This section reports on activities of the National Center for Geographic Information and Analysis other than those directly associated with the new Varenius project. NSF’s funding for 1997 under SBR 96-00465 included support for a transition from the previously funded activities under SBR 88-10917 to the new Varenius structure. These activities include efforts in research through the continuing NCGIA research initiatives; efforts in education; and outreach. The following sections report on these activities, and the appendices include summaries of publications, presentations, visitors, degrees granted, and courses taught.
Initiative 16: Law, Information Policy, and Spatial Databases
The closing session for Initiative 16 on Law, Information Policy and Spatial Databases was held as GIS/LIS '97 in Cincinnati, OH, in October, in a session titled Law and Information Policy for Spatial Databases: Research Results and Future Needs. The following papers were presented:
At Maine, Yvette Pluijmers, M.S. student, has spent the past year addressing the limits of protection for intellectual property in geographic datasets in the commercial sector. In her work she explores competing theories from both legal and economic views on protecting intellectual property and evaluates those theories in the context of the commercial sale of geographic datasets. A survey with two objectives was distributed to sellers of spatial datasets. The first objective was to determine the current level of satisfaction with current means of intellectual protection of spatial datasets; satisfaction both in the ability to build on the datasets of others as well as the ability to restrict use of their own datasets by others. The second objective was to outline competing models for protecting intellectual property in geographic datasets and determine commercial seller preferences in regard to these alternatives. The results of this survey work should have value in assessing and developing new models and methods for protecting intellectual property in spatial datasets.
Initiative 17: Collaborative Spatial Decision Making
Leaders: Marc Armstrong (University of Iowa), Paul Densham (University College, London) and Karen Kemp (University of California, Santa Barbara)
In April, Densham met with Goodchild to discuss progress on the initiative and draft a program for its closing. This program was subsequently approved by the other two co-leaders of the initiative, Armstrong and Kemp. Discussions about the closing continued in October, when Armstrong met with Max Egenhofer in Maine.
Initiative 17 will close during the summer of 1998. The closing meeting will be at a European workshop on Collaborative Spatial Decision Making (CDSM). This workshop is being organized by R. Laurini, and Densham is on the advisory committee.
Several papers on CSDM topics were presented at GIS/LIS ‘97 in Cincinnati and at AutoCarto 13 in Seattle. For example, Armstrong and Densham co-authored a paper that was published in the Proceedings of the Thirteenth International Symposium on Automated Cartography (Auto-Carto 13). This paper describes a new way to decompose solutions to locational problems into a set of analytical objects (vectors and matrices) that are then used to summarize collections of solutions to locational problems. This "network map algebra" is also used to support visualization of solutions to such problems. Armstrong presented a paper on the role that emerging technologies will play on work practices by groups using GIS.
Armstrong and Densham have published several papers that relate to I-17; some recent publications are listed in the publications section of this report.
Armstrong has been awarded two grants, one from NSF ILI and the Geography and Regional Science program, the other from NASA. These projects involve investigating different aspects of the use of immersive visualization technologies in education and research. A particular focus of these projects concerns their use by groups of individuals. With the acquisition of an "ImmersaDesk" a group of five users can visualize three-dimensional solutions to environmental and socio-economic problems.
Initiative 19: GIS and Society: The Social Implications of How People, Space, and Environment are Represented in GIS
Initiative 19 is in its second and final year. In this report activities for the first eleven months of 1997 are identified. Most activities for this year have focused on implementing the research agenda identified from the Specialist Meeting held in Minnesota in March, 1996. The I-19 steering committee includes eight people at six sites (Helen Couclelis, UCSB; Michael Curry, UCLA; Trevor Harris, WVU; Robert McMaster, U. of Minnesota; David Mark, SUNY-Buffalo; John Pickles, U. of Kentucky; Eric Sheppard, U. of Minnesota; Daniel Weiner, WVU). 1997 reports for each site are as follows:
State University of New York at Buffalo - David Mark
The Critical History of GIS (CHGIS) Project submitted a proposal to the National Science Foundation, but learned in June that NSF declined to fund the project. David Mark has interviewed two important figures in the history of GIS, Thomas Poiker (TIN) and Donald Cooke (DIME), and those interviews have been transcribed. Mark presented an overview of the project at the UCGIS Annual Assembly in Bar Harbor in June; the presentation was well-attended and seemed well-received, and the paper was published in the electronic proceedings of the meeting. There were two GIS history sessions at the GIS/LIS meeting in Cincinnati in late October, also very well attended. The first session involved paper presentations by John Pickles, Pat McHaffie, and David Mark; Mark's paper appeared in the proceedings. The second session, organized by Tim Foresman, was a panel discussion including two authors of chapters in his recent book on the history of GIS, Nick Faust and Don Cooke, plus Ian McHarg.
The Initiative 19 working group on the critical history of GIS met in Buffalo November 6–9. Michael Curry, Ken Hillis, John Krygier, David Mark, Pat McHaffie, John Pickles, and Dalia Varanka participated, and Pickles, McHaffie, and Mark presented material to the SUNY Buffalo geography colloquium. The project members met with Hugh Calkins to talk about early GIS developments at the University of Washington, and with Mike Woldenberg to talk about the Harvard lab. Plans were made for a workshop proposal and a specific research project.
University of California-Los Angeles - Michael Curry
Curry attended planning meeting for the critical history project in Buffalo in November. He is now in the process of writing a grant proposal (for a conference, to NSF) on "Representing Terrain: From Bird’s Eye View to Bosnia" based on discussions at that meeting.
Curry is continuing work on the project on "Ethics of Spatio-Visual Representation". This fall, he continued collecting, analyzing, and sorting maps of health, crime, and environmental data, and looking at the literature on privacy, particularly with respect to groups, and expects to begin the next phase of the project in the spring. Curry received funding from the Academic Senate of UCLA to continue this project.
University of California - Santa Barbara - Helen Couclelis
Helen Couclelis participated in a panel discussion on "socio-economic research and GIS" as part of the GISDATA Final Conference, "Geographic Information Research at the Millenium", held in Strasbourg, France, 13–17 September 1997. The panel was chaired by Michael Wegener (Germany). The other panel members were Einar Holm and Sture Oberg (Sweden), and Munroe Eagles (US).
The panel addressed some of the ground covered by I-19: issues of democracy, equity, privacy and surveillance; the promise and threat of geodemographics; the emancipatory potential of grass-roots GIS; and the difference between academic and commercial views of GIS. Helen Couclelis focused her presentation on some differences between Europe and the US in how academics have approached GIS in the three domains of applied research, applications-oriented research, and theoretical/critical research, arguing that for political and institutional reasons, there has been much less of the latter (addressing theoretical issues of GIS and society) in Europe than in the US.
University of Minnesota - Robert McMaster and Eric Sheppard
The NCGIA Initiative 19 project at the University of Minnesota consists of two parts: (1) building a risk assessment model for the Twin Cities that integrates geodemographic, institutional, and hazardous materials information; and (2) determining how neighborhood organizations might utilize GIS technologies in decision making and conflict resolution.
Progress over the past six months has included the following for the risk analysis component of the project: (1) mapping of TRI toxic indices; (2) mapping of Petrofund, Superfund, and Land Recycling data (3) mapping of census and institutional data, including day care centers, schools, retirement homes, and homeless shelters; (4) city- and regional-level analyses; (5) Conversion of data from MapInfo to ARC format; (6) Spatial analysis; (7) Initial experimentation with simulation modeling, and (8) Submission of an Environmental Protection Agency proposal.
(1) Toxicity index. Using the Pratt Toxicity Index, developed at the Minnesota Pollution Control Agency, we have now mapped out all sites, given their relative toxicity. This provides a much different view of risk to these sites than a simple measure of proximity.
(2) Other hazardous materials now mapped for the City of Minneapolis include all Superfund, Petrofund, Land Recycling, and Pollution Permit Holder sites.
(3) Institutional/census data. We have now acquired, geocoded, and mapped all Day Care Centers, Schools, Retirement Homes, and Community Centers. Many census variables at the block, block-group, and census tract level have been mapped.
(4) Analysis. Analyses of TRI and both City and County level data have been completed. This analysis has been reported in the journal Cartography and GIS, and at several conferences. The results indicate that, in the City of Minneapolis, while there is not evidence of environmental inequity based on race, there is evidence based on class, where both TRI sites and lower income populations cluster together.
(5) Conversion to ARC format. In order to integrate plume-dispersion and other modeling techniques, all boundary files, census data, institutional data, and environmental data were converted to ARC format. We now have the ability to take our project on the road for demonstration purposes, yet complete higher-end analysis in a Unix computing environment.
(6) Spatial analysis. We are now computing the geodemographic characteristics around the buffered TRI sites at a variety of distances to establish the population characteristics of these regions. This type of analysis is in contrast to (4) above that looks at the characteristics of the enumeration units near each site.
(7) Simulation of TRI sites. A problem typically encountered in this type of equity work is in establishing what is equity. We are now generating a set of simulated TRI sites to establish a sampling distribution of random TRI sites. The question being addressed is: What would equity look like if TRI (and other) sites were randomly distributed?
(8) Submission of EPA Environmental Justice proposal. Last February we submitted a $350,000 proposal to the EPA's Environmental Justice Program. Although the proposal was not accepted, it was ranked highly and, after discussions with EPA, we were encouraged to resubmit in early 1998. We plan on doing so.
Report on NCGIA-Sponsored Workshop, Geographical Methodologies for Technological Risk Assessment - Robert McMaster, Susan Cutter (co-leaders)
On September 6-9, 1997, a group of experts on environmental risk assessment gathered together at SUNY Buffalo to hold a small workshop on geographical methodologies for technological risk assessment.
On Day 1 (Sunday), the participants discussed and debated an array of topics, including:
Review of research in risk assessment; critique of existing themes;
2. Can a comparable index be developed? What are some comparative methodologies?
3. How do time, scale, and resolution affect the measurement of risk and their spatial representation?
5. What should an idealized risk map be like? What set of maps are needed?
6. What is the place of local knowledge and neighborhood activities in the production of spatial information?
7. How do we incorporate the multiple perspectives of interested and affected parties, technical specialists, policy and decision makers into spatial methodologies for risk assessment?
2. Look into the potential for a broader Workshop on Spatial Methodologies for Hazard Exposure and Risk Assessment, funded by EPA, NIH, or NSF
3. Identify 3–4 four areas for possible comparative risk assessment research
4. Acquire sample data sets for testing spatial methodologies
5. Develop a comprehensive bibliography
6. Establish a mechanism for an internal Web-based review of papers
7. Created design guidelines for risk maps (collection of risk maps)
8. Review paper for GeoInfo Systems
9. Create a draft document, perhaps for NSF, that would address the following topics:
Most of John Pickles’ activities this past year have focused on completing two books on Eastern European transitions and not on Critical History of GIS (CHGIS) matters. However, parts of this work are related, and there are several projects tied directly to CHGIS. Most of the latter deal with epistemological and methodological issues related to the project, rather than work on the case studies. Publications from this work are listed in Appendix 4.
West Virginia University - Trevor Harris and Daniel Weiner
Activities during 1997 at WVU have focused on administration, networking, and research. During the year the I-19 Web Page was further enhanced and developed (http://www.geo.wvu.edu/I19/). This was the second and final year of the NCGIA project entitled "Local knowledge, multiple realities, and the production of geographic information: South Africa and West Virginia case studies". Early work focused on the Kanawha site which led to an AAG presentation and participation at the technological risk assessment workshop in Buffalo (see Minnesota report). However, the acquisition of a major NSF grant for the South Africa work, and the loss of Philip Burkhart to a teaching position, forced us to re-prioritize and focus on the South Africa research. Harris and Weiner hope to return to the Kanawha case study in the future.
On January 1st 1997 Weiner and Harris began a 30 month and $150,000 NSF-funded GIS and Society research project on South Africa. As a result, a graduate student was appointed to the project and summer field work was undertaken in Mpumalanga Province. Within South Africa, the project has been located within the Department of Land Affairs which is the agency responsible for the implementation of land reform. It is also a primary user of GIS technology. The major accomplishments of this work this year include: a reseach team of South African participants was identified; two field sites were selected; data and source materials were identified; appropriate institutional contacts at the local level were made. Furthermore, an experimental ‘community-integrated’ GIS is being developed for the central Lowveld District of Mpumalanga. Work is currently focused on traditional data coverages in preparation for field work in 1998. In addition conceptual development of a multi-media-GIS-WWW platform has been ongoing.
An extensive report on this I-19 topic appears under the report of the Panel on Geographies of the Information Society.
Initiative 21: Formal Models of Common-Sense Geographic Worlds
Co-leaders: David M. Mark and Max J. Egenhofer
The objectives of I-21 are: to identify basic elements of common-sense conceptualizations of geographic space, entities, and processes, and to develop an integrating framework; and to investigate GIS users’ reactions to intuitive geographic inferences, and compare the inferences with the results obtained with current GIS technology. The Specialist Meeting was held October 31 to November 2, 1996, in San Marcos, Texas. This meeting, which was attended by 42 researchers from North America and Europe, was organized and led by David Mark (NCGIA-Buffalo) and Max Egenhofer (NCGIA-Maine). The Department of Geography, Southwest Texas State University, and UB Geography Graduate F. Benjamin Zhan served as its local hosts. There were 48 applicants for this meeting, the largest number ever to submit to an NCGIA Specialist Meeting up to that time, and the backgrounds of the attending participants included such diverse fields as psychology, philosophy, computer science, and engineering, as well as geography.
The multidisciplinary Steering Committee for the Initiative included the following: Roger Downs, Geography, Pennsylvania StateUniversity; Andrew Frank, Geoinformation, TU Vienna, Austria; Janet Glasgow, Computing and Information Science (AI), Queen’s University, Canada; Pat Hayes, Computer Science (AI), University of Illinois; Dan Montello, Geography, Santa Barbara; Barry Smith, Philosophy, SUNY Buffalo; and Barbara Tversky, Psychology, Stanford.
The research agenda defined by the specialist meeting included 49 researchable questions. The report summarizing these was published in the NCGIA Technical Reports series in October 1997.
A paper summarizing the initiative goals and the Specialist Meeting findings was presented at the GIS/LIS'96 meeting in Denver (November 1996), and a lively panel discussion session on the topic was held at the Association of American geographers' meeting in Fort Worth, Texas (April 1997).
Discussions at the I-21 Specialist Meeting and in the Panel Session at the Association of American Geographers meeting led to considerable progress toward clear definitions of the differences between ‘Naïve’, or common-sense, (N-) and ‘Scientific’, ‘Scholarly’, or ‘Sophisticated’ (S-) versions of knowledge areas. The differences between N-Physics and S-Physics are fairly obvious, but the analogy does not so easily hold when trying to define S-Geography in contrast with N-Geography. Mark, Egenhofer, and Barry Smith are working to develop these ideas into a paper for publication.
David Mark (Geography) and Barry Smith (Philosophy) at Buffalo have been working toward establishing an "ontology of geographic entities", and testing the cross-cultural universality of such an ontology. A collaborative research proposal is being developed by Smith and Mark with Varenius Cognitive panel members Barbara Tversky (Stanford) and Stephen Hirtle (Pittsburgh). In the fall of 1997, considerable progress was made on the identification of an ontology of geographic water bodies.
Several student research projects at Maine (see next section) are pushing forward our knowledge of formal underpinnings for spatial representations. Several projects focus on image-schemata, which are expected to be at the core of cognitive universals for space.
Barry Smith (Philosophy/NCGIA, SUNY Buffalo) organized the two-day conference "History of the Concepts of Space", which was held at SUNY Buffalo, April 18–19, 1997. The purpose of the conference was to bring together philosophers, geographers and others working on space in order to explore the ways in which concepts of space have evolved since the first written evidence of spatial theories in the time of the Ancient Greeks. A special focus was on the evolution of concepts of space in the 20th century as a result of developments in the GIS field. Approximately 25 people attended, including Mark, Egenhofer, Smith, Adrijana Car (Pittsburgh and Vienna), Martin Raubal (I-21 participant and Maine graduate student), and Neil Pickard (another graduate student at Orono). A list of the papers presented at the conference is included under the conference section.
At the University of Maine, several master's and Ph.D. theses are under way in the area of common-sense reasoning. Kathleen Hornsby is working on her Ph.D.dissertation in the area of spatial reasoning and GIS. Her thesis on "A Change-Based Representation of Spatio-Temporal Phenomena" focuses on developing a classification of change based on object identity—a concept familiar in database and programming languages—and the operations that either preserve or change identity. The identification and formalization of the basic components of change lays the foundation for a new generation of formal data models that capture the semantics of change and lead to improved interoperability between GISs and process models or simulation software. Preliminary results of her research were reported at the UCGIS annual assembly in Bar Harbor, ME and she presented a paper at the COSIT ‘97 Conference in Laurel Highlands, PA; the latter paper was published in the refereed proceedings volume for the meeting.
Martin Raubal is working on his Master’s Thesis at Maine in the area of human wayfinding and spatial cognition. This thesis presents a methodology based on image schemata and affordances to structure people’s wayfinding tasks. Image schemata are recurring mental patterns (e.g., the CONTAINER or PATH schema) that people use to understand a spatial situation. They are highly structured and grounded in people’s experience. Affordances, offered by space and spatial objects, generate different human activities (e.g., a pathway affords moving). Airport space is used as a case study. We compare two selected airports in regard to the ease of performing a common wayfinding task. As a metric for comparing the two, the number of image schemata and the number of affordances that occur within the task in each of the airports are taken. Our assumption is that a passenger’s wayfinding performance in an airport increases when the number of image schemata and affordances per task decreases. Preliminary results of this work were reported at the COSIT ’97 conference.
Also at Maine, M. Andrea Rodriguez completed her Master’s thesis in June 1997 in the area of spatial reasoning. Based on a case study of a room space, she formally specified spatial relations derived from the behavior of objects associated with the image schemata container and surface. Preliminary results have been reported in the paper "Spatial relations based on Image Schemata" presented at the UCGIS annual assembly in Bar Harbor, ME and the paper "Image-Schemata-Based Spatial Inferences: The Container-Surface Algebra" also was presented at COSIT’97 and published in the refereed proceedings. She has continued working on this area focusing on the composition of spatial relations and a journal article is in preparation. Andrea is continuing as a Ph.D. student, working on Spatial Similarity funded by a grant from the National Imagery and Mapping Agency.
Roop Goyal is doing his Ph.D. dissertation at Maine on models of direction relations for extended spatial objects and reasoning about their similarities. Common models for direction relations, which are either point based or use the approximation of minimal bounding rectangles, are counter-intuitive particularly for non-convex objects. Roop has designed a formalism that captures the semantics of direction relations at different levels of detail and has developed a formalism to derive the composition of such direction relations. This work will be extended to compare direction relations not only for equivalence, but also for similarity. A paper on some early results was published in the electronic proceedings of the UCGIS annual assembly in Bar Harbor, ME and a journal article is under preparation.
John Florence finished his Master’s thesis at Maine on the estimation of the frequency of topological relations in spatial data sets. Such estimations are invaluable meta-information for the optimization of spatial query processing. He distinguished between complete partitions, hierarchically structured regions, and overlapping regions and compared for three regular space partitions (triangular, square, and hexagonal) the frequencies with those of a wide range of datasets. The result is a set of equations with which the estimated distribution of the frequency of topological relations is calculated. The frequencies converge with large data sets, but may be significantly different for small (< 50 objects) datasets.
At Buffalo, Leo Zaibert continues to work toward his Ph.D. in the Department of Philosophy under the direction of Barry Smith. In the summer of 1997, Zaibert was funded by the Varenius project to work on "Landed Property, Cadastral Registration, and Geographic Information Science", a component of his dissertation work. The work includes, in part, cultural differences in concepts of ‘land’, a key element of Naive Geography.
With Zaibert's help and David Mark as a co-PI, Barry Smith submitted a proposal entitled "Land As a Social Artifact: A Cross-Disciplinary Study in Law and Geography" to the National Science Foundation. Unfortunately, NSF declined the grant, but aspects of the work continue. Smith is working on a book on this topic, tentatively entitled "The Metaphysics of Real Estate".
Barry Smith is involved in two other projects under Initiative 21. One is being conducted with Roberto Casati (CNRS/CREA, Paris), an I-21 workshop participant, on universals of spatial cognition and formal models of spatial objects, relations and processes, leading toward a book to be called "Space: A User's Guide". With Achille Varzi (Columbia University), Smith is working on axiomatic theories of mereotopology. and especially dealing with various kinds of geographic boundaries.
Also at Buffalo, Smith and Mark also have continued to work on geographical oddities, such as the non-contiguous towns of Baarle-Nassau and Baarle-Hertog in the Netherlands and Belgium, in relation to the limits of possible geographic ontologies.
I-21 will likely continue until a 1-day closing workshop just before or after COSIT ‘99. in the vicinity of Hamburg, Germany. At this time, the workshop is expected to be self-financed by the participants, based on their travel money to COSIT. We are planning to use this workshop as a key stage in the production of an edited book on Naïve Geography.
Under SBR 88-10917, NCGIA initiated a series of collaborative projects with institutions outside the immediate three-site consortium. Three of these projects remained active in 1997; their activities are reported below.
Gulf of Maine Data and Information Management System
Kate Beard, University of Maine, PI.
The Regional Marine Research Program (RMRP) for the Gulf of Maine has been working toward a suite of models that collectively simulate how the Gulf of Maine ecosystem and its interacting components function naturally and under stress. Given the complexity of the research, its scope and duration, the need for real time interaction among disciplines, institutions, and PIs the RMRP felt the need for a carefully designed approach to data management and information flow. In 1995 the Gulf of Maine RMRP funded a collaborative team to develop a Data and Information Management System (DIMS) for the Gulf of Maine that would ensure timely and easy distribution of both new and archived data, the highest standards of quality assurance, feedback to the field program, and permanent archiving of a large volume of new RMRP data and information.
The purpose of the DIMS was to allow RMRP researchers to determine what information is available, to acquire data or subsets of data easily, and to analyze the data without regard to where it resides and in what form. The goal of the NCGIA collaboration on this project was to work with the PIs and other marine scientists on the DIMS project on issues specifically relevant to spatial data/information distribution and analysis. The objectives were to address some of the unique issues of searching, browsing, retrieving, and analyzing geographically referenced data sets. Specifically the collaboration:
Predator-Prey Modeling of Fish Populations within Three-Dimensional GIS:
Ling Bian, Hugh Calkins, Joseph DePinto, PIs. Stephan Brandt, Collaborator
The project "Predator-Prey Modeling of Fish Populations within Three-Dimensional GIS" is an NSF-funded project conducted in collaboration with Dr. Stephen Brandt at Great Lakes Center, Buffalo State College. The PIs of the project are Hugh Calkins, Joseph DePinto, and Ling Bian at the NCGIA Buffalo site. The primary objective of the research is to develop more realistic spatial models for predator-prey populations in Lake Ontario. The collaborator institution has developed spatial models that predict potential growth rates of predator populations in a two dimensional space. The current research effort is to adopt the existing models to a three dimensional environment and extend the models to simulating the spatial and temporal dynamics of the predator-prey populations.
One of the key research objectives was to interpolate a two dimensional data set into a three dimensional environment. The data available were in a matrix of two dimensional transects in Lake Ontario. The transects are arranged in both north–south and east–west directions. Because of the pattern differences in sampled data, the work was conducted in two studies, one for prey density data which were collected using underwater acoustics, and another for water temperature and plankton data collected using OPC technique. Both sets of data are basic input information for fish modeling.
The first of these two studies was to identify the optimal method to interpolate the prey data to a three dimensional data set. Geostatistical and GIS methods were used to analyze the data. Given the preliminary observations, different spatial interpolations were experimented with. The second study was to interpolate water temperature and plankton data. These two types of data are sampled at the same time and location with the prey data but in a different pattern. The sample data display a undulating pattern within each transact, posing a challenge for spatial interpolation. Linear, inverse distance, and Kriging methods were used to interpolate each transect in order to generate a series of continuous, two dimensional data set. Different search strategies, diel sampling, and isotropies were applied to the interpolations. Cross validation measurement of mean absolute error, mean squared error, and minimum and maximum error are used to compare accuracy between interpolations. The preliminary results of both studies were presented at the 1997 annual meeting of the Association of American Geographers at Fort Worth, Texas. A complete report was presented at the UCGIS conference held at Bar Harbor, Maine. A manuscript describing the work has been submitted to a journal.
The last objective of the project is to extend the current two-dimensional, static fish growth models to a more realistic, three dimensional framework. The three dimensional data interpolated previously were used in this study. An object-oriented framework is used, in combination with a bioenergetics model and movement rules, to simulate the growth of individual fish. The simulation implemented the behavior of individual predator fish in their selection of habitat and their subsequent growth in the three-dimensional environment. Results of this work were presented at the GIS/LIS '97 conference. One manuscript documenting the work has been prepared, and a second is in preparation.
The project concludes on December 31, 1997.
Worlds of Information: the Geographic Metaphor in the Visualization of Complex Nonspatial Information
Helen Couclelis, PI
The purpose of this collaborative project with a group of researchers from the Pacific National Laboratory (PNL) is to investigate the potential of the geographic metaphor in the design of visualizations of complex non-spatial data. The advantages of the geographic metaphor are twofold. First, geographic space constitutes a generally accessible and very comprehensive realm of experience for humans. This minimizes the amount of learning effort necessary for people to make sense of the vast array of possible geographically-structured representations. Second, over the more than two millennia of its existence, geography has developed a formidable arsenal of theories and tools for the representation and analysis of spatial information. Moreover, the quintessential presentation format of spatial information—the map—has always been visual. GIS, the electronic-age addition to this long tradition, successfully integrates the theories and tools of geography and its preferred, visual mode of data presentation, with the computer's capabilities of data storage, manipulation and retrieval. Thus GIS can serve as the technological bridge for moving the geographic metaphor over to the domain of general data representation and analysis.
A first paper based on this research was presented by the PI at the 1997 UCGIS Annual Assembly held in June in Bar Harbor, Maine. The paper investigates the possible significance of certain fundamental geographic concepts, such as place, way, and region, in an information space that is only metaphorically geographical. These metaphorical meanings are then associated with specific functions and user actions that make sense in the context of exploring and searching an information space consisting of textual document objects. The work continues by extending the vocabulary of geographic concepts that can be given practical interpretation in terms of interacting with complex nonspatial information databases.
Currently, the PI has submitted a manuscript (same title as the project) to the refereed journal Cartography and GIS, and is working on the followup paper to the one presented at the Bar Harbor assembly.
The Amherst Deer Problem: Nature and Society in the Suburbs
David Mark has been involved in studying nature–society conflicts involving White-tailed Deer in a suburban community (Amherst, NY), and the role that geographic information plays in decision-making and public participation. The deer problem is also the Ph.D dissertation topic of Steven Parkansky, who is working under Mark's supervision. Amherst is an affluent Buffalo suburb with a population of about 130,000 people and about 1,100 White-tailed Deer. There is no legal hunting in the Town, which has experienced a great deal of development and population growth over the last couple of decades. "The Amherst Deer Problem", as reported in the local press, began as concern for deer damage to individual and commercial gardens, but quickly switched to a focus on deer-related vehicle accidents (DRVAs). Geographic information is playing an important role in the debates over methods for reducing the chance of DRVAs. In 1996, there were over 500 DRVAs reported to the police, almost 10 percent of all the vehicle collisions reported to the Amherst police. Parkansky obtained deer population survey data from the New York State Department of Environment and Conservation, DRVA data from the Amherst police, deer warning sign data from the Amherst highways department, and deer carcass locations from the contractor who picks up road-killed deer for disposal. All of these data were digitized or geocoded, integrated in ARC/INFO, and moved to ArcView. Biogeography students also prepared a detailed land cover map from air photos, and that also was integrated into the GIS. In April 1997, the Town Supervisor (‘mayor’) invited Mark to serve on the "Deer Management Task Force", which has been charged with advising the town of deer population targets and means of reducing numbers if necessary. One meeting of the Task Force was held in UB’s Geographic Information and Analysis Laboratory, and was attended by the Supervisor, as well as by two members of the Information Systems department of the town. The politicians and other task force members seemed to be surprised by many aspects of their own organization’s data, when shown in a geographic context. One task force member stated: "At least now we know there really is a deer problem". Parkansky's dissertation will model deer population distributions, then use that output plus traffic and road data to model deer-vehicle collisions. He also will analyze the 240 questionnaires (out of 1200 mailed) that asked questions about the nature of the deer problem, respondents’ personal experiences with and losses due to White-tailed Deer in Amherst, and their attitudes toward various prevention or mitigation solutions.
Research at Buffalo in the human capital area has followed on from the NSF/NCGIA sponsored workshop Geographic Information Analysis and Human Capital Research, held in 1995, and led by Munroe Eagles and Hugh Calkins. Focusing primarily on the issue of "building strong neighborhoods", research is currently being conducted by Eagles and Calkins on the impact of block clubs and other police outreach activities on neighborhood safety and stability. A currently funded project under the Community Oriented Policing Services (COPS) program is investigating the potential for geographic information technologies to assist in the reduction of crime and community disorder problems in neighborhoods.
Another research project, led by Ezra Zubrow, is developing a broad-based GIS crime analysis application. The use of GIS greatly enhances the ability of law enforcement agencies to visualize patterns of criminal incidents, and makes possible the optimization of police resources given community goals. The design of the software provides predictive modeling and algorithms to facilitate proactive policing. The software is being developed in collaboration with various police departments and the Environmental Systems Research Institute, a leading GIS software vendor. The tools include computerized beat books, gang territory demarcators, drug market analyzers, beat optimization, officer safety monitors, and others.
Spherekit is a spatial interpolation software toolkit developed at NCGIA as part of Initiative 15: Multiple Roles of GIS in U.S. Global Change Research. Developers include: Cort Willmott (University of Delaware), Rob Raskin (Jet Propulsion Laboratory), Chris Funk (NCGIA), Scott Webber (University of Delaware), and Mike Goodchild (NCGIA). The alpha (preliminary) version was released into the public domain in October 1996. Version 1.0 will be released in 1997. The source code is freely distributed over the internet. The package features several unique capabilities.
Spherekit allows interpolation over continental or global domains by computing distances and orientations (among data and interpolation points) from geodesics on the surface of the globe. Conventional interpolations typically are based upon Euclidean distance in Cartesian 2-space which involve planar projections that produce distortions of some kind. In Spherekit, projections are applied only for display purposes after the interpolation has been carried out using spherical geometry. Users can select from several interpolation algorithms that have been adapted to the sphere: inverse distance weighting, thin plate splines, multiquadrics, triangulation, and Kriging. Portions of the GSLIB package have been modified for the sphere and are used in Spherekit to compute variograms for the Kriging algorithms.
Spherekit enables the user to incorporate knowledge or information about the processes that produce the underlying spatial variations into the interpolation model. A built-in equation editor and a collection of nonlinear transforms allow the user to create and experiment with new, physically meaningful variables from the independent and dependent variables available. This ‘smart’ interpolation capability allows Spherekit to intelligently interpolate using auxiliary information. One use of the smart interpolation feature is to incorporate elevation information when interpolating variables that are correlated with height. A digital elevation model (DEM) is included with the package for this purpose.
Error analysis is an integrated component of Spherekit making the package particularly useful for comparing interpolation methods and parameters. Interpolation method performance is measured using cross-validation. Cross-validation error is defined at each observation point as the difference between its actual value and its interpolated value estimated from the remaining points. The resulting error field can be displayed either at the data points or by interpolating to a regular grid to reduce spatial biases. Error difference fields, comparing methods or parameter settings, can be created and displayed with ease.
Spherekit software is compatible with most computers running the Unix operating system. The software uses Tcl/Tk for its Graphical User Interface (GUI), Generic Mapping Tools (GMT) for display of output fields, Ghostview for display of PostScript files, and netCDF for storing the DEM. All of these auxiliary packages are required and can be downloaded together with Spherekit. The package is available through the ‘Products’ section of the NCGIA web site www.ncgia.org.
The Vehicle Intelligence Testing & Analysis Laboratory (VITAL) was established at NCGIA in January 1997, as a testbed for spatial data interoperability in Intelligent Transportation Systems (ITS). Startup funding for the lab was provided by the Testbed Center for Interoperability (TCFI) of the California Department of Transportation. Continuing funding is imminent, from the Federal Highways Administration (FHWA) through Oak Ridge National Laboratory (ORNL).
The mandate of the lab is to examine theoretical and industrial issues in GIS as they relate to ITS. Of immediate concern is the ITS community need for solutions to spatial data interoperability problems. Vehicle and incident location are central data items in ITS. In a world of competing vendors of street network data—and inevitable discrepancies and errors in position, street naming, addressing and classification—the success of ITS hinges on the ability to communicate a location message unambiguously across dissimilar map bases. Prototype vehicle tracking and navigation systems have been in circulation for several years; what is new about the current set of problems is communication of location between systems, i.e. interoperability.
There are two components to this research. First, at the industrial level, datum and messaging standards are being proposed by organizations such as the Society of Automotive Engineers (SAE) and the International Organization of Standards (ISO). These need to be tested comprehensively in laboratory and field simulations—this is becoming a requirement for national standards.
Second, there are research questions raised by ITS problems, e.g. methods for interpreting locational messages so as to reduce ambiguity; minimization of the cost of infrastructure such as coordinate datums. Over the first half of 1997, with TCFI funding, VITAL created a testing infrastructure consisting of moving map software running on a laptop computer carried in a vehicle, with continuous location readings from differential GPS, a Distance Measuring Instrument (DMI), and two-way wireless communications with a fixed server using Cellular Digital Packet Data (CDPD) exchange. Six commercial databases of street centerlines were acquired for Santa Barbara county for the purpose of cross-vendor comparison. In terms of this infrastucture alone, VITAL is a unique facility. Since August 1997, VITAL has been engaged in contract research funded by Viggen Corporation, Tennessee, to test the Cross Streets Profile of the Location Reference Messaging Specification (LRMS), a standard under consideration by SAE. The test has required field surveys using the test infrastructure, and lab simulations, to identify the types of error, and measure their severity, when transmitting a location from one database to another using the Cross Streets Profile. On the academic front we are studying positional discrepancies between centerline databases, to develop means of measuring and visualizing error. This will lead to methods for real time correction of this error with a minimal set of control points. This project currently involves four graduate students and three faculty—Goodchild, Church, and Clarke—with specialties in GIS, operations research and cartography. There are commonalities between this and other research in spatial error, for which the Department of Geography has recently received funding from the National Imagery and Mapping Agency (NIMA). Further details on VITAL research are posted on our web site, http://www.ncgia.org/vital.
The NCGIA sites offer numerous opportunities for visitors for periods from a few hours to many months. Reports on the role of visitors can be found at many places in this document, notably in Appendix 7, and in the sections below.
Phaedon Kyriakidis, a Ph.D. student in geostatistics at Stanford University, was a Visiting Researcher at NCGIA from 06/16/97 to 08/22/97. He worked with several NCGIA faculty and graduate students, and participated in the NIMA-funded project Uncertainty in Geospatial Information Representation, Analysis and Decision Support. Other long-term visitors included Sachio Kubo, Keio University, Japan; Lola Gulyamova, Tashkent State University; Val Noronha, Digital Geographics; William Albert, Boston University; Wilmar Amaya, IPC, Ecopetrol, Colombia; Masatoshi Arikawa, Hiroshima City University, Japan; Kurt Brassel, University of Zurich; Germana Manca, Centre for Advanced Studies, Research & Development, Sardinia, Italy; Micha Pazner, University of Western Ontario; and Tapani Sarjakoski, Finnish Geodetic Institute.
Several visitors worked with Hugo Loaiciga during 1997. They included Dr. Ramanaran Yadava, Regional Research Laboratory, Bhopal, India; Eli Skop, National Environmental Research Institute, Copenhagen, Denmark; Dr. Jong Dae Kim, Department of Mining and Mineral Engineering, Dong-A University; Dr. Takeshi Sato, Department of Civil Engineering, Gifu University, Japan; Dr. Kyoo-S. Lee, Department Landscape Architecture, Sung Kyung Kwan University, Suwon, Korea; Shih Hsing Yang, Department of Geography, National Technical University, Taipei, Taiwan; Dr. Igor Zektser, Fulbright Scholar, Russian Academy of Sciences, Moscow; and Dan Rogers, Clayton Environmental, Detroit, Michigan.
The following individuals visited the Buffalo site in 1997 as participants in the Visiting Fellows Program:
Long-term visitors funded by the visiting scientists program included Dr. Robert Rugg, Virginia Commonwealth University and Dr. Carol Bult from the The Institute for Genomic Research. Dr. Rugg worked during summer 1997 on the formalization of attribute similarity. Based on his work, he is organizing a special session at the AAG in Boston with papers by Werner Kuhn, Andrea Rodíguez, and Linda Hill. Dr. Bult worked with NCGIA researchers Beard and Egenhofer on the application of GIS modeling techniques to genome data. She was successful with a proposal to NSF’s Biological Infrastructure program and organized a 2-day workshop on Spatial Genomics (see description elsewhere in this report). Other Maine visitors are listed in Appendix 7.
NCGIA continues to provide leadership in GIS education though sponsorship of GIS education conferences and the development of GIS curriculum-building materials. Over the years NCGIA has supported GIS education efforts at all levels. In recent years the focus has primarily been on the rapidly expanding area of GIS in the community colleges. Flexibility in these institutions has led to a number of imaginative approaches to GIS education including merged transfer/technican programs, short courses, contract education, and distance learning. The NCGIA education program has been working closely many of the colleges attempting to provide GIS-based education opportunities.
The Secondary Education Project, which has worked with early K-12 adopters of GIS, continues to function, though at a more limited level. Hardware cost reductions, software ease-of-use improvements, and the option for GIS functionality served up over the Internet, are all increasing the possibility of widespread use of GIS in the schools. There are still needs for curruculum-sensitive GIS learning materials and teacher training. The NCGIA is collaborating in the early stages of design of a potential project in this area.
Additional information about NCGIA education projects can be found below and on our education World Wide Web home page (http://www.ncgia.org/education/ed.html).
The NCGIA Core Curriculum in Geographic Information Science
Progress on the GISCC in 1997 has been slow. The outline listing 170+ units, the editorial structure, website, formatting guidelines and editorial assignments were all completed in 1996. Most editors had identified the authors for the individual units in their sections by early 1997. Progress in the current phase of unit writing is dependent upon a great many very busy academics around the world. Unfortunately, we have found it very difficult to encourage these authors to move the GISCC to the top of their priority lists. While the GISCC is envisioned as a GIS community project, NCGIA faculty have produced a few units which we hope will encourage the rest of the community to get working. As well, we have significantly relaxed the formatting requirements for submitted units—rather than requiring contributors to submit HTML-formatted materials, we now accept any text format, including ASCII, which we now convert to HTML and edit to match our standard GISCC format. We believe we have now managed to move the project higher on priority lists and reportedly should receive a flood of completed units in the new year. In November 1997, completed units are arriving at a rate of about 2 per week. We now hope to have a well-filled website by Spring 1999. At that time, a number of related projects such as developing the navigation and curriculum builder tools will be revived. The GISCC can be found on the web at http://www.ncgia.org/giscc.
The Core Curriculum for Technical Programs
NCGIA has continued its work begun in July, 1996 under a two-year grant from the National Science Foundation Advanced Technological Education (ATE) Program. The grant funds the development of a core curriculum in GIS for two-year colleges. This GIS Core Curriculum for Technical Programs (CCTP) is related to the Core Curriculum in Geographic Information Science, but is focused more towards technician training in the community colleges. It approaches the information from the perspective of what the practitioner needs to be able to do, rather than just what they need know.
Like the other Core Curriculum, it is World Wide Web-based and includes access to a number to resources for GIS instructors in technical programs. Currently the CCTP is beginning its testing phase. Curriculum units are being posted to the Web site and will be used by evaluators/resource developers to create resource materials for the CCTP such as lesson plan sets, student exercises, short course outlines, example applications, web tutorials, and multimedia leaning meaterials. The CCTP also is collecting and referencing various resources for GIS instruction in the community colleges such as model course outlines, lists of available text and lab materials, and links to useful Web sites. The CCTP is on-line at http://www.ncgia.org/cctp.
The Third International Symposium on GIS in Higher Education GISHE'97 was held in Chantilly VA, Oct 30–Nov2, 1997 with almost 200 GIS educators from all education levels and from 15 countries attending. This conference was co-sponsored by Towson State University (who handled all logistical details) and NCGIA (who organized the program). Over 60 papers were presented on topics ranging from "Teaching GIS across the Curriculum" to "GIS education and the Internet", "Capacity building in developing countries", "Designing GIS courseware", "Foundations for GIS education", "GIS in Community Colleges" and "Teaching GIS in K-12". Scholarships provided by ESRI, Intergraph, UNIGIS and FGDC covered travel expenses for 6 international educators and one community college instructor. The conference was structured to include indepth discussion sessions. On Sunday morning a final plenary and working groups produced a conference summary. The conference program and summaries with action items can be seen at http://www.ncgia.org/conf/gishe97.
The NCGIA helped organize and present a one day meeting on GIS in the community colleges at the University of California, Los Angeles. This meeting was for college instructors interested in GIS education at their institutions. The meeting led to a California state grant administered by the North Orange County Community College District which provides 4 day GIS training sessions for a total of 120 college faculty around the state.
Although UCGIS is an activity beyond the NCGIA, members of NCGIA played significant roles in this summer's UCGIS meeting. Harlan Onsrud (UMaine) and Karen Kemp (UCSB) were the co-chairs for the conference which combined paper presentations on research in GIScience with working sessions on GIS education. Dr. Onsrud was successful in obtaining funding from NSF to provide limited travel support for 45 graduate students at US institutions who actively participated in the program. He also organized a series of short workshops on fundamental GIScience themes. Dr. Kemp, working with Richard Wright of San Diego State University, organized the plenary and working sessions which culminated in the development of the UCGIS’s education agenda.
As noted above, two-year colleges are an important and rapidly growing sector for GIS education and training. GIS courses and programs exist or are being developed in hundreds of colleges in North America. The NCGIA continues to support efforts to provide GIS learning experiences in the colleges through the conferences and curriculum efforts listed above, especially the CCTP. In addition, Steve Palladino, the NCGIA Education Projects Manager, in conjunction with partners at ESRI, is keeping a master list of community colleges with GIS instruction and parties interested in community college GIS. NCGIA-Santa Barbara members have provided advisory support in the development of a GIS Program at the local community college, Santa Barbara City College. The lessons learned in this exercise have benefited the CCTP development efforts.
Computational Methods for Watershed Analysis CD
A related project, not funded under Varenius but mentioned in our original proposal, is a course on "Computational Methods for Watershed Analysis", prepared by Prof. Leal Mertes of UCSB with support from UCSB Instructional Development. This course uses GIS (ARC/INFO) as a tool to support the instruction of watershed analysis. Extensive student laboratory materials have been developed and tested in the classroom. Discussions on collaboration with ESRI to produce a CD containing all of the necessary data and AML files are currently underway.
The Buffalo site has been working with faculty at Erie Community College (ECC), a regional two-year SUNY institution, to develop a program to teach GIS courses. In May of 1997, Hugh Calkins and David Mark helped ECC’s Jason Steinitz prepare a proposal for external funding to develop a GIS curriculum and teaching program. This proposal, which was submitted to the National Science Foundation’s Education and Human Resources directorate, includes Hugh Calkins and NCGIA Buffalo as a participants in the project.
At UCLA, Michael Curry has developed a new course, entitled "Place, Identity, and the Networked World," that grows out of I-19. Curry is also working with Larry Smith (Geography, UCLA) to revise the introductory GIS course, to include weekly discussion sections (led by Curry) on social/cultural/ethical issues.
Geographical Methodologies for Environmental Risk Assessmen, September 6–8, 1997, SUNY at Buffalo, Buffalo, NY
NCGIA Buffalo hosted the invitational workshop "Geographical Methodologies for Environmental Risk Assessment", September 6–8, 1997. This workshop, which was led by Robert McMaster (Geography, University of Minnesota) and Susan Cutter (Geography, University of South Carolina), explored the state of current research in risk assessment and the ways that GIS use and mapping can influence risk assessment and public perceptions of risk from hazardous materials. The outcomes of the workshop, which was part of NCGIA's Initiative 19 (The Social Implications of How People, Space and Environment are Represented in GIS), are discussed in the Research section of this report. Robert McMaster delivered a public lecture on the outcomes of the meeting on September 9.
Visiting Fellow participants included Marc Armstrong, University of Iowa; Susan Cutter, University of South Carolina; Trevor Harris, West Virginia University; Robert McMaster, University of Minnesota; Mark Monmonier, Syracuse University; Timothy Nyerges, University of Washington; Jeffrey Osleeb, Hunter College-CUNY; Michael Scott, University of South Carolina; Eric Sheppard, University of Minnesota; and Daniel Weiner, West Virginia University. SUNY Buffalo participants included Professor John Krygier, and PhD candidate Elizabeth Kent.
First Annual Assembly of the NCGIA Consortium, Oct 10–13,1997, SUNY at Buffalo, Buffalo, NY
This gathering brought together researchers from the three NCGIA sites to discuss their research projects, share information and approaches, and make plans for future consortium activities.
Geographic Information Systems and Political Districting: Social Groups, Representational Values, and Electoral Boundaries, October 24–26, 1997, SUNY at Buffalo, Buffalo, NY
The purpose of NCGIA’s conference on GIS and Political Districting was to bring together political scientists, philosophers, geographers, electoral cartographers, and legal experts to discuss current controversies in the process of political redistricting. Recent advances in GIS have revolutionized the process of drawing district boundaries, and have opened new possibilities for both the achievement of representational goals and for potential abuse. The conference was led by Munroe Eagles (Political Science/NCGIA, SUNY Buffalo). Co-sponsorship was received from NCGIA-Buffalo, SUNY Buffalo’s Conferences in the Disciplines, the Faculty of Social Sciences, the Office of the Provost, and the Department of Political Science. More than 60 persons attended, and over 25 papers were presented in the eight sessions of the conference.
Visiting Fellow participants included Kimball Brace, Election Data Services; Bruce Cain, UC Berkeley; Carmen Cirincionne, University of Connecticut; Michael Curry, UCLA; Chandler Davidson, Rice University; David Ely, PacTech; Jonathan Entin, Case Western Reserve University; Christian Grose, University of Rochester; Winett Hagens, Norfolk State University; Lisa Handley, Election Data Services; Richard Katz, Johns Hopkins University; Jonathan Leib, Florida State University; James Lennertz, Lafayette College; Karin MacDonald, UC Berkeley; Richard Morrill, University of Washington; Richard Niemi, University of Rochester; Timothy O’Rourke, University of Missouri St. Louis; Mark Rush, Washington and Lee University; Harold Stanley, University of Rochester; Edward Still, Edward Still Law Firm; Donley Studlar, West Virginia University; Jon Taylor, University of Kentucky; Ronald Weber, University of Wisconsin, Milwaukee; Jerry Webster, University of Alabama, Rudolph Wilson, Norfolk State University.
Qualitative Research in Human Geography, November 6–9, 1997, SUNY at Buffalo, Buffalo, NY
Led by Geography Assistant Professor Meghan Cope, this small discussion group involving Lynn Staeheli of the University of Colorado, Boulder and Victoria Lawson of the University of Washington, Seattle, who met to discuss the development of curriculum materials and potential workshops to train geographers in qualitative methods. Professors Lawson and Staeheli were Visiting Fellow participants.
GIS History Project Workshop, November 6–8, 1997, SUNY at Buffalo, Buffalo, NY
The GIS History Project is a multi-investigator study of the history and development of GIS and related technologies. Key innovations will be related to social, economic, and institutional contexts, and to contemporary development of computing technologies more generally. The project is described in more detail in the I-19 research section of this report. The objective of the project workshop was to assess progress on the project and to plan and prioritize future activities. The workshop was be led by David Mark (NCGIA Buffalo) and involved John Krygier of Buffalo. Visiting Fellow participants were Michael Curry (UCLA), Ken Hillis (University of North Carolina), Patrick McHaffie (DePaul University), John Pickles (University of Kentucky), and Dalia Varanka (Independent Scholar).
On November 7, the panel delivered an open colloquium on the Critical History of GIS.
GIS in Epidemiology, November 14–16, 1997, SUNY at Buffalo, Buffalo, NY
Professor Michael Woldenberg led this NCGIA-sponsored workshop, which featured presentations on research in Breast Cancer, Lead Poisoning, and Asthma, and the role GIS could play in furthering these efforts.
Visiting Fellow participants included Cynthia Brewer, Pennsylvania State University; Susan Cutter, University of South Carolina; Erin O’Leary, SUNY at Stony Brook; Linda Williams Pickle, Center for Disease Control and Prevention; Gerard Rushton, University of Iowa; David Stea, Southwest Texas State University; Deborah Thomas, University of South Carolina; Dan Wartenberg, Environmental and Occupational Health Sciences Institute; and Carmelle Cote, ESRI. A number of people from SUNY Buffalo’s Department of Geography, Department of Social and Preventative Medicine, and Department of Medicine also participated. The meeting also featured a public lecture by Erin O’Leary and Gerard Rushton. There were also presentations from Cindy Brewer and Linda Pickle
History of the Concepts of Space, April 18–19, 1997, SUNY at Buffalo, Buffalo, NY
Barry Smith (Philosophy/NCGIA, SUNY Buffalo) organized the two-day conference "History of the Concepts of Space", which was held at SUNY Buffalo, April 18–19, 1997. The purpose of the conference was to bring together philosophers, geographers, and others working on space in order to explore the ways in which concepts of space have evolved since the first written evidence of spatial theories in the time of the Ancient Greeks. A special focus was on the evolution of concepts of space in the 20th Century as a result of developments in the GIS field (see Initiative 21). Approximately 25 people attended.
A number of papers were presented at the conference, including: Istvan Bodnar (Philosophy, Buffalo/Budapest) "Ancient Theories of Space"; Paul Cornish (Political Science, Buffalo) "St. Augustine's Critique of Empire: A Defense of Small Republics"; Barry Smith and Leonardo Zaibert (Philosophy, Buffalo) "The State as Work of Art: Theories of Space in Baroque Philosophy; Timothy Engstrm (Philosophy, Rochester Institute of Technology) "Virtual Art and the History of Space"; David Zubin (Linguistics/Cognitive Science, Buffalo) "Space in Language and Child Development"; David Mark (Geography and NCGIA, Buffalo) "Geographic Space in Cognitive Science"; Max Egenhofer (Computer Science and NCGIA, Maine) "The History of Concepts of Space in Geographic Information Science"; Adrijana Car (Geoinformation, Vienna/Pittsburgh) "The History of Spatial Hierarchies"; Achille Varzi (Philosophy, Columbia) "Theories of Space in Formal Philosophy"; David Koepsell (Philosophy, Buffalo) "The Metaphysics of Cyberspace".
NCEAS Workshop on Uncertainty in Spatial Data for Ecological Models, September 29–October 1, 1997, Santa Barbara
This workshop was part of a longer term project directed by Carolyn Hunsaker of the Oak Ridge National Laboratory, and funded by the National Center for Ecological Analysis and Synthesis. Approximately 20 people—ecologists, geographers, and statisticians–met at the NCEAS facility in downtown Santa Barbara to discuss issues related to the impacts of spatial data uncertainty in ecology. The purpose of the workshop, which was co-organized by NCGIA and represents the most significant collaboration between the two centers to date, was to develop plans for a book, to be published later in 1998. Further information on the project is available at the NCEAS web site, wwww.nceas.ucsb.edu.
UCGIS 1997 Annual Assembly and Summer Retreat, June 15–21, 1997, Bar Harbor, ME
NCGIA-Maine and the College of the Atlantic hosted the "UCGIS 1997 Annual Assembly and Summer Retreat", June 15–21, 1997, Bar Harbor, ME. Organized by Harlan Onsrud and Karen Kemp, with the assistance of Blane Shaw, the conference attendance exceeded 160 participants. Two primary goals of the meeting were (1) to disseminate advanced knowledge on geographic information science and technology topics among leading researchers, professors, and graduate students in the field, and (2) to identify the nation's highest priority education needs for advancing geographic information science and develop initiatives to respond to those needs. The assembly began with a full day session devoted to identifying high priority geographic information science education needs and creating working groups to work on those needs throughout the week. During the remainder of the week, morning sessions began with a plenary paper session followed by concurrent paper sessions on a range of geographic information science topics. The paper sessions were presented by graduate students, faculty and researchers from across the nation. Working group sessions addressing high priority education issues also occurred during the mornings. Afternoon sessions were dedicated to attending optional advanced workshops or enjoying the outdoors, Acadia National Park, and island life in general with fellow attendees.
Workshop on Spatial Genomics, October 23–25, 1997, Northeast Harbor, ME
NCGIA-Maine and the Asticou Inn hosted the "Workshop on Spatial Genomics", October 23–25, 1997, Northeast Harbor, ME. Organized by Carol Bult and Kate Beard, with the assistance of Blane Shaw, the purpose of the meeting was to broaden the discussion of the idea of spatial genomics by bringing together a small group of scientists from the genome and GIS domains. The goals of the Spatial Genomics Workshop were to refine theoretical concepts related to the GenoSIS project (NSF funded "Application of Spatial Concepts for Genome Data" project) and to lay the groundwork for the path from concept to application.
At Buffalo, Hugh Calkins continues his work with a working group of the New York State Legislature to look into GIS standards and development for New York State. Calkins is chair of the committee on Standards. The following NCGIA publications appeared in 1997 to date:
97-1: Multipurpose Land Information Systems Development Bibliography: A Community-wide Commitment to the Technology and its Ultimate Applications, by David L. Tulloch, Bernard J. Niemann, Jr., and Stephen J. Ventura, Univ of Wisconsin - Madison; and Earl F. Epstein, The Ohio State University, brings together literature covering the many aspects of MPLIS development, including initiation, implementation, operation and maintenance.
97-2: Formal Models of Commonsense Geographic Worlds: Report on the Specialist Meeting of Research Initiative 21. Edited by David M. Mark, SUNY-Buffalo, and Max J. Egenhofer and Kathleen Hornsby, U. Maine, documents the discussions held during the meeting in San Marcos TX, 30 Oct–3 Nov, 1996. It includes a set of researchable questions that forms the basis for future research in this area.
97-3: Report of the ICA Workshop on Map Generalization (Gävle, Sweden, 19-21 June 1997) by W.A. Mackaness, Univ of Edinburgh; R. Weibel, Univ of Zurich; and B.P. Buttenfield, Univ of Colorado-Boulder, reports on the discussions and key findings of a workshop held in Gävle Sweden, 19–21 June 1997. Discussion focused on impediments to automated map generalization, the current state of knowledge and progress on specific problem areas.
Other NCGIA publications added
Annual Report Year 8 (January 1, 1996–December 31, 1996)
Abby Caschetta began employment with NCGIA Santa Barbara as the Publications Coordinator on June 18, 1997, filling the job vacated by Elan Sutton.
Michael Goodchild was appointed Chair of the National Research Council’s Mapping Science Committee.
Faculty members joining the NCGIA Buffalo during 1997 were: Barry Lentnek, Department of Geography; Christopher Rump, Department of Industrial Engineering; and Aidong Zhang, Department of Computer Science
Dawn Becker, NCGIA Secretary, left the department in May, 1997. In July, Linda Doerfler joined the NCGIA staff as part-time Secretary.
In July, Martin Camacho, GIS and Workstation Support Specialist, was promoted to oversee Unix operations for the Faculty of Social Sciences, which includes but is not limited to the Geographic Information and Analysis Laboratory. In September, Jian [Justin] Fan joined the staff of the Geographic Information and Analysis Laboratory as a Programmer/Analyst and GIS specialist in residence.
Carol Bult has been appointed Project Manager with the NCGIA, effective July 1, 1997. She will oversee the management of the recent NSF-awarded project "Application of Spatial Concepts for Genome Data".
Douglas Flewelling has been appointed Project Manager with the NCGIA, effective June 1, 1997. He is responsible for the management of the research projects "Heterogeneous Geographic Databases" and "Similarity Assessments Based on Spatial Relations and Attributes".
Scott Overmyer, Assistant Professor in Business Management, accepted a new position at Drexel University as of September 1, 1997.
Tony Stefanidis has been appointed Project Manager with the NCGIA, effective January 1, 1997. He is responsible for the management of the research project "Spatial-Query-by-Sketch".
Nectaria Tryfona, Post-doctoral Research Associate, accepted a position as Assistant Professor in Computer Science at the University of Aalborg, Denmark, as of June 1, 1997.