1. Introduction
From a January "Computer World" editorial:
It's 1994. Welcome to the world of wireless, multimegabit, rightsized, group-enabled, GUI-based, object oriented, user-empowered, decentralized, multimedia, massively parallel, totally open, mission-critical, almost-impossible-to-cost-justify computing.
Information management technology provides application capabilities today that could help us deal more effectively with critical and complex modern issues, and that can vastly increase efficiencies in the daily operational business functions of our governments and our private sector. We have not yet built the "information infrastructure" that will support the known capabilities of information technology. We have a long and expensive road to travel before the information is captured and converted into data, before all the networks are established, and before the rules that will allow appropriate access to expanded information management capabilities are established. The question of 'How do we get there from here?' raises many practical and ethical issues, some of which cause us to reexamine some fairly established tenets of modern jurisprudence.
The more a government built information management system is used by the entire community (non-profits, press, universities, and those supporting the tax base of the community) the more valuable its information management capabilities become. The taxpayers get their value from the system without expanding its use into a consortium between utilities, local governments, and various kinds of data exchange partners. However, such a consortium can be formed that would provide the most efficient organization for supporting ongoing maintenance and enhancement of databases. Reasonable cost sharing among the primary users of the information management system can provide the most information to the most users, and even be further tapped to finance increased subsidies of individual citizen access to the system or its use by general welfare organizations.
An examination of our shared public and private needs, the role of partnerships, foreign models, and some trends in domestic legal settings driven by new information management tools provide a basis in this paper for suggesting a model information management system access and use policy . We need a new semantic to help draw some distinctions that were not needed before the concept of "information" was expanded by new technologies. Using that new semantic, we can establish a more sophisticated legal setting that can enhance the accountability of government, enhance information access, and improve the institutional setting to allow for more synergy in financing the needed information infrastructure.
2. Highways and Data: Institutional and Financial Challenges
A "database," no matter how copious, is not information. It is information's ore. For raw material to become information, it must be organized for a task, directed toward specific performance, applied to a decision. Raw material cannot do that itself. Nor can information specialists. They can cajole their customers, the data users. They can advise, demonstrate, teach. But they can no more manage data for users than a personnel department can take over the management of the people who work with an executive.[1]
Copyright law applies to works of original authorship. Tort law covers issues of negligence, justifiable reliance, duty of care, and remedies for breach. State and federal versions of the Uniform Commercial Code deal with the sale of goods. These and other bodies of law that impact the management of new technologies were not written to work together. Software is a very confusing creation for the law since it can take different forms that seem to fit under many different formally distinct areas of established law. In the form of a custom service it is subject to negligence and contract principles. It can be mass produced, shrink wrapped, labeled with a product warranty disclaimer, and put on the market as a movable good. It can be developed by a federal agency and become a public good, while private proprietary rights can be established in derivative works based on this public good. It can be copyrighted as a work of original authorship. [2] Finally, a great deal of software ownership protection relies on concepts of trade secrecy as defined by the Restatement of Torts and codified in many state statutes. The case law dealing with software involves multiple options as to characterization of rights and remedies, forum shopping on the part of litigants and movement away from bright line tests by the courts.
Authors have pointed to the serious doctrinal problem with copyright protection for machine readable software because of its utilitarian nature. Traditional copyright law had not previously deemed utilitarian works with a function beyond merely conveying information and displaying an appearance to be protectable. Works that were pictorial, graphic, sculptural, or of artistic craftsmanship were protected only insofar as their form, but not so far as their mechanical or utilitarian aspects were concerned.[3] The Office of Technology Assessment stated in its 1986 report, Intellectual Property Rights in an Age of Electroncis and Information, that there was no clear distinction between "idea", which is not protected by copyright law, and "expression", which is protected, using traditional copyright analyses. OTA suggests that when we force the application of obsolete law to novel circustances we may end up skewing the policy objectives that the statute seeks to promote. The idea/expression dichotomy offers an even greater challange when applied to issues of proprietary rights in databases.
How much more complex, then, is the issue of ownership and characterization of databases, particularly when the most important databases are dynamic, changing with updates constantly in our efforts to capture the dynamics of the systems we wish to model, manage, and understand. Like software, you can sell it as a movable good when the expression is bottled in some form of magnetic or CD media, but you still have it after you sell it. As a "unique compilation of facts" with requisite originality, a copyright can be established and enforced for a database.[4] How much more complex is it to fit databases into our current legal setting than software, especially when they are built by government on behalf of different groups of taxpayers and rate payers? This paper is premised on the existence of an intellectual property right in databases built and maintained by government, opening up the potential for more affordable financing through partnerships. And further opening up the possibility of providing more information in more usable forms to the public, private, and individual users.
The federal grant of copyright does not apply to the federal government but it can be invoked by state and local governments. Very little investigation into the conflicts between open records law and copyright of government compiled databases has been done. Arguably, all copying of records cn fit under the concept of the "fair use" exceptions to copyright, and local governments can continue to stay out of the whole list of unsolved problems that apply to the application and enforcement of copyright and other bodies of law to databases. There is a distinct financial need, however, for governments to work very hard to solve some of these problems with legal theory and practice.
According to the National Research Council's Mapping Science Committee and other estimates, the construction of a national spatial data infrastructure will be at least equal to the infrastructure costs of building the national highway system, and vendors have estimated data conversion for automated mapping to be approximately $4.5 billion annually.[5] The National Information Infrastructure, including the National Biological Survey , "electronic government" focused more on efficient citizen service, and all the commercial sector information access and application needs will no doubt cost even more. There are opportunities for synergy to cut these costs if we establish a legal setting that promotes public-private enterprise to build and maintain our future information management infrastructure. Neither the government financed federal highway model, nor the privatized railroad model provide the necessary synthethis of investments, the necessary guarantees of universal access, or the joint management between the commercial and governmental stockholders in this new infrastructure.
One of the most important jobs for modern jurisprudence is to examine the current legal setting, understand its historical roots, and reformulate certain basic principles that are impediments to progress. Information management technologies are already providing capabilities far in advance of our institutional abilities to put them to use, and this gap is getting wider. We must revisit some fundamental principles of modern jurisprudence and select those that must be preserved and those that must be discarded in light of the many changes that information technology is bringing to our economy, our culture, and our understanding of whole systems. These include the checks and balances placed on public-private enterprise (e.g., OMB Circular A-130, model procurement codes, antitrust concepts, multi-agency consortiums, limitations on governmental authority), the concepts of "information" and accountability principles (e.g., FOIA and state public records laws, archiving and management of records, public access and public information dissemination practices), and intellectual property concepts (e.g., copyright of databases and works of authorship on networks, technology licensing, enforcement of third party use restrictions).
"Our intellectual property system has jumped into prominence in this decade. Discussion about the scope of patents, copyrights, and related property controls engages a much broader audience than previously. The ranks of those with a direct stake in our intellectual property system have increased sharply. There is no mystery about the causes of this heightened concern. As new technologies have become the basis of new economically critical industries and transformed older industries, the technologies have become the focus of debates over proprietary protection. However, because our innovative technology includes novel features, it is not easily protected by established policy. The new sorts of "products" and "processes" created by the biotechnology and software industries, for instance, challenge traditional classifications of patent and copyright law."[6] In the five years since this introduction was written the investment of billions of tax dollars by federal agencies, every state, and a majority of the non-rural counties and municipalities in the creation of computerized base maps and the conversion of records as attribute data for GIS applications has exponentially increased the interest in the intellectual property law legal setting and its impact on the traditional management of information resources by government. According to the response to OMB Bulletin 93-14, federal agencies alone spent $4,058.4 million in 1993 and the 1994 budget includes $4,406.6 million for geographic data expenditures. State and local governments spend a great deal more on even more application rich, larger scale data as this technology gains acceptance. Many local governments are viewing information services as one of the highest potential revenue sources since property taxes, while the federal government with its copyright restrictions and OMB policies is not so free to discuss tax spreading plans.
Much of the focus is currently on who will build, own and control the broad-band network that constitutes the "information highway", and modifications to telecommunications laws (see, Senator Hollings' S1822 and Representative Markey's HR3636). This paper focuses on the more subtle problems with our characterization and management of the digital forms of information that will move over those networks, and roadblocks to better institutional arrangements to deal with the ownership and control of the information itself. The paper is further focused on Geographic Information Systems, because solutions in this arena should not vary greatly from the more broadly defined National Information Infrastructure needs.
3. Shared Needs for Information Technology
The advent of geographic information systems, and overall system integration technology is quickly changing the setting between the local government as the source of information management resources and an increasingly sophisticated group of outside technology users. Most government institutions desire to provide the "widest" possible access to public records, even though such records are stored on magnetic media, and accessed and managed through computer technology. Such access policy must take into account not only the purpose of providing broad access with minimal impediments in cost, and complexity. It must balance this goal with limitations in custodial agencies' operating budgets, the required resource support to the host agencies from the computer resources, personal privacy issues, and the procedural requirements of the public record laws of the state. Local governments typically recognize that government data has commercial value and utility, and that they have a duty to manage the capital investments on the part of the taxpayers that financed them. They also realize that the intended use of a public record is irrelevant to the right to inquire, inspect, or copy public records as long as the information requested lies within the public domain. They will not create records to respond to a request, and a majority of these custodians still make no distinctions between requests for identifiable information and requests for entire databases.
Geographic Information Systems have become a lightning rod for many of the data ownership and control issues. Governments, particularly local governments, are the natural source of funding because of the enormous front-end costs, but the commercial utility of these systems is attracting private sector use in a way that has little precedent in the previous implementation of information technologies. Geographic information system technology allows traditional databases made up of text and numbers to be linked with computer maps (or graphic data), allowing for powerful and useful information management applications. These are particularly useful for infrastructure, real property tax, natural resource and economic development/growth management programs. Most information relates to a place, and this technology has greatly expanded the opportunities for information processing and analysis, map production, and modeling. The initial costs of building the computer maps and linking the corporate or agency databases to the maps are significant, although very cost-effective over time. The map managed as a database becomes the fundamental organizational tool and index for all other information "pools", whether or not they originally had a graphic element or a location based field. The legal and policy issues raised by this new tool are also significant, and legislative actions to establish new information policy for funding, managing, and defining public access and use of this technology are critical to its evolution in every state.
Commercial, for-profit applications of taxpayer-owned GIS include demographic analysis, marketing, market research, trade area analysis, market monitoring, network management, vehicle tracking, routing and scheduling, sales territory management, cannibalization analysis, and many others. Vertical industry applications for this technology are growing rapidly in many areas including financial institutions, insurance, real estate, manufacturing and packaged goods, retail trade, and numerous middleman value-added information industries.[7] Government remains the largest user group and education the second largest user group according to Dataquest Inc.'s GIS Strategies, a quarterly analysis of the international GIS market. They show significant and growing use of this technology and the data generated to support applications by consultants, retail, engineering, transportation, forestry, utilities, publishing, software, health care, direct mail, service bureaus, database developers, oil and gas, map publishing, mining, manufacturing, agriculture, restaurants, advertising, aerospace, archaeology, emergency management, lottery, oceanography, pharmaceuticals, and others.[8] While GIS will disappear as a separate technology and be rolled into the "systems integration" concept along with document imaging, GPS, remote sensing, and traditional MIS technology in the near future; it clearly provides a ubiquitous and powerful forum today for reexamination of related legal concepts. Particularly those that impact public-private partnerships, information ownership and control, and liability.
While the technology is widely accepted, the multi-agency organizational structure best suited to cost-effective financing and management of this information management technology is not supported by the current state and federal legal setting. So called "input models" (like the financing of cadastral mapping for Wisconsin counties with land transfer taxes) are the exception rather than the rule. That particular funding source has been used to its limit in most states, will not approach the estimated funding required for the total need, and its application for funding data conversion displaces other critical funding needs.[9] Without a sufficient input model from tax sources, government agencies and utilities that wish to utilize this technology must find other funds. Cost /Benefit studies, and more recently post-implementation studies and better definitions for benefits than those offered by models based on manufacturing productivity show financial efficiencies over the long term from GIS. But significant front-end costs defeat plans for individual agencies that want to build databases, or at least cause them to go forward with database designs that are limited in applications to their needs, or even less than their real needs. Redundant investments are then made when the next application need is defined or when a different agency begins to build its independent databases.
Traditional laws dealing with information access, copyright, negligence, and limits on partnerships with the private sector were primarily put in place before this technology existed. The current amendments and proposed amendments to state laws around the country provide an opportunity for a significant step forward as we work to establish new information policy tuned: 1) to the exponentially changing capabilities of information management technologies, 2) to the economic realities of our local and state finances, and 3) to the traditional values of government accountability. There are substantial economies of scale in GIS construction. A regional, multi-agency approach that takes into account potential private sector needs and financial contributions could help local governments afford the front-end costs of this important tool without state or federal assistance, or new taxes. Partnerships also hold the promise of financial synergy necessary to support data maintenance, which is the recurring killer of GIS implementation efforts that target ongoing operational support rather than one-time projects.
Local governments, in particular, have begun to recognize that the ability to "create" information products and services is distinct from the access to identifiable public records required for accountability. Public records law does not generally require the GIS custodian to create records to respond to a request, with the narrow exception of the requirement for redaction of exempt material from an identified group of records. Such electronic products and services (that would not exist 'but for' a request requiring an application program to be run on the government-owned and managed databases) is increasingly valuable to commercial users, citizens, non-profit organizations, and other government agencies. Copies of entire databases and update services, special analyses, macro/application/software development, and many other such products and services could be made available to a community of information technology users through cost recovery based user fee programs.
Such access and use of the system could not be made available at all under the custodial agencies' primary missions and general fund budgets. The capabilities to provide such products and services has exponentially increased. In anticipation of a comparable increase in demand, local and state governments are beginning to establish electronic information access policies for their geographic information systems. Most are investigating the appropriate level of subsidized cost recovery (not full cost recovery) necessary to enable the government custodian to meet demands for electronic products and services at the lowest possible cost to users, while maintaining a strict public records access policy. User fee revenues from for-profit commercial users can then be used to finance system use by journalists when they enhance the public's "right to know", to provide individual access and participation in government through public terminals on-line to systems, and to subsidize non-profit use for established general welfare purposes appropriate to the jurisdiction.
4. Partnerships
A popular term right now is partnerships. The National Research Council has just published "Promoting the National Spatial Data Infrastructure Through Partnerships",[10] President Clinton's April 1994 Executive Order that established the NSDI emphasizes partnerships, and every state and regional strategic plan for information management calls for partnerships.[11] The agencies promoting the partnership concept seem to be assuming that the funding will primarily come from the other partners, and therefore have less problems with proceeding without a good basis for establishing any proprietary control over the resulting system. While good work is being done to establish who the logical partners with overlapping needs might be, such ideals tend to fall apart when it comes to issues of:
1) placement of the actual staff and equipment in existing or new agencies,
2) who pays what parts of the front-end costs and ongoing maintenance costs,
3) what structure or rules should be established for "horizontal" management of the shared resource,
4) how to account for like-kind contributions,
5) how to respond to non-partners that do not share front-end costs and risk,
6) how to resolve government public access missions with proprietary concerns of non-government partners.
Many chilling impacts on shared resource financing and management remain in the current legal setting that make most of this talk about partnerships rather idealistic without a real working model. The multi-agency GIS consortiums that have survived these first few years of experimentation are usually extremely unique creations that were formed more from the regional differences than from the common problems they were trying to overcome. They have survived based on the force of will of individuals that established and lead them, and have not provided a model based on the force of law that can be objectively promoted throughout the many communities that must find a way to work together as partners to build and pay for the NII vision.
Before front-end cost sharing will take place in most settings, there needs to be adequate assurance to the partners that they can retain some control over the database once it is built. The idea that those that do not pay their share will have it for free tends to act as a major disincentive to many deep pockets from participating as a partner. Regardless of any agreements to the contrary by the partners, proprietary control over the database may be taken away by third parties if one or more of the partners is an agency subject to open records laws. Consortiums have been formed with this issue in mind in many jurisdictions. Some have established "independent" non-profit agencies to own and manage the resulting database, only to discover that the maintenance of some control by the government partners also required the continued application of open records requirements.[12] Licensing arrangements where the database is owned by a private organization and licensed back to the local government have been used with varying degrees of success, but they establish more future problems with the records treatment and ownership of derivative works that result from the ongoing updates to the original database by the government licensee. They cannot adequately anticipate all the needs for use of the data that have legitimate accountability purposes, and often place an impossible enforcement duty on the government agency licensee.
The Mapping Science Committee (MSC) has recognized that partnerships cost money to organize and facilitate properly, and that partnerships can seem cumbersome when compared to a lean, efficient, organization that carries out its mandate as inexpensively as possible. However, it is always inefficient in the broader perspective if one organization duplicates the products of others, and this keeps driving a recommendation to move forward even without a clear understanding of the intellectual property rights involved. The MSC has delineated three major long-term NSDI cost reduction impacts of partnerships:
* Partnerships are an effective way of achieving consensus. Instead of each agency acting independently, partnerships create a sense of shared responsibility for the product and its use. Partnerships broaden the basis of support for projects and help to ensure that they survive to meet the needs of society. Partnerships between federal, state, regional, and local governments act to dispel the perception that one level of government ignores the others or knows better. In economic terms, partnerships broaden the resource base by sharing costs while enhancing the benefits of spatial data.
* Partnerships can encourage a clear division of responsibilities even when the data needs are shared. Historically, responsibility for spatial data has been divided between different levels of government in the United States on the basis of map scale and types of data. For example, the federal government has concentrated on producing maps at 1:24,000 scale and smaller while leaving larger-scale mapping to local and state governments and the private sector. States have mapped themes that match their areas of responsibility, such as transportation. These familiar divisions are becoming confused because of the radical changes resulting form the introduction of digital technology. For example, the federal government is supporting a digital orthophoto quarter-quad (DOQ) program with spatial resolution and positional accuracy higher than that of the 1:24,000 scale mapping programs. GIS is being used by all levels of government to take advantage of these new data types and to integrate data from a wide variety of sources. In a world of high-speed communications and distributed databases, we may need entirely new concepts of ownership of data, or responsibility for its creation. This may take the form of a division of responsibility along entirely new lines, with the federal government responsible for data standards and quality control, and state and local governments responsible for data collection and maintenance.
* Division of responsibilities within partnerships can promote investment so that we develop entirely new ways of reducing costs. Salaries account for by far the largest share of the costs of spatial data, whether they are paid to digitizer operators, programmer analysts, or field workers. The most effective ways of reducing those costs lie in better technology and better training. Spatial data handling and GIS have grown to the point where creative strategies are needed to promote new methods and better technologies and better development of human resources; however, no agency or level of government has assumed a leadership role in such developments. Partnerships could foster a sense of shared responsibility between all levels of government, the educational sector, and private industry. We need partnerships that foster more efficient data collection activities while at the same time fostering a more productive and responsive human resource sector.[13]
The MSC defines a partnership as a joint activity of federal and state agencies (and potentially local agencies and the private sector), involving one or more agencies as joint principals focusing on geographic information. The key elements of such partnerships are:
* Shared responsibilities-the parties to a partnership will have made a formal agreement that defines each party's responsibilities in the activity, in the form of a memorandum of understanding, contract, or other binding document.
* Shared costs- the costs of the activity will be shared between the parties according to some agreed formula.
* Shared benefits- each party to the activity will derive some benefit that is consistent with its mandated role as an agency. In addition, benefits from the partnership will likely accrue to parties outside the partnership and to society at large. This is especially important when the partnership is seen as contributing to the evolution of the NSDI.
* Shared control- decision-making control of the project will be divided between the participants.
5. Domestic Problems and Foreign Models
The universal good sense expressed by these partnership recommendations begins to fall apart when the reality of paying new money for data integration and maintenance comes up and none of the potential partners have the budget to do it, or the authority to seek cost recovery from "clients" (i.e., users who were not part of the original partnership). Data standards and data maintenance are sacrificed as budgets begin to get strained. Cost sharing formulas agreed to by one administrator are not acceptable to successor administrators after recurring costs are not recovered in the form of tangible benefits. Private partners including utilities have problems with distribution of copies of the database for the cost of a tape and don't join in the first place, or worse, drop out after government partners have come to rely on their cost share. Original partners that agreed to cost share turn to open records requests rather than pay their share when it becomes due. The lead partner (established by having taken the primary financial risk, often through the issuance of a bond) is then left with a greater share of the incremental payments than they budgeted. Idealistic policies to fund electronic government access or expanded citizen services as an aspect of the information management plan are the first items to go unfunded.
Even if one or more of these events does not occur, the fact that they are all possible has a chilling impact on the creative efforts to put together the partnership, especially one with utility and private sector participants. At the core of all these impediments is the lack of governmental proprietary control over the "database". And the solution has to do with granting some proprietary control over the information management system, distinguishing the database as a dynamic part of the system not as a "thing", and further distinguishing the database from "records". Outside the United States working models have dealt with such issues in advance of the formation of the partnerships.
For example, the Guide for Government Managers established in 1991 by the Federal Treasury Board of Canada states in part, "...ongoing provision of information is expensive and should be undertaken only when: 1) there is a clear duty to inform and 2) users are willing to pay for the information. The full costs of providing information to satisfy proprietary interests of individuals should not be borne (i.e., subsidized) by taxpayers at large; where appropriate, external user fees and charges should be applied for provision of information and related services." Winnipeg has built a state-of-the-art GIS, amended its ordinance dealing with access to records for accountability so as not to apply to defined electronic products and services (including copies of the database), and marched forward to form partnerships with utilities, the legal and real estate community, private sector consultants, and others.[14]
While establishing a GIS to support the operations of Ontario Hydro, executives noted that a developing controversy in the USA with parallels in Canada is the clash between freedom of information, encouragement of GIS database development and services, and the traditional role of government agencies as providers of baseline data at nominal prices. "Too many GIS users are scientists, and they tend to regard information as something to be freely disseminated rather than owned, subject only to due acknowledgment. At present, charging for the incremental costs of serving a client seems acceptable to most users, but recovery of the cost of discharging one's intrinsic function is not, except where some extra service is performed such as a rush priority. Even within our own corporation there are three conflicting policies awaiting resolution: freedom of information, a requirement or option to charge for internal services, and a prohibition on charging corporate accounts twice for the same good or service. Regardless of what pricing formula may be argued, it is essential that GIS user establish the principle that geographic information is a company asset that must be accounted for. Also, geographic information should be treated as part of the official corporate records. Typically, there would then be a suite of generic company policies and procedures to provide some guidance, even if only as a base from which to create more appropriate ones for GIS. Otherwise, it is all too easy for valuable files to be misplaced, sold or given away inappropriately, or damaged."[15]
Unlike Canada and many other countries (most notably the United Kingdom and Australia) this kind of policy has not formed the basis for cost effective management of taxpayer-owned information resources here in the United States where most of the technology at issue originated. The United States has legal, historical, and policy problems with most concepts of public-private enterprise, especially when we move away from the local government setting towards federal government setting. We have problems with the definition of intellectual property rights in information assets, especially when it comes to government or taxpayer "owned" assets. We lack a philosophical attitude that prevails in most of the rest of the technologically advanced world. Namely, that business and government are not all that different, and collusion is not all that wrong so long as there is a positive cost/benefit and a continued guarantee of fundamental rights. We do allow for the creation of legal entities called authorities (i.e., port authorities, sewer authorities) and utilities in the United States that potentially can have a mix of legislated governmental powers, some tax and user fee assessment authority, as well as private sector funding and investment tools. These can be constructed to look a lot like the foreign government/private competitors that range from Crown Corporations in Canada to the European government run and competitively subsidized airlines, but such authorities are almost never utilized to support a program of national or international scope in the United States.
There is "legal" room for more creative and more cost competitive organizational structures if the policy side can be made to work with the tools we do have. We look back on the railroads with their robber barons, government supported monopolies, and attendant abuses and are happy that we have built so many safeguards against "undue competition" and monopolies ranging from the antitrust statutes [16]to OMB Circular A-130. But today, we look at the mounting debt in terms of failure to maintain the infrastructure already built in this country-pipes, bridges, and roads. We look at the many versions of taxpayer revolt in the form of state level "Proposition 13's". We hear demands for smaller government from the Libertarian Party and the National Performance Review. The idea of a taxpayer financed information infrastructure that takes full advantage of known technology capabilities is not realistic when we are not adequately maintaining the current infrastructure, or even the interest on the debt. Local tax initiatives to finance GIS are even less politically palatable. But user fees that spread the cost for infrastructure and services to those that use them have been growing faster than any other local government revenue source. Included within the proprietary rights to charge user fees for information management system use is the authority to establish enforceable multi-agency agreements to build it in the first place. This works when the user fees are based on services like garbage collection or municipal swimming pool use or government owned parking lot access. But confusion results from starting down the same road when the commodity is information, a well-known public good, and the service is the creation of information from government records.
"It is perhaps a contradiction in this culture that Americans are often indifferent to the plight of the poor but find their heartstrings pulled at the thought of the poor being denied a chance of survival or cure that the rich can afford. Is it ethical to withhold a cure from the poor or the infirm? Put that way, it is hard to say yes. But the question can be formulated differently, as indeed it must when considering the question of whether to allow patents to issue for medical advances. This is because in the absence of patent rights, the level of investment needed to bring about the advances that will provide cures may fall off, and there may consequently be fewer advances and, overall, fewer people cured."[17] A direct analogy to the appropriate limits for government's management of information resources is at stake. If we limit proprietary management of databases under the theory of open records (or a failure to adequately mature the copyright law to keep up with the most important commodities we are producing) we limit the practical use of multi-agency organizational structures that would help to adequately finance the provision of more information resources. As a result less information resources exist, and the infrastructure to access them is never built.
6. Access To Records vs. Systems Use
Many problems have been identified that arise from the application of older Freedom of Information policy to newer information management technologies. Can we preserve accountability while distinguishing commercial use of these systems? Good examples are illustrated by the current debates over legislative recommendations in New York, Florida, Washington, Vermont and South Carolina. Other states including Tennessee and North Carolina are avoiding the broader issues by legislating exceptions for GIS commercial use from traditional records treatment on a county basis. Still other states ignore the FOIA issues of local government cost recovery programs, proceed to establish cost recovery programs under local ordinances, and await litigation to define the need for broad-based legislation (e.g., Kansas, Ohio, and California.) In the face of local and state government efforts to develop information technology cost recovery schemes we find an unusual coalition for resistance forming between large corporate value-added information vendors and public interest groups, both of which claim this to be an eroding of Jeffersonian principles that will not serve their interests so well as application of the status quo treatment of paper records to entire information management systems.
Reports from diverse groups including the Information Industry Association, the Reporters Committee for Freedom of the Press, and Ralph Nader's Taxpayer Assets Program call for clarification of legislative language to expand access to "electronic information," and promote placement of extreme limits on the proprietary control and fees that might be applied by custodians who manage computers and databases. The "everything is free" philosophy fails to see the encouragement of further creation of useful, accessible data and information systems as a public and economic benefit in itself. There is a gap between the ideal of granting access to everything (non-private) to everyone (without distinction as to intended use), and the operational realities which include the potential for exceeding budgets or offsetting primary missions in an unfunded service bureau role. There is a need to distinguish identifiable records from potential electronic products and services. This includes the distinction that "entire databases" are not identifiable information. "Entire databases" are merely a part of the information management system used to 1) produce records that did play a part in the business of government, or 2) to produce information products that did not previously exist. The primary distinction between these two applications of the database is the first involves "access" and the second "creation". It is the difference between granting access to information and providing information dissemination services that must be emphasized, not the delineation of the entire database as a record.
Political response (whether from Congress, state legislators, or policy formulated by a particular government custodian) to a request for authorization for cost recovery too often results in a "politically correct Catch 22." Provide the new capabilities of information management technology and increase service to the citizens. No new taxes or budget increases for conversion or maintenance of information. And, retain the universal definition of "records" in open records law so that entire databases are included. This is summarizes the self defeating impacts of many of the efforts to implement innovative technologies. One good example is WINDOW (S. 564, PL 103-40) which constitutes an unfunded mandate to the GPO to make digital federal clearing house available to federal depository libraries. Other federal efforts add more good sounding mandates without the means to implement them, including the Improvement of Information Access Act (HR 629) , which directly addresses dissemination (not mere access) of federal agency data at the cost of distribution.
Traditional records treatment applied to determine access procedures for government managed information management systems not only ignores the opportunity for revenue generation that offsets tax burdens for the community as a whole, it establishes a potential free service bureau role that can displace agency missions. A semantic problem creates the most confusion and defeats solutions to the Catch 22. There is a distinct enough difference between " access to identifiable government records for purposes of guaranteeing accountability", and "use of taxpayer-owned information management systems for individual and commercial purposes" to provide a bright line test. We should not be worrying with the concept of a database as the watershed issue at all. However, there is no bright line drawn at the database vs. records end of the spectrum without some modification to the definition of "records" in the open records laws. And as long as we are stuck with historical treatment of "databases" rather than information as something that can be inspected and copied for accountability, the fundamental problem will not go away.
Several time-honored open records principles work to defeat responsible management of this taxpayer resource. The watershed question is raised by the open records request, "First, provide me with a listing of all your databases and explanations of their contents, the data dictionary, etc. Next, give me the following databases in my format for transaction costs...." Always implied with such requests are numerous hours of time over the next few months while the requestor tries to get your database(s) up and running. Can the custodian ask them why they want it so as to:
a) limit liabily and other risk management for misapplications,
b) avoid doing free work for a for-profit firm that has no issues of accountability in mind, c) protect the available copyright held and managed on behalf of the local taxpayers,
d) determine appropriateness of subsidizing the production of information products for an economic development or general welfare purpose expressed in local policy ,
e) limit redistribution to third parties in furtherance of privacy, copyright and liability limitation considerations,
f) put off the service oriented response until the agency's internal use needs are completed and time to perform such extra service activities is available ?
Not if databases are records. Then inquiring into intended use, delays beyond specified statutory time frames, as well as limits on redistribution are not appropriate. Custodians should not be able to inquire why a requestor wants some identifiable information. Custodians should for many reasons ask why a requestor wants an entire database.
The application of relational database technology to "records" is central to the semantic confusion and practical problems that surround electronic access issues. We are no longer providing access merely to electronic filing cabinets full of the original identifiable information. Electronic manipulation of the data contained in publicly financed databases can create new information that did not exist "but for" the request. Since government information custodians are not bound to "create records" to meet a request, much of the potential use of relational database technology is lost under this "old" view of records. The potential to manipulate records to create information products is typically available only to commercial information vendors that will produce only at a cost + profit rate. The public policy position of getting the most information to the most people is limited by the impact of rules designed before the potential of the information age was recognized. If state legislation distinguishing use of government owned information management systems for commercial purposes from access to records for accountability is passed, the government agencies with large investments in databases will proceed to document the advantages of multi-purpose, multi-participant database creation and maintenance. The agencies will further be able to show increased overall access to information that can result under new rules designed with the capabilities of current technology in mind.
7. Legislative Agenda
The following list summarizes some of the factors that have hindered the state and local governments' efforts to establish GIS to date, and that work to defeat multi-agency or public-private enterprise database building projects around the country:
* Taxpayer subsidies to commercial "value added resellers" of information products result from mandates that allow entire databases to be turned over for the price of copying, while public digital custodians are prohibited from applying staff expertise and software to those same databases to "create" an information product for the average taxpayer.
* The lack of express proprietary authority over databases on the part of the public agencies which created them can result in an unfunded service bureau role.
* Limits on the formulation of public-private partnerships in information systems construction and operation are inherent because of the private partner's risk lose of control and ownership of all information jointly funded and placed in the database. This, in turn, inhibits the sharing of costs for creation of expensive digital land bases, limiting the cooperative and cost-effective establishment of multi-agency consortiums.
* Confusion between "programming" and "operating" a computer can be used to defeat legitimate demands for public records searches.
* Confusion as to the application of governmental immunity and tort claims acts to the provision of electronic products and services.
Legislative modification is still needed for the copyright law and in most states' open records laws to set the stage for creative use of expanding, taxpayer-financed information assets. Suggested changes follow:
* "Information" must be viewed to include a broader universe than records. (Suggest change definition in FOIA and state open records laws to distinguish specified electronic products and services as seen in the Alaska model)
* Records need to be tied tightly to the subset of information in some way associated with government accountability. (Establish a "central purpose" test for requests that is balanced between protection of privacy, guarantee of accountability, and that still preserves the opportunity for discretionary cost recovery based services).[18]
* Information in certain forms should be distinguished and managed as a commodity. (Establish more direction on what constitutes the requisite Feist test for "originality" required to establish a copyrightable database, and less lumping of all databases under one rule, when databases vary as much from each other as they do from books. Again, a central purpose test is needed based on the promotion of dissemination of ideas by guaranteeing authors proprietary rights.)
* Appropriate legislative direction must be given to custodians for establishing fees for electronic products and services that enhance funding and public enterprise without unduly competing with the private sector. (Express authority to provide discretionary services for user fees, and cost based fee setting formulas that provide for equal treatment and subsidies or exceptions for general welfare purposes.)
* A framework for encouraging the creation of broadly useful information management systems should be established that promotes "outside" access to the system capabilities and creation of new information. (Put the custodians in the position of trying to expand use of their information systems by letting them retain user fees for system maintenance that they generate from marketing and selling the taxpayer asset to cost sharing partners and system clients.)
* A framework that promotes interagency cooperation and coordination through shared financing and shared management of information management systems. (Authorize and mandate horizontal management structures that reach across traditionally vertically funded and managed departments, divisions, bureaus, sections, etc. for funding, management, access and use of systems).
Legislative formulas for dealing with these issues are available, and a review of those efforts in place or under consideration in other states can illustrate the solutions that may gain acceptance as a model. Recommendations for changes to copyright law are being formulated by an Intellectual Property Rights Working Group of the federal Information Infrastructure Task Force implementing the NII.[19] Twenty examples of state legislation and proposed bills currently under consideration have been identified that to various degrees have attempted to address one or more of the identified problems with joint financing or cost recovery for GIS. Each example of existing or proposed legislation deals with access to geographic information systems (GIS) or information technology generally and the authority to establish fees in excess of the partial transaction costs allowed for records. Most are direct modifications to open records laws, and some are codified as Data Management legislation separate from the public records provisions. They include:
* Alaska 1990 SLA CH.200 (codified at [[section]]09.25.115 et seq.)
* Arizona CH.1, [[section]]39-121.03
* California Gov. Code [[section]]408.3, [[section]]409 Assessment, [[section]]6256 computer records
* Colorado HB 92-1995 (codified [[section]]24-72-203)
* Connecticut [[section]]1-19(a), and Pub.L. 91-249
* Florida [[section]]119.07(b), [[section]]119.085 and Draft amendment to FSA 119.01 and 282.303
* Iowa [[section]]22.2.3 Executive Branch
* Kentucky KRS 61.872 to 61.884 (as amended 1994)
* Maine Title 12 [[section]]1753-A 5
* Maryland HB 1538 (passed 1992 session, effective October 1, 1992)
* Minnesota [[section]]13.03 Data Practices Act, and [[section]]375.85
* North Carolina 1991 Acts CH 285 (1991 HB 356)
* Oklahoma Title 55 [[section]]24A.5.3
* Oregon ORS 268.357
* South Carolina CH 4 Title 30 (proposed amendment)
* Tennessee TCA [[section]]10-7-506, (expanded 1992 HB 2125)
* Utah [[section]]63-2-201(7)
* Vermont SB 26 proposed 3VSA [[section]]2202 et seq.
* Wisconsin [[section]]19.35(g)
* Washington proposed SHB 1752.
Basically, statutory guidance is needed to authorize custodians of government information management systems to make policy regarding access to and use of the systems when accountability is not the issue, and to enact that policy using differentiated, partially subsidized user fee structures to the extent general funds are not available.
They should put the custodians in the public policy position of wanting to get the capabilities of their information management systems out to the broadest base of users. A large base of individual users should drive down the average costs of operating the system. This is done by giving the custodian the authority to form consortiums of participants, and to market to non-participants to generate system clients; and by giving them an incentive by requiring them to recover a certain portion of their operating budget. One time users seeking a custom maps or analysis, subscriptions to the entire database for on-going businesses that have programmatic applications, special combinations of layers and geographic coverages that are suited to professional use for financed projects, etc. should be made available. Custodians should provide such services and products based on user fees calculated to recover a portion of the system's annual equivalent costs minus the value of use by the custodian for taxpayer purposes. Custodians should give user fee breaks where the taxpayers financing the system believe they are appropriate and establish financed policies through their elected representatives. These might be typically provided for academic, journalistic, and non-profit use; but might also be provided for economic development incentives, promotion of certain land use practices, etc.
We need to empower and finance the custodians of taxpayer's intellectual property assets in information to achieve broad and equitable dissemination of information. Some critics will always claim that the issue is black and white: If general funds are not allocated, then the jurisdiction does not deserve this technology. All talk of user fees for information management systems' access involves a threat to participatory democracy. While any new and flexible authority granted to government agencies will involve opportunities for abuse, this black and white position will only hurt the taxpayers as a whole in favor of the few who actually do use the information net both for participation in government and for commercial profit. The long term goal of free access to the most information for the most people is not at issue. The issue is how do we get there from here. And the answer is to establish reasonable funding and control for information management system implementation within our governments' custodial institutions, at least during this time of paying for the initial spike of conversion costs and education and change necessary to establish a real National Spatial Data Infrastructure.