Licensing1 has both advantages and disadvantages, and is one of a number of procurement options available to government. Agencies considering data acquisition and distribution alternatives must weigh all options in the context of their mandates, missions, and user needs; government efficiency and accountability; the public interest; the effect of their decisions on private markets; and budgetary realities. The culture of licensing is evolving as the geographic data community experiments with this tool, and licenses are becoming more flexible. In this environment, society will be best served by agencies (1) sharing contract negotiation experiences and techniques; (2) frequently refreshing their understanding of data acquisition and dissemination options and user needs; (3) encouraging unambiguous, standardized, and automated licensing; (4) using licensing to improve coordination of data acquisitions; (5) enhancing government institutions that coordinate acquisitions; and (6) investigating options for building a National Commons and Marketplace in Geographic
License or licensing of geographic data or a geographic work means a transaction or arrangement (usually a contract, in which there is an exchange of value) in which the acquiring party (i.e., the licensee) obtains information with restrictions on the licensee’s rights to use or transfer the information. Geographic works are works incorporating geographic data that have been collected, aggregated, manipulated, or transformed in some manner. Geographic services refers to the processes of obtaining, processing, or providing geographic data or geographic works. For a glossary of terms, see Appendix E.
Information. 2 Progress on all items could be seen in the near term, although some aspects of items 5 and 6 would require more time and resources.
The shift from sale of physical books, maps, and other intellectual works to the licensing of digital data, information, and affiliated services represents a significant change in the communication of knowledge. This shift is altering the balance between public and private interests in geographic works and data, a balance that has heretofore provided a level of creativity and innovation unequaled elsewhere in the world.
The balance that prevailed until recently was that the federal agencies generally paid for full rights in geographic data, including the right to publish it freely, permitting the data to enter the public domain and become readily accessible for follow-on uses. The prevalence of digital media as the platform for large databases, however, has made it feasible3 for data providers to contemplate multiple licenses of geographic databases, even when the data are made available online to large numbers of potential users. Thus, vendors in some instances would prefer to license data to government agencies, with restrictions on redistribution and reuse of the data. In this environment, such data may not enter the public domain, and the cost of access may preclude otherwise beneficial uses, including the development of new products and services and informing public discourse. The fundamental issue addressed by this report, therefore, is under what circumstances and to what extent should agencies accept limitations on the further distribution or use of geographic data they acquire from private and other governmental vendors.
The number of uses of geographic data has expanded rapidly with the evolution of geographic information systems that manage geographic
data,4 improvements in remote-sensing technologies, the advent of inexpensive Global Positioning System receivers, the decreasing costs of personal computing and digital storage, the increasing reach of the Internet, and the increasing pervasiveness of wireless, location-aware telecommunications services. These developments have been accompanied by increased use of licensing as an alternative to the outright sale of the data and data products. Licensing has become commonplace because of
the realization that many geographic data, as opposed to geographic creative works, are difficult to protect through copyright alone;
a shift away from supplying distinct datasets to providing access to databases;
the rise of business models that stress multiple subscribers despite the reality of digital networks and media that allow others to distribute perfect and inexpensive copies;
increased concern over potential liability and a desire to limit liability through explicit license language; and
the rise of shared cost and data maintenance partnerships.
Expanded mapping activities have increased the potential for duplication of data gathering and processing. Initiatives such as the U.S. Geological Survey’s (USGS’s) The National Map, the Office of Management and Budget’s (OMB’s) Geospatial One-Stop, and the U.S. Census Bureau’s MAF/TIGER5 modernization program seek to leverage local investments in geographic data and avoid unnecessary duplication. Because states, tribes, regional groups, counties, and cities have a wide range of data-sharing policies, the federal government is increasingly forced to address licensing issues for the data it acquires. Confusion and uncertainty have arisen as a result of
a proliferation of nonstandard licensing arrangements;6
difficulty in designing licenses that track the legal, economic, and public interest concerns of different levels of government;
difficulty in designing licenses that accommodate all sectors of the geographic data community;
an imperfect appreciation for the licensing perspectives of different sectors of the geographic data community; and
lack of effective license tracking and enforcement mechanisms.
THE COMMITTEE’S TASK
Given the confusion and uncertainty surrounding licensing, the National Academies, at the request of the Federal Emergency Management Agency, the Government Printing Office, the National Oceanic and Atmospheric Administration, the U.S. Census Bureau, and USGS, convened the Committee on Licensing of Geographic Data and Services in 2002. The committee was charged with six tasks:
to explore the experiences of federal, state, and local government agencies in licensing geographic data and services from and to the private sector using case studies such as the Landsat program;
to examine ways in which licensing of geographic data and services between government and the private sector serve agency missions and the interests of other stakeholders in government datasets;
to identify arguments in favor of and in opposition to spatial-data licensing arrangements;
to dissect newly proposed license-based models that could meet, concurrently, the spatial-data needs of government, the commercial sector, scientists, educators, and citizens;
to consider potential effects on spatial-data uses and spatial-technology developments of competing license/nonlicense approaches within the commercial sector, and
to analyze options that will balance the interests of all parties affected by licensing of spatial data7 and services to and from government. Each of these tasks is now addressed in turn.
GOVERNMENT EXPERIENCES IN LICENSING GEOGRAPHIC DATA AND SERVICES FROM AND TO THE PRIVATE SECTOR
Despite recent interest in licensing, most federal agencies still prefer full ownership rights in the data that they acquire when this option is available. Their reasons vary from increased flexibility in the use of such
data to supporting agency and federal mandates relating to access, dissemination, duplication avoidance, waste avoidance, and saving money. None-theless, all federal agencies that provided input to the committee have acquired data under license from commercial vendors. Their reasons for doing so vary from being able to make maps faster and less expensively to having no alternative source of data meeting specific needs. Reactions to licensing differ from agency to agency, although there appears to be a general consensus that any cost advantage offered by acquiring licensed data must be weighed against constraints on current and possible future use and the interest in free exchange of information. In some cases, the coordination, negotiation, and administration costs associated with licensing are higher than those of other procurement methods.
Federal agencies almost always distribute geographic data at or below marginal cost of distribution. Since the 1990s, however, many state and local governments have experimented with distribution of data using licenses to generate revenue from their data.8 Ten years later, many of these entities have concluded that fee programs (1) cannot recover a significant fraction of government data budgets, (2) seldom cover operating expenses, and (3) act as a drag on private-sector investments that would otherwise add to the tax base and grow the economy. The use of licenses to provide data to users may, however, be useful to enforce proper attribution, minimize liability, enhance data security, and formalize collaboration.
WAYS IN WHICH LICENSING SERVES AGENCY MISSIONS AND THE INTERESTS OF STAKEHOLDERS IN GOVERNMENT DATA
Agency mandates and missions can be broadly grouped into those requiring broad, limited, or internal data redistribution; those requiring distribution of derivative products; and those ensuring adequate citizen access and judicial review. In addition to utilizing outright purchases of data, agencies have experimented with a range of licenses to acquire data to perform their missions.9 So far, results have been mixed. For the most part, agencies whose missions require broad dissemination find acquisition
Examples included Hennepin County, Minnesota; the State of Maryland; and various European weather services. See also examples cited in Open Data Consortium, 2003, 10 Ways to Support Your GIS Without Selling Data, available at <http://www.opendataconsortium.org>.
of licensed data less useful than agencies that have small numbers of users or need licensed data as an input for making derivative products. Over time, some agencies have learned to negotiate new types of licenses that potentially offer better value to both the agency and commercial data suppliers.
Commercial data vendors have mixed attitudes toward licensing. In general, providers whose business models depend on adding value to data gathered from local, state, and federal agencies tend to oppose government data acquisition through licensing. Providers whose primary business models involve selling imagery or low-value-added geographic products to government generally welcome the prospect of licensing data to the government.
Academic users and producers are among the strongest advocates for the free flow of government geographic data as well as the free flow of any other publicly funded data and information of use to the scientific community. Nonetheless, the interests of students, teachers, researchers, libraries, and university administrators in gaining access to geographic data are not necessarily the same. For example, students and teachers may need legal and convenient access to data to accomplish class demonstrations, laboratory exercises, and class projects, but may care little about the right to openly publish datasets or derivative products. Researchers, on the other hand, need the legal and practical ability to access, use, and extend the datasets and work products of others, including the right to publish derivative works.
Ultimately, however, although agencies often are charged with promoting the public interest, the interests of actual and potential user groups may be discounted by agencies faced with budgetary constraints and vendors’ demands.
ARGUMENTS IN FAVOR OF AND OPPOSITION TO LICENSING ARRANGEMENTS
The advantages of licensing data as opposed to outright purchase (e.g., through a for-hire service) include reducing acquisition costs (i.e., cost to the customer) in many instances, making data immediately available, enabling faster build times for operational information systems, structuring data release after a given embargo period, supporting specific agency projects as opposed to ongoing operations or decision-making functions, updating or correcting existing government databases, supporting national security uses, allocating risk, ensuring proper attribution,
and supporting commercial markets.10 Disadvantages of licensing can include increased acquisition cost in some instances; increased negotiation, coordination, administration, and enforcement costs; uncertain use and redistribution conditions; limited redistribution rights; inability to meet specialized needs; and loss of public domain effects. Restrictions on dissemination of data obtained by government from the commercial sector can impose large costs if these data are important inputs to research and development by businesses or the academic community. Nonetheless, licenses continue to evolve rapidly and are likely to improve over time. Suggestions from the commercial sector for promoting licenses include better contract design, validating licensed data to increase user confidence, developing standard form licenses, and simplifying negotiations.
LICENSE-BASED MODELS THAT COULD MEET THE GEOGRAPHIC DATA NEEDS OF ALL STAKEHOLDERS
The suitability of a particular license model depends on the intended use of the data. Furthermore, agencies must weigh the perspectives of many different stakeholders to find licenses that best fit their missions and mandates; legal, economic, and public interest considerations; and budgetary realities. Depending on the circumstances, the advantages of licensing may outweigh social and economic drawbacks of acquiring geographic data with some level of use restriction.
Licensing of geographic data and works has come of age because of the limited protection afforded by copyright and other intellectual property doctrines in the digital environment. Providers also have turned to technological means to control access and copying, measures that are reinforced by the Digital Millennium Copyright Act for works that have at least some copyright protection. Moreover, courts have upheld contracts or licenses that limit the uses that a licensee can make of data, or that prohibit further distribution. Data providers’ rights are likely to be further strengthened if Congress adopts database protection.
Federal agency data acquisitions are also constrained by a variety of federal laws and regulations. Some federal laws and policies embody a strong preference for making data available to the public. Additionally, government accountability may require further public access, particularly in light of changes to the law regarding data access and data quality. Even so, documents such as OMB Circular No. A-130, the Federal Acquisition Regulations, and the Freedom of Information Act (FOIA) recognize the possibility that some government information will be subject to proprietary restrictions and cannot be disseminated or made public.
Society makes data investment decisions through two very different institutions: governments and markets. Deciding which sector should acquire and distribute a particular product has profound implications for economic efficiency. For goods such as data, markets are a good solution where the initial decision to invest in a particular product is controversial or uncertain. Conversely, government procurement is most useful when uncertainty about whether to invest is small, so that distributional efficiency becomes the dominant concern. Beyond these generalizations, additional considerations may apply to particular cases.
Agencies affect the government/market balance each time they acquire or distribute data. The challenge is to make these choices with an eye toward economic efficiency. License design can be an important tool for setting this balance. In general, licenses for data obtained from the private sector that contain modest use and redistribution rights promote markets by allowing the original suppliers to pursue additional sales. Conversely, the agency may think that certain geographic data products have proven their worth, but that high prices are preventing many people from using them. In this case, the agency may wish to make the data it acquires widely available by acquiring them through licenses that give the agency broad redistribution rights.13 Although such rights will limit any remaining private market for the product, that will be reflected in the price that the vendor demands, and the agency pays, for such a license. In return, efficiency in distribution is more likely to be achieved.
Traditional licensing models are not the only—or, in some cases, the best—ways to promote economic efficiency. For example, some firms will
See Chapter 6 for a more complete discussion.
In the limit, the agency may wish to purchase the data outright.
not mount large data acquisition programs unless agencies commit upfront resources. This can be done in a variety of ways, including private–public partnerships, licenses that obligate the agency to pay for large volumes of data, and cooperative research and development agreements. From the agency perspective, such transactions offer a good mix of efficiency in both production and distribution. Efficiency in production is achieved because the project must still realize significant private-sector sales to be profitable. Efficiency in distribution occurs because the agency often has significant leverage to demand license terms that permit widespread dissemination on favorable terms, perhaps by requiring the vendor to donate its data to the public after a fixed period of years, or otherwise limiting the private partner’s ability to impose high prices. Alternatively, government might bear the entire cost of data production and acquire unlimited rights in order to promote efficiency in distribution.
The Public Interest14
Public discourse, equality, and innovation are benefits that are not easily assessed but accrue for society as a whole. These benefits have been well served by public domain15 data, which have been the norm under a legal regime in which geographic data, once published, were free for anyone to use. Such data also serve government accountability and transparency, although some license restrictions also may support these public interests in some cases. National security, law enforcement, and privacy issues present a common challenge to policy makers considering geographic data access issues: how to weigh potentially harmful or intrusive uses against legitimate uses. Blanket restrictions and classification on national security or law enforcement grounds are inadvisable except in unambiguous cases. Furthermore, the potential benefits of classified data beyond the national security arena make timely declassification important. When classification is deemed appropriate, licenses can be used to limit access to specified users. Government also can use licenses to promote reuse of geographic data by negotiating terms that limit commercial firms’ ability to discard data prematurely, promoting uniform and high-quality metadata,16 and encouraging standards that make geographic
data interoperable across a wide range of hardware, software, and data products.
Recommendation 1: Before entering into data acquisition negotiations, agencies should confirm the extent of data redistribution required by their mandates and missions, government information policies, needs across government, and the public interest.17
Recommendation 2: Agencies should experiment with a wide variety of data procurement methods in order to maximize the excess of benefits over costs.18
Circumstances in Which the Need for Public Access Is Strong
When government uses data to promulgate regulations, formulate policy, or take other actions that affect the rights and obligations of citizens, there is a compelling interest in making these data available so that the public may understand, support, or challenge government decisions.19 This interest often will be served by acquiring unlimited rights in data, but also may be accommodated in some circumstances by licensing data under conditions that permit access for more limited purposes. For example, in some cases the public may only need access to views derived from satellite data, rather than the original satellite data. The important principle is that access to information cannot be so limited, its distribution so difficult, or its content so closely held by government that outcomes of political debates are determined by unequal access to data.
Recommendation 3: When geographic data are to be used to design or administer regulatory schemes or formulate policy, affect the rights and obligations of citizens, or have likely value for the broader society as indicated by a legislative or regulatory mandate, the agency should
Federal Geographic Data Committee, 1998, Data Content Standard for Digital Geospatial Metadata, available at <http://www.fgdc.gov/standards/documents/standards/metadata/v2_0698.pdf>.)
evaluate whether the data should be acquired under terms that permit unlimited public access or whether more limited access may suffice to support the agency’s mandates and missions and the agency’s actions in judicial or other review.20
Strategies for Government Data Acquisition Under License
Compared to other procurement methods, the benefits and costs of licensing tend to be complex. The importance of particular terms usually depends on context. Thus, there is no “golden rule” for determining which license restrictions are appropriate. That said, agencies usually need to weigh such terms as price, dissemination restrictions, enforcement, available uplift rights,21 and liability.
Recommendation 4: Agencies should agree to license restrictions only when doing so is consistent with their mandates, missions, and the user groups they serve.22
Although agencies are familiar with their own internal needs, it is important that they confirm that use restrictions are also acceptable to outside user groups included in their mandates and missions. This usually requires repeated, direct discussions with affected parties. Agencies also are trustees for the taxpayers they serve. This status includes an obligation to adhere to government’s information policies by acquiring data that meet needs across government and serve the public interest. Because beneficial downstream uses and the public’s interest in the free flow of information cannot be fully anticipated, agencies should exercise caution in construing their mandates and missions to permit licenses that restrict such uses.
Recommendation 5: Agencies that acquire data for redistribution should take affirmative steps to learn the needs and preferences of groups that are the intended beneficiaries of the data as defined by the mandates and missions of the agency. Agencies should avoid
The Agency as Licensor
Most federal agencies and many state and local agencies routinely limit the fees they charge for data to the marginal cost of distribution.25 However, federal agencies sometimes charge higher fees under specific statutory exceptions or to honor commercial restrictions on data previously obtained under license. Some state and local government agencies may choose (or are legally required) to set fees above marginal cost. In this case, agencies could limit the impact of such a decision by adopting the following strategies: (1) adopt price discrimination to mitigate the economic inefficiencies associated with user fees, (2) charge the lowest price consistent with covering their variable costs (if the goal is to finance ongoing operations while still providing affordable public access), and (3) present minimally restrictive contracts that offer a broad menu of licensing options. Even when cost recovery is not a goal, agencies sometimes may use licenses to pursue other, nonfinancial policy goals (e.g., ensuring attribution, negating implied endorsements, and managing risk). These provisions should impose minimal restrictions on licensees’ ability to use and redistribute data.
Accommodating a Culture of Licensing
Most data vendors’ standard terms and prices are negotiable, particularly for large transactions. When circumstances permit, agencies may want to demand fewer rights in exchange for lower prices. Agencies also
may be able to offer in-kind payments to vendors in order to lower dollar costs still further.
Given the expansion of licensing of geographic data in the marketplace, agencies cannot help becoming more sophisticated consumers when licensing provides the best value or is the only available means of acquiring geographic data. The committee learned of many positive examples of agency negotiation. However, some agencies seemed to believe that they could not negotiate from a position of strength or found negotiations burdensome. As a result, some agencies indicated that they accepted vendors’ opening offers at face value with little or no negotiation. Not coincidentally, these same agencies tended to have the most disappointing licensing experiences.
Recommendation 6: Agencies should dedicate resources to training and knowledge-sharing among agencies in order to extract maximum public benefit from licensing. The Federal Geographic Data Committee’s working group and subcommittee structure provides a convenient venue through which agencies can report and learn from their experiences.26
POTENTIAL EFFECTS ON GEOGRAPHIC DATA USES AND GEOGRAPHIC TECHNOLOGY DEVELOPMENTS OF COMPETING LICENSE/NON-LICENSE APPROACHES WITHIN THE COMMERCIAL SECTOR
An earlier section of this summary27 presented the contrasting interests of various groups in licensing and some of their concerns about effects of license and nonlicense approaches on geographic data use and technology. In addition, vignettes between the chapters of this report present visions for the effects of license and nonlicense approaches. Realization of these visions hinges on whether policy and/or technological solutions can be developed to address a license or nonlicense option. With each vignette, the vision builds in complexity to illustrate a possible future that accommodates the broadest range of stakeholders in geographic data and services. Chapter 9 lays out specific strategies and institutions that could help the geographic data community reach this goal through positive effects on geographic data uses and in directing technology developments.
OPTIONS THAT WILL BALANCE THE INTERESTS OF ALL PARTIES AFFECTED BY LICENSING
Standard Licenses and Form Agreements
Today’s geographic data contracts span a wide range of language and levels of complexity. At a minimum, it should be feasible to standardize provisions covering liability, indemnity, attribution, jurisdiction, and choice of law. Standard language and (eventually) standard form licenses are key to advancing many of the ideas in this report. Likely benefits include reduced negotiation costs, reduced uncertainty, improved market efficiency, and greater ability to automate transactions.
Recommendation 7: Agencies, trade associations, and public interest groups should exercise leadership in promoting standard clauses and form licenses throughout the geographic data community.28
Coordinating Government Acquisitions
Agencies often develop interagency approaches to prevent duplicate procurement of data. One strategy is to rely on a “lead agency” to purchase licenses on behalf of a larger group. Alternatively, agencies can purchase uplift rights when they acquire licensed data that are likely to be reused elsewhere in government.29 Institutions such as data brokerages or automated business-to-government purchasing systems could strengthen this strategy.30 However, these reforms may not be practical in the near term.
Recommendation 8: Agencies should continue to keep abreast of data brokerage and automated purchasing system developments that might help them coordinate data acquisitions from competing vendors.31
Data brokerages enable users to search for previously licensed data. Business-to-government purchasing systems enable automated purchasing of standardized commercial products by government.
Toward an Integrated National Commons and Marketplace in Geographic Information
Facilitating the sharing of and trade in data through the development of an efficient and user-friendly system, including a well-organized commons connecting users and contributors and an efficient market connecting buyers and sellers, would be a valuable endeavor.32 Although no such online environment currently exists for geographic data, The National Map, Geospatial One-Stop, and the National Spatial Data Infrastructure provide first steps.
The National Commons
The overarching goal of the geographic information commons is to create a broad and continually growing set of freely usable (i.e., no monetary charge for use) geographic data and products at local scales similar in effect to the public domain datasets and works created by federal agencies. To succeed, the commons could provide easy, effective, and integrated mechanisms that
enable any geographic dataset creator to construct a license that grants others permission to use his or her data,
enable novice creators to quickly generate accurate and substantive standardized metadata for a geographic data file,
enable data contributors to take advantage of form liability disclaimers,
embed identifiers automatically in any commons dataset so that future users can link back to and recover detailed metadata and license conditions,
allow for deeper search capabilities of geographic data and metadata than are currently available, and
provide a long-term archive for commons geographic datasets.
Not all local governments, private citizens, or private companies will want to make any or all of their geographic datasets or products available in the public domain or in a commons licensing environment. Nevertheless,
more people likely will make their data freely available when the integrated mechanisms described above become available.
Recommendation 9: The geographic data community should consider a National Commons in Geographic Information where individuals can post and acquire commons-licensed geographic data. The proposed facility would make it easier for geographic data creators (including local to federal agencies) to document, license, and deliver their datasets to a common shared pool, and also would help the broader community to find, acquire, and use such data. Participation would be voluntary.33
The Internet has greatly enhanced the ability of commercial businesses, government, nonprofit organizations, and citizens to find comercial geographic data meeting their needs. A National Marketplace in Geographic Information would provide an online environment where any seller or licensor, no matter how small, could efficiently post its offerings in a searchable form using a menu of standard licenses and metadata reporting. Would-be purchasers could search through thousands of offerings to find the geographic data that meet their technical and license condition needs.
In the simplest implementation of the marketplace, purchasers would obtain the data directly from the vendor after “clicking through” to contact its server. Minimal investment could provide a combined license–metadata creation capability for sellers and search capability for consumers within a short time. In more advanced implementations, the seller or licensor might define detailed but standard license or sale conditions tied to sellerdefined pricing formulas and participate in automated financial transactions and downloading of products. Buyers would be able to accomplish efficient comparison-shopping and buy or license desired geographic data within minutes of finding it. Seller’s accounts could be automatically credited with funds from product sales, and sellers would be able to alter their geographic data offerings, descriptions, license conditions, and pricing formulas at any time.
Recommendation 10: The geographic data community should consider a National Marketplace in Geographic Information where
individuals can offer and acquire commercial geographic data. The proposed facility would make it easier for the geographic data community to offer, find, acquire, and use existing geographic data under license. Participation would be voluntary.34
Encouraging Data Donations to the National Commons and Marketplace
A potential add-on to the basic commons and marketplace facility is a “timed donation strategy.” To encourage donations, the following rule might be adopted: Creators who post a data file for sale over the “marketplace component” must at the same time deposit a copy of the data file in escrow to the secured archives of the National Commons and Marketplace. Escrowed files become available after five years through a commons license selected by the creator at the time of deposit or, if no commons license is generated, enter the public domain.35
This strategy is a natural extension to current USGS policies that use licensing to draw data into the public domain.36 The benefits of such a strategy include (1) offering voluntary participation, (2) encouraging agency culture to become more sensitive to commercial concerns and foster greater coordination between private and public sectors, (3) improving data archiving, and (4) reinvigorating the public domain in geographic data.
Recommendation 11: The geographic data community should consider a system of “data donations” in which anyone who sells data using the National Marketplace in Geographic Information automatically agrees to donate their data to the commons after a commercially reasonable time, which we provisionally set at five years.37
Operating the National Commons and Marketplace
Because they require similar software and hardware, the commons and marketplace components could be built simultaneously as a unified or closely integrated facility. Assuming consistent standards and processes, separate entities conceivably could host and operate the different components of the system. Whatever the chosen path, strong agency leadership will be needed to ensure that maximum benefits are achieved.
Recommendation 12: Federal agencies should investigate options for and encourage development of a National Commons and Marketplace in Geographic Information.38