It was shown how Lynch's nodes, paths, districts, edges and landmarks could begin to be identified using structural clues and GIS, but the most important idea Lynch put forth in his work is that the people who live and work in a place have the best answers to many design questions. This part of his theory has been the most neglected by planners--Lynch says that his model was used as a way of further distancing people from the design process instead of making them a part of it. We have another chance to test his theory using new information sharing technologies that can integrate the designer, the design process and the citizenry. What is needed is an information system that can gather information from people on issues of concern to the design process, synthesize this input into some kind of shared understanding of the city and report back the results and allow for comments on these results (like the design process, the information sharing process should be iterative).
The premise of this paper was that if design concepts could be integrated into a GIS two benefits emerge. First of all, the designer can analyze a larger portion of the landscape because once their methods are operationalized and integrated into the GIS, computing power allows a relatively quick processing of as much of the landscape as exists in the database. This benefit was seen in the nodes prototype, where a GIS function for finding concentrations of activity was developed based on familiarity with a small area of a city and then applied to the rest of the city database. Second, having a computerized information system will lead to new kinds of studies and working methods. The pattern finding prototype suggests a new way for urban designers to take a comprehensive look at how areas are similar and how they are different.
Designers should add GIS to their arsenal of analysis tools. However, new spatial analysis interfaces will be needed to attract this group of professionals to GIS, because their modes of thought are often non-linear and less rigidly structured than those who have already adopted GIS technology. Although progress was made in creating a "sketch plan" of the city, detailed data relating to the movement of people and vehicles would have been valuable because it could have helped to explain how people move from home to node and from node to node. It also could have helped answer some interesting questions that were raised regarding how similar areas in the city relate to one another.
This study has shown, however, that GIS analysis can play an important role in urban design. The pattern finding application stimulated a great deal of exploration and discussion among the planners that tested it. The node finding application proved to be a powerful tool, as it accurately located the busiest areas in the city using only a detailed land use coverage and a well thought out algorithm.
Model development is concerned with the construction of, as Rowe puts it, descriptive, explorative, predictive and planning models, as well as the ability to go back and forth between design concepts/models and make refinements. For example, in the nodes model, after developing the nodes and using them in an analysis, I might want to go back and remove the constraint that institutional uses be a requirement or make the node radius smaller. The GIS must accommodate this kind of refinement easily. The construction of models is also an important element of the system. The models presented here were constructed with GIS scripting languages which are almost as complicated as generic programming languages. A more visual-oriented model building system would be necessary for designers.
Interface issues fall into two categories, hardware and software. The standard GIS interface currently consists of a keyboard and mouse for input (digitizing tablets are usually used only for data entry), and a monitor for visual output. It would be more natural for the monitor (visual interface) to rest on a table and tilted only slightly to mimic the orientation of a map or reading table. Regarding input, it would be best to use a stylus that wrote directly on the screen because people are used to drawing on the place they want a mark to be, not twelve inches away. These recommendations would not be made if their purpose was only to make the computer interface more like one to which designers are more accustomed. If there were significant advantages to be gained from the current interface, then the designers should adjust--for example, most people can type faster than they can write freehand, so perhaps the keyboard should stay--but the current interface only exists because it was easy to create, not because moving a mouse on a flat surface is better than moving a finger across the screen.
The software issue is concerned mainly with text-based input versus more graphical methods. As software packages (and the hardware to run them) are evolving, their interfaces are becoming more graphical and less text-based due to user preferences, and urban designers share this preference. It was mentioned above that scripting/model building should become more graphical, and this is the main extension of the general argument that graphically-oriented user interfaces are preferable to text-based ones. A less obvious software issue is inter-application communication. Communications, word processing, database, spreadsheet and GIS software should work together seamlessly in order to promote maximum work efficiency.
One glimpse at the way this might look in practice is the Netscape Navigator software package with "helper" applications, "plug-ins" and the Java scripting language. Navigator is a piece of software that provides Internet networking capabilities. When the network software accesses text or graphics, it simply shows the text and graphics to the user. When it accesses a kind of data that it does not recognize, it uses a helper application, a plug-in, or Java to process it. The idea is that the user simply tells Navigator what he or she wants and lets that one piece of software figure out what other software might be needed to complete the task. This may soon be the model for operating systems and that would be a boon to the development and adoption of information systems by all disciplines.
As for private firms, their should be a change in policy regarding digital data. Often when a government project is done by a private firm, the firm retains the rights to the data, meaning the government can not distribute it. This is a complicated issue to solve legally and financially, but it would be in society's interest to create a system where all data created by public and private entities could be accessed from a centralized place (so that everyone knew where to find data). This system does not have to only contain spatial data sets. Maps, reports and other sources of information can be in digital form and be accessible through the network.
Once these technical issues are resolved, a designer should have access to any information for any place without leaving their office. This is not to say that doing urban planning without leaving the office is a good thing, but rather to emphasize that the overhead involved with the current system of identifying, obtaining and converting data severely limits its use, and government can play an important role in alleviating this problem.