Speculation about the future of an area involves experienced individuals extrapolating measurable phenomena from past experience and applying it to the future using informal mental models. A much more formalized mechanism for such speculation (some may call it prediction or modeling) has been made available to collaborative processes through the use of computer-based analysis tools, and most recently GIS.

Augmentation of collaborative settings with analytic tools such as GIS has traditionally been handicapped by a lack of immediate response, and abstract output that tends to exclude from such conversations those who are not technically-sophisticated. A degree of immediacy in response to user queries about alternative scenarios can be afforded through the implementation of direct manipulation interfaces. Such interfaces translate human desires into commands that the computer can understand. Multimedia representational aids support information flow in the other direction by augmenting numeric values with graphical representation and associated imagery to transform abstract data into concepts that the human can understand. Both direct manipulation and multimedia representational aids have been made possible by recent increases in computing power available to the masses. While the tendency to apply such power to the undertaking of previously unattainable analyses is important, the ability to improve the communicability of existing analysis tools, especially in collaborative contexts, should not be overlooked.

Augmentation of Planning-related Conversations

To better support these activities it is important to develop a conceptual and practical means of implementing emerging technologies so that cognitive information can be effectively augmented. Such augmentation can be facilitated using both a spatially-referenced associative information structure and a directly manipulable mechanism that affords multiple representations of future scenarios in a spatial context. Spatially-referenced associative information structures, (such as those now made possible through the modularization of WWW and GIS tools), can aid in the description of past and present conditions by providing a means of juxtaposing maps, imagery and other relevant information in a manner that can effectively augment conversations. The employment of directly manipulable analysis tools with multiple representations of otherwise abstract data has the ability to bring a level of analysis into conversations about future scenarios that is significantly more robust than the mere speculation afforded to many collaborative settings. Furthermore, desktop video conferencing and collaborative software has the ability to include individuals who would have otherwise have been excluded from such conversations. Nonetheless, the employment of these technologies alone is not enough to ensure successful augmentation of planning-related conversations. A deeper understanding of the nature and structure of human communication in these settings is necessary to provide a starting point from which one can begin to pose relevant CDSM-related research questions. What follows are some opening questions and issues that can hopefully clear the way to such investigations.

Issues and Questions for Research

How does this technology change the balance of power in planning settings that are collaborative in nature? Access to these tools (or a lack thereof) may put certain parties at a disadvantage. Access is more than having a machine available running relevant software. It also involves having an understanding of the information base and functionality of the software contained in the machine. While policies of equal access may look good in writing, such a balance may be difficult to strike. This research has the ability to enlighten others about which levels of access are easily attainable, which are difficult, and how to determine the level of infrastructure and knowledge necessary to attain the desired level of access.

How does it change the way people interact in the context of face-to-face communications? The availability of broadly manipulable tools in group planning contexts can lead to the consideration of a much broader range of alternative scenarios. While one might argue that this can lead to better-informed conversations, it can also make it difficult (or impossible) for a group to reach consensus. Furthermore, while multiple representations may minimize arguments based on "apples and oranges", they can also confuse or mislead. Just as this technology has the ability to create compelling representations of spatial scenarios, it has the ability to create compelling misrepresentations. While this issue is not new to spatial analysis, it can be exacerbated through the use of multiple representations in collaborative contexts, therefore it is important to understand the potential pitfalls of unintentional misrepresentation so that measures can be taken to minimize it.

How can these tools be effectively, inexpensively, and fairly delivered to the contexts that can best benefit from their use? Modular tools with spatial referencing (such as GIS components combined with WWW client and server software) are a step towards the effective delivery of a mechanism that augments recollection, description and speculation. However it is still necessary to develop conceptual designs so that these tools can be properly implemented. Such designs can result from an iterative process of observation, development, testing, and feedback.

Finally, evaluative mechanisms need to be developed that provide feedback about where these computer-based tools fall short in their design, implementation and execution. Such mechanisms will have the capacity to expose whether the use of planning support systems such as these can lead to better-informed conversations, planning, and/or decision making.

The best way to address these questions is through collaborative research initiatives such as that which is proposed here. It is hoped that such interactions can lead to a broader understanding of the methods, tools, and techniques used by others in previous and ongoing research.


Michael J. Shiffer is on the faculty of the Massachusetts Institute of Technology Department of Urban Studies and Planning as a Lecturer and Research Scientist. There, he teaches courses in analytic methods and new and emerging technologies for planning and decision support. Shiffer investigates how information technologies (such as multimedia, video conferencing, and world-wide networking) can better inform public debate and decision making with a specific focus on geographic information.

Shiffer received his Ph.D., and Master's of Urban Planning, from the University of Illinois at Urbana-Champaign. Shiffer has had additional training in communications, human-computer interaction, and psychology. He has earned a Bachelors degree in Geography from De Paul University in Chicago.

In addition to his academic work, Shiffer consults with a variety of public and private agencies on information technology and communications issues. He is an active professional speaker and author on the application of information technology in the area of city planning and development.


Shiffer, M.J., (1995) "Multimedia Representational Aids in Urban Planning Support Systems", in Marchese, F. (ed.), Understanding Images, (New York: Springer-Verlag), pp.77-90.

Shiffer, M.J., (1994) "A Geographically-Based Multimedia Approach to City Planning", in Plaisant, C. (ed.), Human Factors in Computing Systems, (New York: Association for Computing Machinery), pp.265-266.

Shiffer, M.J., (1993) "Environmental Review with Hypermedia Systems", in R.E. Klosterman and S.P. French (eds.), Proceedings of the Third International Conference on Computers in Urban Planning and Urban Management, vol. 1, pp. 587-606 (also forthcoming in Environment and Planning B: Planning and Design, volume 22).

Shiffer, M.J. and Wiggins, L.L., (1993) "The Union of GIS and Multimedia", in Castle, G.H. (ed.), Profiting from a Geographic Information System, (Ft. Collins, CO: GIS World, Inc.), pp. 336-341.

Shiffer, M.J. (1992) "Towards a Collaborative Planning System", Environment and Planning B: Planning and Design, volume 19. pp. 709-722.

Wiggins, L. L. and Shiffer, M. J. (1990), "Planning with Hypermedia: Combining Text, Graphics, Sound, and Video", Journal of the American Planning Association, Spring, 1990. pp. 226-235.