Distributed Geolibraries: Spatial Information Resources, Summary of a Workshop (1999)

Implementation of a distributed geolibrary presents a host of challenges, ranging from the technical to the societal and institutional. The latter are discussed in this chapter; technical issues are discussed in Chapter 4.

The policy challenges presented by distributed geolibraries include the following:

• What are the legal, ethical, and political issues involved in creating distributed geolibraries? What problems must be addressed in the area of intellectual property rights? How will these issues affect the technical development of distributed geolibraries?
• Who will pay for the creation and maintenance of distributed geolibraries? What components might be in the public domain versus those provided by the commercial sector?

This chapter addresses many of these issues from the perspective of geoinformation at the local level, how distributed geolibraries might build off the library model (and how traditional libraries have addressed or handled some of these societal and institutional issues), and some of the additional issues introduced by the digital context of distributed geolibraries. These issues are not necessarily unique to distributed geolibraries as many have been discussed extensively within the context of recent digital library programs. The intention here is not to review or paraphrase excellent surveys of the social context of digital libraries, such as that of Borgman et al. (1996), which readers interested in a broader perspective should consult.

Five years ago discussions regarding geospatial data in the United States focused on the rapidly increasing use of such data throughout society and the need to create a more formal infrastructure to coordinate geospatial data coverage across the nation, minimize redundant data collection at all levels, and create new opportunities for use throughout the nation (National Research Council, 1993). Much has been accomplished. Concepts such as metadata standards, standard framework databases, and thematic databases have been developed and pursued (see www.fgdc.gov). The federal government in cooperation with state and local governments has been and continues to be well positioned to lead the development of the basic concepts and public domain databases upon which the NSDI is being built.

The NSDI now involves many stakeholders as a result of activities over the past five years. Its basic data will be assembled from diverse institutions throughout the nation, with institutions contributing those parts that are most relevant to their roles (Tosta and Domaratz, 1997; Moeller, 1998; Rhind, 1999). At the core of this vision is the concept of local generation of geoinformation. Geoinformation is inherently local in nature and of greatest importance to those in that local area. It makes sense that the tens of thousands of units of local governments in the United States understand their own geoinformation assets and needs far better than do higher levels of government.

New developments in technology make it possible for local people to gather local data germane to their own needs more readily, extract data from online and other electronic repositories, develop the information products they need, use the products for decision making, and contribute their locally gathered geoinformation and derived products to libraries or other repositories. Developing the technical and institutional means to support incorporation of local knowledge into networked repositories presents a novel challenge.

Stakeholders across the nation are beginning to think and act around more common visions for the NSDI. A library service model

provides an initial way to consider the organizational and institutional arrangements for finding and accessing the geoinformation assets and digital products being generated by numerous stakeholders across the nation.

In considering possible institutional arrangements for distributed geolibraries, we begin with the assumption that libraries are social institutions that will continue to change but will not be made obsolete by the advent of electronic publishing. Indeed, distributed geolibraries and digital libraries in general will complement the traditional activities of libraries and related institutions. Libraries respond to many complex societal needs. They are used for research, teaching, self-learning, and entertainment. They serve as social and activity centers for many communities, whether these be small towns, neighborhoods, or institutions. The opportunities that libraries provide range from learning about practical matters to exploring science, art, history, or literature for the sheer pleasure of doing so. They are places for children to learn how to read and places for disadvantaged members of communities to seek solutions and solace (Crawford and Gorman, 1995, p. 118). The library system serves as a repository and by doing so preserves most aspects of our culture. Libraries range from small to large, urban to rural, and public to private but cooperate through a common professional culture and set of procedures, sharing information for mutual benefit. In short: "libraries exist to acquire, give access to, and safeguard carriers of knowledge and information in all forms and to provide instruction and assistance in the use of the collections to which their users have access" (Crawford and Gorman, 1995, p. 3).

Libraries have incorporated information technologies in all aspects of library services. Most recently, libraries have embraced network-based programs that support collaboration among

institutions and the sharing of resources. In addition, consortia have been established on state, regional, and library-type bases throughout the United States to share information, negotiate licenses, engage in collection development, and for many other purposes. A useful distributed geolibrary of the future will need to participate in these activities as an entity that will accumulate, make available, and conserve electronic carriers of georeferenced knowledge.

Existing public libraries do not buy most books or subscribe to most magazines or journals, yet they are highly valued by the estimated two-thirds of American adults who use them (Crawford and Gorman, 1995, p. 127). A typical robust public library will lend out 10 items per person per year based on the population served by the library and will answer two questions per person per year for its service population. Typical circulation of a robust library is twice its content (i.e., a library with a collection of 1 million volumes will lend out 2 million volumes during the year). In-library use of volumes in poor and rural communities often exceeds circulation, and in-library use at academic libraries often exceeds circulation by two to three times.

Public libraries provide these high use and service rates at a cost of approximately five cents per day per capita for their service population, while public libraries in economically healthy areas aspire to 10 cents per day per capita as a reasonable starting point for funding a robust library (Crawford and Gorman, 1995, p. 139). These expenditures appear to be a bargain for the access and services provided, and any proposal for supplanting current library services with electronic services would need to compare costs realistically.

Conversely, would an electronic digital library be available to at least the two-thirds of American adults who currently use existing libraries? Would it serve children and the disadvantaged to the same extent or greater than existing library facilities and resources?

There is an economic conundrum that in the face of a proportionately higher demand some communities might not have the available resources to support distributed electronic delivery services, even though the delivery technology is dropping in price. In terms of distributed geolibraries, this may be an issue, as a recent survey of public libraries in Colorado (Gayon, 1998) indicates that rural libraries receive a larger than expected proportion of requests for geographic information (maps, images, and digital data).

Libraries have the effect, although not a priority purpose, of introducing library users to works, authors, and publishers. Libraries thereby serve the economic function of creating markets for intellectual works. Would a geolibrary have the same effect? These are some of the institutional questions that will need to be addressed as the technological capabilities for distributed geolibraries are built over time.

Might distributed geolibraries develop as part of existing library arrangements or complement them? Although the possibility exists that distributed geolibraries might develop in tandem with libraries and be interconnected with them, the duplication of all the roles of libraries in a new institutional environment would make little sense. A useful analysis of these issues is presented by Hawkins (1994) in the context of digital libraries. Indeed, the way distributed geolibraries evolve will depend in large part on access to resources in existing library institutions.

Some of those things that traditional libraries have never been able to do well might be better done by digital means. One of these functions might be the provision of access to geoinformation. The size and shape of the sheets on which paper maps are produced often depend on the information or the story that the cartographer is attempting to convey graphically, the scale required to present information adequately, and the shape of the geographic area being addressed. Owing to the wide variability in map sizes and the nonstandard placement of information on them, the classification, cataloging, and storage of maps have been far

more problematic for librarians than handling books, journals, magazines, and recordings. Thus, in some instances, maps may be ineffective uses of print on paper, and many maps might be better represented, accessed, and used in digital form.

Thus, the advent of distributed geolibraries is likely to alter the relative advantages of paper and electronic map production. Paper map collections in libraries are unlikely to be completely eliminated. Because of the increasing user friendliness of mapping software and the ready availability of digital geoinformation, the ability to produce sophisticated maps and communicate through them is now available to many more people. As a result, we may witness substantial increases in both paper and digital maps that may be of interest to members of communities and made available in their local libraries.

Although a geolibrary is defined earlier in this report as digital in nature, any practical or useful geolibrary from an institutional perspective will need to be able to accommodate a multiplicity of forms for conveying knowledge. The various means and forms for conveying geographic knowledge each have weaknesses and strengths. Diversity in the means for conveying knowledge is a good thing. The institutional geolibrary must maintain a complex multidimensional web of mixed media, knowledge sources, collections, and services (Crawford and Gorman, 1995, p. 78). The expectation is that this will be accomplished by merging and embedding geolibrary technological advancements into the existing library information infrastructure of the nation.

Geolibraries should play a key role in providing access to carriers of geographic knowledge. In addition, some geolibraries also will want to focus on providing access to the ability to process geographic data. In one sense, all a distributed geolibrary need consist of is a good gazetteer in which users can look up information based on location. The "look up" might be accomplished by drawing a box around an area on a computer screen or by indicating a name

of a place or specifying other information contained in the metadata for a particular item of geoinformation. This would allow a user to find out whether geoinformation covering an area of concern exists in the geolibrary network. If databases exist, the system returns metadata on them so the user can further assess the nature and utility of the databases. Performing this role is consistent with the traditional role of libraries. In addition, gazetteers are one of the prime examples of library documents that were never very efficient in paper form. Conversion to electronic form makes sense since both searching of the gazetteer and updating are made much easier.

If it is legal to copy the databases located through the electronic gazetteer (e.g., public domain geographic databases) or to "check them out" from the holdings within the distributed geolibrary (e.g., the conditions of lending might be determined by licensing agreements), the distributed geolibrary as an institution should be capable of supporting these functions. That is, direct access to the library's holdings should be provided. Again, this function is parallel to and compatible with the traditional roles of the library as an institution.

The level of services and functions (see Chapter 4) provided by the traditional library can be different in geolibraries. Should the services and functions of distributed geolibraries extend beyond providing users with efficient access to the geoinformation in the library's holdings? Or do the technologies that could be provided by distributed geolibraries extend the services and functions in an attempt to provide answers to complex questions rather than guide users to resources where answers may (or may not) be found?

To place this concern in context, we should first define some terms, although no attempt is made here to add to the extensive literature on the nature of information (see, for example, Buckland, 1991; Losee, 1997). The terms listed in the following order are sometimes used to describe an ascending continuum: data, information, knowledge, understanding, and wisdom. Crawford and Gorman (1995, p. 5) define these terms as follows:

Whereas the substantive content and focus of geographic infrastructure building have focused on data and information (e.g., the NSDI), the substantive content of traditional libraries has focused on collections of knowledge and to a lesser extent collections of information. Traditional libraries collect and catalog primarily knowledge works for good reason. The reading and contemplation of works of knowledge such as books and journals provide context and convey meaning. Currently, such works are one of the best means by which we are able to acquire understanding. "Works of knowledge" are largely synonymous with "intellectual works" and are thus the primary expressions protected by our intellectual property laws.

The goal of copyright law, and the effect of copyright law in library settings, has been to strike a balance between giving authors sufficient incentive to make their works available on the one hand and supporting the rights of users to use the intellectual works of others for socially constructive purposes on the other. This balance is complex, but the balance in interests supported by our current intellectual property laws has made libraries highly successful and valued social institutions. A similar balance of interests has not yet been achieved in the online world. A background discussion of current intellectual property and copyright issues

possibly related to distributed geolibraries appears at the end of this chapter.

There is a growing collection of geoinformation available online that is in the public domain because no copyright can exist in some databases due to their nature (e.g., those with no creativity or originality in the arrangement of facts). Claims of copyright in some databases have been rescinded by the authors, and the copyright for other works has expired. Additionally, there are the equivalents of online bookstores and online mapstores that sell or license databases to customers. For the commercial products, libraries have explored several licensing arrangements that attempt to bring together commercial interests and public rights interests to arrive at solutions that support the interests of all stakeholders (see, for example, Barker et al., 1995, and Gladney and Lotspiech, 1998).

In the vision of distributed geolibraries, there is a possibility of creating knowledge and making it available through the distributed geolibrary itself; this raises additional concerns about the status of such derivative knowledge from the perspective of rights and intellectual property. Collections of information, by contrast, gain very little protection under copyright law principles.

Suppose a student wishes to know more about Yosemite National Park and has access through a distributed geolibrary to two different types of information: a digital elevation model (DEM), giving the elevations of points spaced 30 m apart across the park; and a landscape description by John Muir. In principle both are descriptions of terrain, but one is a raw database of measurements

in the public domain, and the other is a creative knowledge work. In another example the couple searching for a home in Chapter 1 might access either a database of socioeconomic statistics or a collection of news reports on the changing characters of neighborhoods.

Both types of information are valuable, depending on the circumstances and the skills and requirements of their users. To a distributed geolibrary they both look like collections of bits with footprints, and both are retrievable using the same mechanisms. Traditionally, one might have looked for the raw data in a data archive, such as the EROS Data Center of U.S. Geological Survey, or a Census data center, and for the description in a library. But distributed geolibraries would provide a unified means of access.

In doing so, however, geolibraries raise issues concerning the relative value of the two types of information. To a specialist equipped with sophisticated tools of analysis, the raw data may be more useful than the landscape description and more acceptable as a source of information for scientific understanding. To a student without sophisticated tools, only the description may be of value. Moreover, the work of the scientist may result in the production of new data, to be fed back into the distributed geolibrary for use by others (such as estimates of solar radiation based on topography combined with a suitable numerical model) or the production of knowledge works in the form of journal articles, which might also be added to the distributed geolibrary. In this sense a distributed geolibrary would be much more than a repository of knowledge because it would support the creation of new knowledge by individuals or groups, in addition to the dissemination of existing knowledge. A student might wish to create personal knowledge as a result of investigation and use a geolibrary to share that knowledge with others in the class.

Both forms of information seem indispensable. There are many questions of a geographic nature that cannot be answered by a right answer but require careful reflection based on both data and prior knowledge works. Providing new data query, search, and display capabilities and services may be important in some distributed

geolibraries but providing access to digital works of knowledge is likely to be important in all distributed geolibraries.

In summary, a distinction needs to be drawn between raw data and knowledge works because they appear different from the perspective of the functions and services of a library and with respect to intellectual property rights. Although the NSDI is concerned primarily with the production and dissemination of raw geospatial data, distributed geolibraries could also provide an effective mechanism for the dissemination of knowledge.

The concept of access in an institutional distributed geolibrary environment has two major aspects. One involves technical efficiency and effectiveness in finding desired geoinformation, determining its appropriateness and authenticity, linking to and acquiring it, and electronically processing it if needed. To enable such access, knowledge works and databases must exist somewhere on the network with sufficient metadata and tools available in the system to allow these tasks to be accomplished.

The second major aspect of access involves the legal and economic ability of the distributed geolibrary as an institution to provide the geoinformation resources desired by its users, either directly or through the network. If access to intellectual works is barred by legal or economic constraints, powerful computational capabilities and user-friendly search software will not be of any use to the user. Legal rights to materials may alone be an insufficient condition, but they are a critical and necessary condition for access. Acquiring legal rights to intellectual works

and databases can cause a financial burden on the distributed geolibrary and the community it serves.

Although distributed geolibrary collections might be anywhere, they must be somewhere. Those institutions or people with the greatest vested interest in ensuring that specific geoinformation is available, maintained, and accessible are logical candidates for providing those specific collections and resources for distributed geolibraries. For instance, local libraries typically focus on the needs of the local community, and therefore local geolibraries would likely be the primary collectors and maintainers of local geographic information of relevance to local culture.

Another major assumption in the traditional library model is that acquisition or access to commercially provided geoinformation will be through institutional, not individual, payments (Hawkins, 1994). Equity is a fundamental principle of library access. To uphold this principle, the community rather than the individual typically pays for the library and its services (Crawford and Gorman, 1995, p. 101). Just as the poorest Americans can freely borrow books from public libraries, so too should they have equitable access to geolibrary services if a community library chooses to provide those services.

An unanswered issue that will be continually debated in the distributed geolibrary vision is that of access to geoinformation in the public domain and traditional library services versus access to geoinformation and services that are only available on a commercial basis. Embedded in this issue are additional issues of public and private rights and intellectual property. These issues—most of which are not unique to distributed geolibraries—are being debated in the broader library community and the digital information arena (see, for example, two 1997 National Research Council reports—Bits of Power: Issues in Global Access to Scientific Data and More Than Screen Deep: Toward Every Citizen Interfaces to the Nation's Information Infrastructure).

In pursuing solutions there is a pressing need to develop new legal, economic, and institutional models that support the public goods benefits of traditional libraries while providing sufficient incentives for private individuals, private publishers, and

government publishers to make their geoinformation available through distributed geolibrary settings. The practical benefits and drawbacks of institutional models will need to be thoroughly explored from economic, legal, and organizational perspectives. Prototype models will need to be developed and tested. It is highly likely that the most appropriate incentive models for private-sector firms will vary from the incentive models that might best encourage local, state, and federal agencies to make their databases available through distributed geolibrary environments.

This chapter discussed many institutional and societal issues that will have to be addressed by distributed geolibraries, especially if they attempt to replicate many of the services and functions of the traditional library. The major issues are summarized in this section, together with other issues that appear important but were not discussed at length at the workshop.