"Shallowing" the Learning Curve
Making GIS easy enough for even an architect to understand

Nils Gore and Joe Hagerman
Mississippi State University, School of Architecture

This paper is based on an ongoing research project at Mississippi State University, School of Architecture. For updated information about this research project (and the revised GIS Interface Modification) please refer to gis.sarc.msstate.edu.

Background
This paper describes a current project that expands the possibilities for the focused use of geospatial technologies in community design¹. The current use of such technologies in architecture is limited; yet the act of community design is inherently geospatial, and could benefit from wider use of geospatial technologies. The use of these technologies in other professions has proven to be very rewarding, in the sense that it has allowed them to visualize things they have previously been unable to, using geographic information system (GIS) analysis based on census data, physical inventories (roads, streams, political boundaries), and demographic data.

Much of the work of community designers is subjective or qualitative in nature². It is not formulaic due to the unique nature of individual communities. We are concerned with such things as: spatial qualities of landscapes, understanding the historic fabric of a town, and understanding district edges, nodes and patterns of development. These subjective data are collected through on-site observation and personal contact with citizens and community leaders. The data are not easily quantified, but are critical for analysis of a town during any community design project. In order for community designers (architects, landscape architects, planners, urban designers) to use GIS effectively, customized data sets-describing subjective, qualitative attributes of communities-will require development, implementation and analysis.

This project will explore 1) the problem of developing a GIS data set, based on subjective observations; and 2) customization of the GIS user interface, to advance the work of the community design professions. The work will take place in the context of a "case study" of downtown Brookhaven, Mississippi, which will undergo significant change with the planned introduction of a fine arts magnet high school on the historic Whitfield College site, which sits adjacent to the Brookhaven central business district. The primary community design issues of concern to Brookhaven deal with parking and downtown housing for faculty artists.

1. In general, the term "geospatial technologies" is meant to describe the digital technologies of Geographic Information Systems (GIS), Global Positioning System (GPS), and Remote Sensing (RS).

2. These professions include architects, landscape architects, city planners, historians, urban designers, etc. their work is primarily concerned with imagining the future of communities, in the context of their peculiar existing conditions (as opposed to managing communities, which is currently the primary use of geospatial technologies).

Geospatial Technologies and Community Design
Community design has always been a geospatial operation: in our work we design buildings for construction on the earth; we draw maps and plans fixing their locations; we read and use others' maps and plans as we are making our own.

Until recently, with advances in digital technologies used to represent our work and our world, the work of the community designer was limited to what could be represented by hand, for a specific project, at a specific point in time. The designer's understanding of the site in question was limited to one's personal observations, the map materials available in the local government office, or to a site survey custom-made, by others, for the project. The constraints imposed by project limits, representation technologies, and reproduction methods forced planners to see their work through a narrow lens, focused on solving immediate problems, but failing to advance the potential of a project through meaningful connections to a larger project context.

Though digital geospatial technologies (GIS, GPS, Remote Sensing) have been developed and used by other disciplines for the past twenty years, their use in everyday design practice is not common due, in part, to a lack of data-data that is qualitative in nature-addressing the subjective, circumstantial nature of the designer's work in a particular place with distinctive qualities.

Geospatial Information Providers and Community Design
Providers of geospatial technologies have not yet embraced the community design professions. For instance, Environmental Systems Research Institute (Esri), a recognized leader in GIS software development, has a special web page for "Your Industry,"³ listing Agriculture, Forestry, Mining, Petroleum, Law Enforcement, and some twenty other industries. As of this writing, none of the community design professions is listed.

Similarly, a study conducted by the Mississippi State University (MSU) Social Science Research Center⁴ has determined that of 37 courses being taught in nine Mississippi institutions, only one course is listed in a community design profession department: Regional Landscape Planning offered in the Landscape Architecture department at MSU.

3. Esri's Your Industry: http://www.Esri.com/industries/index.html

4. Liesel Ritchie, Remote Sensing and Geographic Information Systems Course Availability in Mississippi's Public Universities and Community Colleges (Social Science Research Center, Mississippi State University, February 1998) pp.7-9.

Types of Data Commonly Available for GIS
Data is the heart of a geographic information system. A query to a GIS will yield results only as good as the data set being queried. And there is an abundance of data available for immediate use in GIS. For instance, several gigabytes of data come supplied with Esri's ArcView GIS software. These include data sets of political subdivisions, natural features such as streams, man-made features such as roads, and population data. The Mississippi Automated Resource Information System (MARIS)⁵ acts as the GIS agency for our state. One can download-for free-spatial data pertaining to county government, public utilities, streams, census block numbering areas, primary and secondary roads, and some dozen other categories. Federal Agencies, such as USGS and the Census Bureau are useful sources for geospatial data.

In general, it is possible to characterize data from these sources as quantitative, meaning that they refer to concrete resources that exist on the earth (roads, streams, and populations) or to abstract overlays that we use to organize ourselves (political boundaries).

5. MARIS http://www.maris.state.ms.us

Types of Data Used for Community Design
Though community designers use quantitative data in our analyses of existing places, we also employ data that can be described as qualitative, meaning that it refers to a subjective description of qualitative aspects of our environment. These data have traditionally been collected by personal observation of trained designers. Our attitudes towards a place are often based on sensory perceptions, on the "feel" of a place, or on our holistic understanding of a place. We react to our impressions based on notions of beauty, comfort, rightness, and other aesthetic judgements. These subjective judgements are, to a degree, our stock-in-trade; they are a critical component in our work as community designers.

For instance, in historic preservation work, one is often required to assess a particular building as "pivotal, contributing, or non-contributing" in terms of how it relates to a larger historic district or to a historical epoch. There is no formula for making such a determination. Rather, it requires a trained observer to go into the field, analyze the historic character of the building and make a subjective judgement.

Another example: in Booneville, Mississippi⁶ the Small Town Center proposed a new entrance to the city that followed the course of an existing street. The street was characterized by a gently winding roadway, lined with historic houses and beautiful, mature trees. It offered a pleasant alternative to the current exit off of the by-pass, down the congested commercial strip and via a longer, tedious route into town. Our assessment of these two routes was subjective, in the sense that there is no formula for determining the qualitative differences between two routes. Yet the difference is substantial, and will have far-reaching consequences in the economic development of the town, public perception and community pride.

We did our geospatial data collection and analysis for the Booneville project using fieldwork with paper maps and hand notations. A GIS would make our work more efficient and substantial, in the sense that these data, based on qualitative impressions, could be used for subsequent analyses of more complex environments, or shared with others. In order to take full advantage of GIS in our work, we need to develop techniques for collecting and classifying subjective, qualitative observations into a data set useful by a GIS, and we need to modify the software interface to simplify data input for community design purposes.

6. Shannon Criss, Booneville, Mississippi: A Case Study Assessing the Possibilities, (The Small Town Center, Mississippi State University, 1998).

Project Plan
This project will consist of three steps, with which we will 1) develop a "taxonomy" of attributes and develop techniques for their collection; 2) develop a modified GIS application using Esri's MapObjects; and 3) conduct an analysis of them using a digital GIS. The work is happening in the summer of 2000.

Step One: Taxonomy Development and Data Collection:
Using past community design projects⁷ as a guide, we will develop a taxonomy of qualitative attributes that are a factor in the community design decision-making process. In the case of the Brookhaven problem, for instance, we will need to develop ways of characterizing the potential of individual properties to support second floor housing,⁸ and to assess the impact of their parking needs on the existing fabric. We will need to develop an understanding of current parking needs and how that relates to public perception about parking in real terms (as opposed to the zoning requirements for parking, which are abstract and quantitative).

Field studies will be conducted to collect the data identified in Step 1. Using a global positioning system (GPS) receiver and laptop computer, we will input our observations into a GIS. Classifications will be translated into point, line and polygon themes, and data tables will be constructed which link to the geospatial attributes.

In our work we make extensive use of photography due to the wealth of information that can be interpolated from photographs. Remotely-sensed raster data, in the form of aerial and satellite imagery, will be analyzed and digitized as part of this work, along with digital reference photos taken at ground level, and linked to the database table. Such data are being collected now in Brookhaven in the context of a related academic project⁹.

The product resulting from this work will be immediately useful for our case study but will also be documented in a way that will permit it to "stand-alone;" to serve as a model for community design fieldwork¹⁰.

 Step Two: GIS Interface Modification
The architectural profession long ago embraced computer aided design (CAD) for the production of architectural drawings. It is now the standard method for generating design drawings at all stages of a design project. For architects, the de facto standard CAD package is AutoCad. For GIS professionals, the de facto standard GIS package is Esri's ArcView (or ArcInfo). The ArcView graphic user interface (GUI) is considerably more complicated, in the sense that it constantly changes depending on whether one has a view, table, layout or script as the active window. The learning curve is rather steep. For this part of the project, we are interested in modifying the ArcView interface so that it is more user-friendly for community design purposes.

 Step Three: GIS Analysis and Evaluation of Data
The GIS will be queried to address the problems posed by the Brookhaven case study.

7. The Small Town Center has a twenty year record of projects from which to draw.

8. These housing issues include: structural capacity of second floor spaces, fire egress, public accessibility, handicap accessibility, owner-driven issues, zoning-driven issues, real estate issues.

9. A current, purely academic, design studio project in Brookhaven (under the direction of Small Town Center Director Shannon Criss) has collected some data already, which will be mined for content. Dr. Roger King of MSU's Remote Sensing Technology Center (RSTC) has also shared with us some aerial data collected as part of the RSTC's transportation initiative.

10. See Architectural Design Guidelines for the Tupelo Historic Downtown Overlay District, by John Poros (The Small Town Center, Mississippi State University, 1998) for example of a field guide to instruct homeowners on how to qualitatively assess proposed changes to their homes in this sensitive historic district.