4S (Society for the Social Studies of Science) Halifax NS October 1998

Two presentations:


The first paper describes the original project to study the different
roles geographic information technology is used to make a political case.
With help of STS concepts, notably boundary objects, we are now able to
present some results that were closed to us before we became STS-aware.
This paper focuses on the problems encountered and how we now approach
them.

Networks of NIABY (Not in Anyone's Back Yard):
Geographic information technology uses to site Low Level Radioactive Waste repositories in the USA

Nicholas Chrisman
chrisman@u.washington.edu
Department of Geography Box 353550, University of Washington
Seattle WA 98195-3550

and
Francis Harvey
Department of Geography, University of Kentucky

Full paper (.pdf)

ABSTRACT


Geographic Information Systems technology (GIS) has been developed with the express purpose of integrating information from different sources ­p; from different disciplines and different ways of knowing. Studying the application of this technology gives particular insight to the debates over translation (trading zones or boundary objects) between different actors in a technoscientific network. This paper presents an example drawn from a comparison of the procedures applied by sixteen states of the United States in addressing a common federal requirement to site a depository for low level radioactive waste (LLRW). Most states applied GIS technology at some stage in the process, each with radically different data sources and radically different logic to combine the sources. Even though the legal requirements were the same and many of the projects were performed by a small group of corporations specializing in LLRW, other forces overwhelmed any tendency to replicate a uniform solution. The authors began this study with the expectation that there would be some common social factor to "explain" the "failure" of the technology. The case proves a lot more complex. GIS technology demonstrates how information technologies are locally constituted, and how "context" does not predict results.

The second paper looks particularly at the social construction of software
for addressing geographical question and utilizes Latour's concept of
irreversibility to illustrate that the presumed arrow of progress in
technology has more complex diversions and returns to prior states
(cycles?). The author critically assesses his own role as insider in the
process of constructing irreversibility and discovering that it didn't
work.

Topological Representations of Geographic Information: Reversing the irreversible arrow of progress

Nicholas Chrisman

ABSTRACT
Over the past thirty years, Geographic Information Systems (GIS) have emerged from tiny, fringe origins to become a significant industry. While this development was certainly made possible in some general way by the rising tide of computing power and the demands of various environmental disciplines, the specific nature of the software technology was a matter of significant contention. Although massive government expenditures were poured into the image-based "raster" technology, other groups promoted a "vector" representation that more closely approximated the line drawing
techniques of mechano-optical cartography. This paper traces how a number of groups promoted an approach to vector data structures that applied simple rules of topology. In particular, the paper will draw on the role of the author in promoting this approach as necessary and inescapable. In the period from 1974 (at the AUTO-CARTO I event) to the NSF-funded Symposium for Topological Data Structures in 1977, my group at the Harvard Laboratory for Computer Graphics expanded the topological alliance from Census Bureau to an international consensus. We also wrote the software that demonstrated its effectiveness. The analysis will draw upon the role of time in the construction of technology, particularly the claim of irreversibility, and general notions of progress as they were used at the time. As an analysis produced by a direct participant, there will be all kinds of issues about how reliable a reporter I can be. The author does not stay at the center of the story: topology becomes a central element in the marketing of GIS software for a number of years. However, current software at the high-end have abandoned the use of topology as an organizing principle of the database. In the light of these changes, the original claims for topology can be reevaluated as contingent upon a set of technical assumptions not recognized at the time. The paper will make an attempt to adapt the linguistic terminology suggested in STS (translation and creoles) to a the representation of two-dimensional figures on maps.