ABSTRACT

Hawaii is an interesting place to examine politics and power relationships arising from the use of GIS technologies. As an island setting, isolated from other communities, it is perhaps easier to trace the development of an innovation such as GIS and to identify how systems have been designed, implemented, and used. Because of a highly evolved planning regime including state and local land use laws, environmental impact assessment procedures, coastal zone management regulations, locally elected neighborhood boards, and other planning apparatus, GIS has been widely embraced. Hawaii is a culturally diverse environment, containing a mix of high and low density urban, rural, and suburban development. Examples of public, private, and non-profit sector uses of GIS in Hawaii are summarized. Government in Hawaii has played an important role in initiating GIS through strategic investments in hardware, software, databases, as well as efforts to standardize collection and use of spatial data At the same time, large landowners, developers, utilities, and various non-profit organizations have also developed their own systems using a combination of public and proprietary data. While GIS can be used to empower marginalized groups, GIS use in Hawaii predominantly reflects the existing economic and social structure. Some opportunities for enhancing empowerment are identified including the formation of an "empowerment" clearinghouse, new partnerships, additional research on localized representation issues, leveraging of new resources, and an expanded role for universities.

INTRODUCTION

This paper consists of three parts. First, a survey of GIS uses in Hawaii will be conducted. The survey will serve two purposes: 1) to present an overview of typical applications in Hawaii; and 2) to summarize who has access to the technology and control over uses of hardware, software, and databases. Second, three case studies involving the development and uses of GIS will be conducted. Third, the role of GIS in community empowerment in Hawaii will be assessed, focusing both on the overview of its uses as well as on the specific findings arising out of the three case studies.

One of the difficulties of this topic is that "there is no unifying or underlying theoretical explanation of empowerment (Rocha, 1997). Aspects of ommunicative theory (Habermas, 1984; Healy, 1995) and Gidden's structuration theory (Bryant and Jary, 1991) may be useful in understanding the ways in new technologies such as GIS can shape or transform social interaction, learning and policy discourse. At the same time too, there is also a "instrumentalist" view of GIS -- that it really doesn't fundamentally change or restructure roles or functions within society -- like a hammer that can be used to pound or pull nails, GIS is just a tool. As such, the fundamental differences in values and perspectives with regard to planning and development issues are still primarily a function of the larger political economy of place.

WHO USES GIS IN HAWAII?

For a small state in a remote location, known primarily as a tourist destination, Hawaii has invested heavily in GIS technologies. This investment has been an outgrowth of two related developments: 1) extensive experience in mapping, remote sensing, and cartography; and 2) strong and centralized system of planning. In the 1970's, the University of Hawaii developed a prototype GIS, known as the Geographic Information and Display System (GIDS) and engaged in cooperative research projects with NASA using LANDSAT data to map land use, land and water cover classifications in urban and rural areas (Kim, 1990). Coupled with a high degree of land use regulation and planning, it is not difficult to see why GIS would be such an attractive investment in Hawaii. In 1961, Hawaii became the first state in the nation to adopt a state general plan and to enact a statewide land use law. Unlike other states with many units of local government, Hawaii has only four counties (Honolulu, Maui, Kauai, and Hawaii). Both the State of Hawaii and the largest county (Honolulu) which comprises approximately 80% of the total state population have invested heavily in GIS technologies.

The state's geography, aspects of which include its remote location, varied topography, and coastal environments have increased the need for spatial information. One of the underlying rationales for the state land use law was to preserve agriculture and open space. Hawaii was one of the first states to also adopt a coastal zone management program. It also developed its own environmental impact assessment laws and procedures (Kim, 1991). Various environmental regulations and procedures were also proliferated to protect the natural environment. Ninety percent of the state's plants and animals, approximately 10,000 species -- exist only in Hawaii. The state, moreover, has the only tropical rainforests in the nation. More species face extinction in Hawaii than anywhere else in the nation. With only, 0.2% of the nation's land area, Hawaii contains approximately 75% of the nation's documented extinct birds and plants.

It is not surprising, therefore, to find that much of the development of GIS, particularly at the state and regional level has focused on resource management. Hawaii's geography is such that there are relatively few places suitable for inhabitation. The most attractive of these areas, principally along the southern coastlines, are subject to various natural destructive forces -- hurricanes, tsunami, flooding, and coastal erosion. In 1984, a prototype GIS known as the Ocean and Coastal Information Management System (OCIMS). OCIMS focused on applications such as ocean waste disposal, coastal erosion mitigation strategies, maritime transportation, regulation and development of fisheries and marine resources, and coastal energy facility development. Using the USGS 7.5 minute quadrangle maps, represented at 1:24,000 scale, this system became the template for the eventual development of the state's GIS. Typical applications at the state level include boundary amendments for state land use classifications (urban, rural, agricultural, and conservation). Even prior to the development of the state GIS, the 1:24,000 quad maps had already become a standard for planning purposes. Agencies such as the State Historic Preservation Office, the Department of Agriculture, and the Department of Transportation have all developed interfaces to the state system. Currently, the Department of Land and Natural Resources maintains a GIS registry of "environmental hotspots" -- locations where activities pose serious threats to the natural environment.

Like other places, GIS development in Hawaii has also been influenced by the federal government and strong military presence. In addition to the federal agencies (FEMA, Corps of Engineers, federal CZM, U.S. Park Service, U.S. Bureau of Census and the USGS), that utilize spatial data, there are numerous military bases and command posts that have access to mapping and GIS technologies. GIS has been used in the management and re-use of military lands. Large projects include base closings and the return of the island of Kahoolawe to the state (used as a military bombing target). Smith (1992) has described the strong linkages bertween GIS and war pointing out that not only was the "war against Iraq in 1990-91, the first full-scale GIS war," but also that "text and contexst of GIS is heavily underwritten by a military agenda...a considerable percentage of students who learn GIS in USA classrooms graduate to military and defense related jobs; the Defense Mapping Agency with 9,000 employees is the single largest employer of USA geography graduates."

The City and County of Honolulu, the largest of four counties in terms of population size has developed an extensive GIS. While the state's GIS is focused on resource management, the city's GIS is more oriented towards permit management, administrative decision-making, and the planning and delivery of municipal services. Unlike the state system that is based on the USGS quad maps, Honolulu's system is based on the city's parcel maps, generally at a scale of 1:200. More than 140,000 separate parcels have been digitized. The parcels have been linked to administrative databases (e.g. real property, zoning, infrastructure, and other urban features). Recently, the City has made much of its data available, for purchase on CD-ROM.

While GIS in Hawaii was largely initiated as public sector enterprise, the past decade has seen growth in GIS utilization in the private sector as well. Most of the utility companies (energy, cable t.v., telephone, gas, etc.) have also developed and implemented GIS technologies. Several of the large landowners in the state also use GIS to facilitate in the planning, management, and development of real estate. Non-profit organizations such as the Nature Conservancy, Hawaii Chapter, the Ala Wai Watershed Project, and other agencies like the Kahoolawe Reserve Island Commission use GIS for an ever widening set of applications. GIS has also found its way into transportation planning -- used both in the planning and development of fixed rail, public transportation and highway planning as well as in routing and operations management of bus and handivan (persons with disabilities) services.

Because the technology is still somewhat new and expensive, the primary GIS users in Hawaii tend to be the large, dominant organizations in both the public and private sector. Yet certain developments such as the adoption of the USGS quad maps as well as the use of particular hardware and software (initially mini-computers and Arc/Info) have structured and constrained GIS in Hawaii. Especially in the early years of GIS development in Hawaii, government worked hard to create a standard for GIS -- which involved not just the definition and selection of base maps, but also the acquisition and maintenance of hardware and software.

If one closely examines the recent history of GIS development in Hawaii, it is possible to see three divergent pathways that have split off the initial investments. These represent three different models or manifestations of empowerment through GIS. The first model might be termed, "GIS as good government." Given the dominant forces involved in creating GIS in Hawaii, the dominant players and users, and the basic conception of this model is that good data and analytical tools will contribute to enhanced planning and decision-making. The second model, extends the first set of ideas to include the notion of decentralization of power and decision-making and community-based empowerment. Putting good data and tools into the hands of citizens to improve the quality of life in their communities. The third model, while similar to the second in terms of its emphasis on grassroots applications, entails the use of GIS as a tool of resistance and struggle against the dominant socio-political forces. Three different cases have been developed as a means of exploring these different aspects of GIS and empowerment.

RARE AND ENDANGERED SPECIES

Who speaks for the birds? Or, in Hawaii's case, the snails and other endangered plant and animal life? To understand the prominence that environmental management has taken in Hawaii requires some knowledge of local political and economic development. Long known as a tourist destination, the state has tried to capitalize on its spectacular scenery and unique topography. With increased demand for nature tourism, adventure travel and alternatives to typical mass tourism experiences -- Hawaii has also recognized that protection of natural environments also makes economic sense. This realization has come along side a cultural and political renaissance among Native Hawaiians. The Hawaiian culture is rich in terms of its connect to the natural environment. Preservation of rare and endangered species as well as the protection of natural areas helps to ensure the perpetuity of cultural practices.

The "good government" model of identifying resources that need protection has been facilitated through the proliferation of various environmental regulations. Planners, developers, and landowners need to know what are the important resources are located. Figure 1. "Snail Sightings in West O'ahu" , for example, contains an illustration of the location of rare and endangered snails (http://www.tnc.org/infield/State/Hawaii/). It is assumed that access to such information would enhance decision-making by both government as well as by landowners and developers. Perhaps it would led to stronger enforcement of environmental regulations by government and increased "stewardship" of natural resources by landowners and the corporate sector. The notion of "community" is broadly construed and includes a wide range of stakeholders. Technical information is shared among the different parties and while there might not necessarily be universal agreement across all aspects of data quality, spatial analysis, and decision-making, GIS is used to bring technical experts together. It is important to note that although over time, the various agencies and organizations have appropriated similar hardware and software and in many cases similar spatial databases, there are variations in terms of the coverages and attributes stored in the respective systems. Although the initial databases were developed using public resources, some of the databases which have been augmented by private landowners or non-profit entities like the Nature Conservancy are treated as proprietary sources. The corporate board and sponsors of the Nature Conservancy in Hawaii reads like a who's who of "land and power" in Hawaii.

The Internet has contributed to increased distribution of GIS data in Hawaii. See for example, Honolulu' s GIS, located at: http://cchnl.oceanic.com/Depts/dlu/gis/. While government is under some obligation to distribute and make GIS data available to the public, private organizations have incentives to not do so. There is some times competitive advantage associated with location data or the belief that because private resources were used to build the databases, there is no compelling need to make all of the information publicly available, or to recover costs associated with the data collection and compilation. Pricing may serve as a barrier to data access. In some instances, there may even be larger public policy reasons for not divulging the exact locations of rare and endangered species. A recent environmental impact statement on a proposed highway project in Hawaii excluded this type of information as a means of protecting these species (U.S. D.O.T., 1997).

The issue of marginalization arises in the context of GIS development for the protection of rare and endangered species. Perhaps this is because the environmental concerns involve specialists -- wildlife biologists, botanists, entomologists and others with scientific backgrounds. It is also complicated in Hawaii because of the cultural anthropology and competing epistemologies -- in the naming and uses of plant and animal life by the indigenous people. It is interesting to note that most classification schemes include three names -- the scientific name, the common name (generally in English), and the name used traditionally in Hawaii. While such conventions may help in making the information more accessible to various parties, there still remains a great deal of missing knowledge and information regarding rare and endangered species and how these data are represented on maps produced by GIS. Issues of concern include not just how data are collected, stored, managed, and manipulated, but also more fundamentally, how the data will be used and by whom. While public resources went into the development of these systems, the hardware, the software, the databases and the spatial analyses are largely, if not entirely controlled by a limited number of people. We have come to rely on these "experts" to exercise their judgment and expertise in deciding how the system is to be used.

CITIZENS! TAKE BACK YOUR STREETS!!!

In 1992, the University of Hawaii, Department of Urban and Regional Planning received a $700,000 grant from the U.S. Department of Transportation to build a Crash Outcome Data Evaluation System (CODES) and to conduct various safety related analyses (http://durp.soc.hawaii.edu/codes/codespp1.html). As part of the CODES Project, a linked database of all motor vehicle collisions, EMS transports, hospital data, and insurance claims in Hawaii over the period 1986-1995 was constructed. A special of the CODES Project in Hawaii involved the development of a statewide Traffic Safety GIS. Different procedures were used to probabilistically link databases and to assign coordinate locations to all the crash data (see for example, Kim and Nitz, and Kim and Levine ). To date, the Traffic Safety GIS has been used both for problem identification as well as for the development, implementation, and evaluation of various traffic safety interventions.

As a university research project, funded by the federal and state government, the CODES Traffic Safety GIS serves a broad constituency. In addition to meeting specific requests for information from government agencies, from consulting firms, legislators, and from a myriad of community groups, the Traffic Safety GIS serves as a central resource for academic research. Numerous papers and articles have been produced using the Traffic Safety GIS. In addition, requests come from government agencies, law enforcement, public health groups, other community-based organizations as to the extent and nature of motor vehicle collisions in a particular region.

Recently, the Traffic Safety GIS was used as part of a project with the City and County of Honolulu to identify and implement 'traffic calming' strategies in various neighborhoods on Oahu (http://www.durp.hawaii.edu/calm/charrette.html). Maps and spatial analyses showing the locations of crashes, pedestrian collisions, bicycle collisions, as well as crashes involving key population groups: children, elderly, alcohol-involved persons, and tourists were carried out. These analyses helped citizens, transportation planners, and traffic engineers to better understand when and where potential strategies might be implemented.

Honolulu, like other communities across the country, has begun to explore new ways of thinking about the relationships between traffic, land use, and quality of life. A movement, known as traffic calming -- which involves improving communities by making streets more safe, accessible (particularly for non-motorized forms of transport), aesthetically pleasing, and integrated into surrounding land uses has become increasingly popular in many cities. In Honolulu's case, the City designed a participatory process where in residents come together to inventory and identify traffic problems and participate in design charrettes alongside city officials and transport consultants to improve conditions in their communities. Mapping, spatial analysis, and GIS are important components of this overall process.

The provision of maps, air photographs, and spatial representations of their own communities invariably invokes comments such as, "I didn't know this was here," or "is the peak hour traffic volume on this street that high?" Spatial registration also helps residents to identify and talk about particular problem areas. Where does speeding occur? Where are hazardous locations for pedestrians or bicyclists? In addition to helping identify problems, citizens are also given a "traffic calming" toolkit. These are physical devices or strategies that can be implemented in certain situations to correct various problems. These include roadway treatments (speed bumps, humps, tables, curb bulbs, road narrowing, chicanes, serpentines, road closures, traffic circles, diverters, and others) and other strategies (increased law enforcement, public education, and urban design improvements).

Again, like the case involving rare and endangered species, the GIS was developed by specialists with experience in accident analysis and spatial modeling. But by putting the GIS within a neighborhood setting with a clearly identifiable task (traffic safety), the system has been much more accessible to the broader public. Three different public uses of GIS information have emerged. The first relates to education and understanding of traffic safety problems in particular areas. The second involves using the spatial tools (maps, air photography, spatial analysis, GIS, and the traffic calming toolkit) to design appropriate actions and strategies for enhancing safety, at the community level. Finally, both the City as well as various neighborhood groups have also expressed interest in using the Traffic Safety GIS to evaluate the effectiveness of the various treatments and interventions.

THE FIGHT FOR WA'AHILA

For more than a decade, the Hawaiian Electric Company (HECO) has planned the construction of additional 138 kV high voltage transmission lines in Hawaii. Most of the power plants are located in western part of the island of Oahu. Power is fed to a network of substations scattered throughout the island. The Pukele substation, located in the back of Palolo Valley, is connected by two aging overhead lines. HECO has proposed a third, overhead powerline to increase system reliability. Because HECO's proposal involves construction of 100 foot tall steel towers in the conservation district, an environmental impact statement is required by Hawaii law.

Resistance to the utility's plans has been strong. In addition to citizens groups, and elected and appointed officials from Hawaii, several national organizations have also opposed the proposed action. In 1997, the National Trust for Historic Preservation declared Wa'ahila Ridge to be one of the nation's most endangered sites because of HECO's plans. A community based organization, known as the Safe Power Action Network (SPAN) produced detailed maps and drawings of the likely impacts using GIS technology.

These technologies also provided a basis for critique of the utility company's draft environmental impact statement. Questions regarding the accuracy, scale, reliability, and validity of HECO's information have been raised. Specific concerns focused on the use of selected views and spatial representations of the environmental, visual, health, and property value impacts of the electric company's plans. In addition to evaluating the existing data, new coverages were developed and added. These include more precise and accurate location data gathered by citizen use of GPS (global position system) equipment. For example, when the utility would not provide exact coordinate locations for existing poles, citizens simply went out with GPS equipment and determined the locations themselves. When the electric company failed to provide profile drawings, using the digital elevation maps and GIS software, citizens generated views and perspectives that more completely captured the project impacts.

The reality is that while HECO has millions of dollars of resources to pour into this important project, community organizations have relatively few financial resources. In this instance, access to GIS technologies have helped both in terms of the quality of critique and review of the environmental impact statement as well as in the larger public policy arena, a place where public opinion and attitudes about the project and its impacts still matters. GIS provided a value tool in terms of challenging the traditional power relationships. Some of the electric company officials admitted that the community's drawings and maps were superior to their own. The use of GIS also helped to shape the topics and nature of discourse on the proposal. There were drawing and representations put forth by the community rather than only those that were generated by the electric company.

PRETTY PICTURES AND PARTIAL PERSPECTIVES

In all three examples, there is a basic problem of creating pretty pictures (maps) with only partial perspectives. Some of the most endangered species are located in out of the way or inaccessible places. Without experiencing the place first hand, it is often difficult to understand the information presented on maps and spatial analyses. While topographic maps, as well as rainfall, vegetation, and other coverages may be somewhat helpful to the trained eye, there really is no substitute for direct exposure to the particular environments. But such exposure could come at the cost of degradation of the very environment that one may be interested in protecting.

The difficulties associated with identifying traffic safety "hotspots" have been well established. Merely using counts of fatalities or perhaps fatalities and injuries, ignores basic questions of exposure. Data on pedestrian and vehicle volumes are limited. Moreover, because traffic accidents are the consequences of human behavior, which in turn is influenced by a variety of demographic, social, and economic factors, a purely spatial model may not adequately capture all causes of the problem.

Limitations in terms of access and use of sensitive data is a common barrier to PPGIS. This is not just a problem with environmental data (when trying to protect rare and endangered species), but also with traffic safety data -- particularly when the databases have been linked to injury and health outcome information. In the CODES Project, for example, concerns regarding privacy as well as proprietary data were identified early on as an important consideration. It is important to note that with the CODES project, individuals were matched without the use of exact identifiers such as name, social security number, drivers license, license plate, etc. Instead a probabilistic matching algorithm based on the characteristics of persons, events, and locations were used to match records across different files (Kim and Nitz, 1994). This allowed for a much higher rate of matching than if only exact or direct matching techniques were used -- due to the various inconsistencies, misspellings, and keypunch errors in all of the linked databases. Unlike an "exact" match -- a probabilistic match may or may not be a "true" match. The degree of uncertainty can be estimated from the matching algorithm.

In addition to not having an "exact" match on individual cases, standard reporting practices such as not including enough specific information to identify an individual or isolated event is also routine. The approach is to usually to aggregate cases rather than report individual instances. The aggregations can be done across time, space, or by the salient attributes of individuals or circumstances. Personal injury attorneys and others seeking to, perhaps, recover damages on behalf of clients might also try to access personal or proprietary information. In addition to not reporting individual cases, another practice used in the Traffic Safety GIS in Hawaii is to introduce a certain amount of spatial error. For example, one could post all collisions to the nearest intersection, regardless of whether or not they really occurred at an intersection. Rather than report exact locations, one could also tabulate events by street segment or some other grouping (zones or polygons).

THE 'OTHER'

All GIS users are not the same. Indeed, different groups may use the same information quite differently. In the case of the battle over Wa'ahila Ridge, various sides chose to report (or not report) certain information, or to represent data in different ways. Issues, conflicts, and understandings that arise out of spatial analysis are framed both by the larger political and economic perspectives as well as the specific technologies, skills, and approaches used in the process. GIS is thus used as both a tool of reinforcement as well as a means of advocacy -- for a particular perspective -- be it that of the utility or of an environmental group or community organization.

Culturally, GIS has its origins in scientific, technological, and "systems thinking" traditions -- yet many of the applications and interpretation of findings occur in more politicized contexts or at least within settings having explicit normative outcomes. The findings translate into prescriptions that rare and endangered species should be protected, or traffic safety should be improved, or that the power line should or should not be constructed. With the exception of the University, there are few institutions or organizations that create space for purely scientific endeavors. It is the "applied" character of GIS that creates contradictions between the "scientific" and "normative" aspects of its use.

In the early years of GIS development in Hawaii, the emphasis was on technology involving new hardware and software and the use of consultants, predominantly imported from the mainland. The University also played an important role in providing technical assistance and basic training on cartography and GIS. Yet over time, those using GIS has widened to include a much broader range of users -- planners, advocates, citizens, social service providers, firms and businesses, and others. While there still are distinctions between experts and others who use GIS, over time, these differences have become increasingly blurred. Part of this has to do with the extent to which the technology has been embraced by different groups. The technology moreover has become more affordable and available across different systems and platforms.

In addition to distinctions between "GIS experts" and "others" which arise in the context of GIS -- there are other "others." Increasingly, recognition that local knowledge whether it is seen in terms of indigenous names for plants and animals, or in terms of resident knowledge of traffic conditions privileges one group over others -- particularly if the others are from outside or have limited experience in the community. With traffic calming, moreover, there may be explicit attempts to "exclude" the outsiders from driving within a particular neighborhood. A common expressed concern (and motivation for participating in traffic calming charrettes) is that "outsiders are speeding and creating traffic safety problems within my community." GIS could be used to reinforce attitudes of "NIMBY" (Not-In-My-Backyard) syndrome. Studies in Hawaii have shown that the most likely violators of speeding laws on residential streets are most likely to be residents, not outsiders (Kim, 1997).

The emphasis on territoriality is at once a double-edged sword. On the one hand, it lends itself to community-based identification and empowerment. On the other hand, it creates a divisiveness between places and encourages exclusion rather than inclusion. In Hawaii, this has been reinforced because of nodal pattern of development and because the division and articulation of space into tourist versus resident space. One need only to look at the state's largest tourism area, Waikiki, some 450 acres of concentrated hotel and visitor industries which remains geographically separated from the rest of Honolulu by means of the Ala Wai Canal to visualize the physical separation between these two worlds.

Race, class, gender, and other factors also constitute implicit and explicit divisions in the larger social order. These divisions are no doubt also embedded within the uses or non-uses of GIS technology. This would be a very interesting topic to study further in Hawaii because of the multi-cultural environment. At present, discussion of these influences on using GIS are largely of a speculative nature. There is, of course, research that suggests that males and females or people of different races or ethnicities or social class perceive space, distance, and time in different ways. Does the presence of GIS technology serve to standardize these perspectives -- because of "adaptive structuration" or other changes brought on by the technology itself? Or, are these "other" factors merely integrated into the uses and applications and products that these individuals create using GIS?

POWERING EMPOWERMENT

In Hawaii, community-based traffic safety programs have provided an important way of introducing mapping and GIS technologies into the community. In addition to providing resources for mapping and identifying problem areas, these programs have also enhanced efforts to promote alternatives to driving, to reduce traffic speeds using strategies such as "traffic calming" and to promote more village-scale design of "walkable" communities.

But the question of who has power in Hawaii remains to be addressed. These three cases provide interesting, yet different views of how PPGIS are designed, implemented, and used in Hawaii. While there may be a relatively high degree of agreement over the uses and benefits of a GIS to enhance safety and quality of life in neighborhoods, conflicts intensify as the interests of either regulatory agencies versus developers or utilities and communities widen. In a pluralistic society, with increasingly divergent viewpoints, it may be difficult to reach agreement over how GIS should be used and by whom. Also, when the stakes are great, issues regarding accuracy, quality, and appropriate uses of data become increasingly prominent. Community access to GIS, however, may help to balance power relationships both by providing a new means of critiquing plans and studies, but also in developing reasonable alternatives not yet on the bargaining table.

The question of who has power and access to GIS is related to the larger issue of who has access to computers, internet and the related technologies. While we can expect that the there will be continued growth in development of these technologies as well as in the markets for new products and services -- we can also expect that socially and economically marginalized groups will, no doubt, be last in line for the new upgrades, enhanced versions, and new models. One need only scan the digital landscape in Hawaii to see that it is, very much, a reflection of the larger political and social structure.

Given these conditions, what can be done to "power empowerment?" First, it is necessary to compile good examples of GIS used to empower communities. More than detailed case studies -- some of these examples could form templates or models of "good practice" for other places. Without demonstration of the benefits of GIS, it will be difficult to justify its costs. Establishing a clearinghouse for GIS empowerment tools would help to share information across geographic space. Second, in a small state such as Hawaii, where everyone pretty much knows everyone else, there is an inclination towards partnership. Some of these may be uncomfortable at first -- such as when environmental adversaries begin to collaborate -- but it does become a way to strengthen community and encourage cross-platform dialog and exchange. Such partnerships, ideally, involve a recognition that all sides have meaningful contributions to a shared understanding. Third, much more emphasis needs to be directed towards the issues of representation -- more than just "how maps work" (MacEachren, 1995) but also the cultural and social construction of meaning. In Hawaii, the use of traditional Polynesian navigational methods with GIS serves as an example of cultural knowledge meeting technological perspectives. Finally, without additional resources earmarked to redistribution of access to data and technology, the road to empowerment will be longer and steeper than ever before. New mechanisms and initiatives to leverage resources into the community are needed more than ever before. Initially, because of scarce resources, there may be the need to combine together various local GIS applications -- having the same hardware, software, and personnel used to manage environmental, social, transport, health, cultural and business concerns. Public agencies may need to take the lead in creating these new systems -- perhaps focusing first on a few key neighborhoods or districts, then replicating the data and models across an entire jurisdiction.

CONCLUDING COMMENTS

The University places an important role in facilitating the diffusion of GIS technologies -- through training and education, development of appropriate methods and applications, and serving as a clearinghouse for data. The University also occupies a unique position in terms of its ability to leverage and acquire resources for GIS. Issues such as hardware, software, data use, confidentiality and release of sensitive information, as well as standard concerns regarding the quality and accuracy of information arise within the context of public participation and community empowerment. Efforts to work with marginalized groups in Hawaii, including Native Hawaiians; communities opposed to development or siting of unwanted land uses, and environmental groups demonstrate the nature of conflicts and agreements over the intended and actual uses of GIS technologies and data.

While the University can play a role in "seeding" the development of new technologies and their applications -- there must be some changes within academia to foster genuine empowerment. The traditional model of empowerment through University education focuses much more on the academic and professional success of individual students. No doubt, marginalized groups have benefited from the success of individuals who have taken this route. At the same time, there has also been broader conceptions of the role and function of a university that go beyond research and teaching -- but to include service and extension into the community. Revisiting the role of Land Grant institutions (Key, 1996) in a digital age (including GIS technology) which traditionally have emphasized the importance of extension services may be an appropriate beginning. At the heart of these changes involves a fundamental recognition -- that non-traditional students and those without meeting formal educational requirements may have a place in academia -- both as students as well as teachers.

Finally, there is also need of a "broadened" view of scholarship (Boyer, 1990). Reform of the university setting would not just "tolerate" multiple meanings and epistemologies of knowledge and learning, but also would provide space (physical, social, and cyberspace) that allows for mutual respect and interaction, regardless of age, ethnicity, class, culture, political affiliation, or academic major.

REFERENCES

Boyer, E. 1990. Scholarship Reconsidered. The Carnegie Foundation for the Advancement of Teaching. Princeton, NJ.

Bryant, C. & D. Jary, eds. 1991. Giddens' Theory of Structuration, Routledge. London.

Habermas, J. 1984. The Theory of Communicative Action. Polity Press. Cambridge.

Healy, P. 1995. "The Communicative Turn in Planning and its Implications for Spatial Strategy Formation." Environment and Planning B: Planning and Design. Vol. 23. Pp. 217-234.

Key, S. (1996) "Economics or Education: The Establishment of American Land Grant Universities". Journal of Higher Education. Vol. 67. No. 2. March-April. Pp. 196-220.

Kim, K. 1997 . "A Community Based Approach to Managing Residential Traffic Speed" Transportation Quarterly. Vol. 51. No. 1. pp. 127-133.

Kim, K. 1990. The development and application of Geographic Information Systems in Hawaii. In L. Worrall, ed., Geographic Information Systems: Developments and Applications. Belhaven Press. London. Pp. 87-108.

Kim, K. 1991. "Environmental Impact Statements in Hawaii: Problems and Prospects." Environmental Impact Assessment Review. Vol. ll. No. 2. pp. l03-127.

Kim, K. and L. Nitz. 1994. "Applications of Automated Records Linkage Software in Traffic Records Analysis" Transportation Research Record. 1467. National Research Council. pp. 50-55.

MacEachren, A. 1995. How Maps Work: Representation, Visualization, and Design. Guilford Press. New York.

Rocha, E. 1997. A Ladder of Empowerment. Journal of Planning Education and Research. 17:31-44.

U.S. Department of Transportation. 1997. Saddle Road Mamalahoa Road Draft Environmental Impact Statement. U.S. Department of Transportation. Federal Highway Administration. Central Federal Lands Highway Division. Denver, CO.

Smith, N. 1992. "History and Philosophy of Geography: Real Wars, Theory Wars." Progress in Human Geography. 16.2. pp. 257-271.