The following statement has been updated and is included here for historical purposes. This version does not represent statements of the AMS that are “in force” at this time.
(Adopted by AMS Council on 13 January 2002)
Bull. Amer. Met.Soc., 83
Publicly financed environmental data should be regarded as a public good and made publicly available at costs commensurate with replication and transmission. Where practical, these data should be provided free of cost, and no limits should be placed on their redistribution.
The existence of privately funded compilations of publicly financed data should not preclude open public access to the original data; nor should it preclude the creation of publicly financed and publicly accessible compilations of the data.
Principal investigators should make publicly financed research data publicly available at minimal cost and as soon as feasible. Arrangements for the exclusive use of datasets by the principal investigators for some period of time (preferably as short as possible) should be negotiated with the funding agency or agencies.
In circumstances in which environmental data are acquired or generated through combinations of public and private funding, every effort should be made by the public institutions involved to ensure that public access to the data is as free and open as possible, in keeping with the spirit of this Statement.
The American Meteorological Society encourages private institutions that serve as sources of environmental data to anticipate circumstances in which some subset of their data may become critical to the public welfare and to prepare channels of communication by which that critical data can be transmitted to the appropriate civil authorities when and if such circumstances arise. In addition, the Society encourages private sources of environmental data to make such data freely available if and when their proprietary utility expires.
Few instances of international cooperation can match the success of the practice of free and open exchange of meteorological data that began when telegraphy was introduced more than 100 years ago. The benefits that accrue to the citizens of all nations as a result of this practice have been so great that meteorological data were exchanged in many cases even between nations at war. Historically, those nations with large economies vulnerable to weather have expended the most resources in acquiring meteorological data; thus there is a rough proportionality among the amount and quality of data provided by each nation, the value of those data to the nation in question, and the ability of that nation to shoulder the costs. For example, the enormous costs of developing and deploying environmental satellites have been borne by a few of the developed nations, while all nations have benefited from satellite data. This proportionality has contributed to a sense that the practice of free and open access to environmental data is both beneficial and fair to all concerned.
It would be difficult to overstate the benefits to society of free international exchange of environmental data. Weather forecasts beyond a day or so depend on meteorological data collected over a vast region spanning many nations and international waters. Monitoring and predicting climate change require global environmental datasets and are of intrinsic value to all nations. Research and education in the environmental sciences depend crucially on the low cost and free accessibility of environmental data, and the peoples of all nations have benefited from the progress thus made possible in understanding and predicting environmental hazards.
The free and open provision of environmental information is one of the most important ways that the developed world assists underdeveloped nations. Attempts to recover the costs beyond reasonable costs required to provide access and delivery of environmental data have a disproportionately negative effect on those nations least able to afford the data.
Several developments over the last decade threaten the international practice of free and open exchange of environmental data. As citizens of many nations demand reduced taxes and increased efficiency from their governments, those governments are under increasing pressure to recover the costs of the services they provide, particularly if those services are seen as benefiting for-profit enterprises or the interests of other nations. And, as the economic value of environmental data inexorably increases, some private interests are pressuring their governments to reduce or cease activities they view as taxpayer-subsidized competition and to take measures to limit or eliminate competition from abroad. At the same time, the private sector itself is increasingly involved in the generation of environmental data and has a legitimate interest in protecting its intellectual property.
The primary purpose of this Statement is to reassert the American Meteorological Society's commitment to a policy of free and open international exchange of environmental data, while at the same time endeavoring to draw critical distinctions among different types of environmental information. In the following section, we detail a few key, guiding principles that serve as the foundation for this Statement. In section 3 we define key terms that are used in the Statement.
2. Guiding principles
a. Public ownership and public goods
Goods and services that are funded by taxpayers and are intended for use by the public are generally conceived of as being publicly owned. Thus, in a broad sense, the U.S. interstate highway system is publicly owned. The "public" in this sense consists generally of the citizens and permanent residents of the political entity to which the taxes are paid, but in some circumstances the sense of "public" extends beyond ownership. For example, the interstate highway system may be used by foreign visitors, and state highways may be used by residents of other states or countries. On the other hand, national health care systems are not intended to pay for the health care of foreign nationals, except perhaps in emergencies.
A good or service that is nondepletable is generally referred to as a "public good," whether or not it is publicly owned. Nondepletable goods and services are those that are not depleted by consumption. A public good is also often nonexcludable. A nonexcludable product is one that once having been supplied to some, cannot easily be denied to others. An environmental datum is an example of a public good; its utility is neither depleted by virtue of being accessed nor easily denied to some after being provided to others.
It is the position of the American Meteorological Society that all publicly financed environmental data should be regarded as a public good. In particular, we assert that environmental data, even when collected over a small region, are of intrinsic value to the public broadly conceived, both through their potential real-time value to public health and welfare, and through their value in enhancing the understanding of the environment through scientific research. For these reasons, it is highly beneficial to make publicly financed data openly available to the global public.
b. Cost recovery
Under some circumstances, use of publicly owned goods and services may incur usage fees. Regulation of demand or compensation for excessive demand sometimes justify the imposition of such fees. For example, the U.S. interstate highway system is supported through a variety of general taxes, including income taxes and taxes on fuel. But commercial trucks pay an extra road usage tax, because heavy trucks are responsible for a disproportionate amount of wear and tear on the highways. On the other hand, police and fire departments protect public property, public gatherings, private residences, and business enterprises without charging fees for basic services.
In the case of publicly financed environmental data, efforts to recover costs from users often result in no more than a reshuffling of funds among public agencies. For example, if a government environmental agency imposes fees for the use of data it collects and disseminates, the fees are often paid by other public agencies, such as public schools or government laboratories, or though government grants and contracts awarded to researchers. Moreover, if one political entity charges fees for the use of its environmental data, it must expect in turn to be charged for data it receives from other political entities. A "trade war" in environmental data can only result in greater costs for everyone involved and increasingly limited public access to the data.
Cost recovery or a sense of competition with the private sector has been used to justify the imposition of usage fees only on for-profit users of publicly financed environmental data. Apart from the difficulty of preventing such users from procuring the data indirectly through "legitimate" users of the data, such a policy can stifle the growth of private businesses that often serve valuable public functions, and it may paradoxically reduce net revenue to the government by reducing or eliminating taxable business profits.
In some cases (outside the U.S.), usage fees imposed on commercial enterprises have been used to expand significantly the range of goods and services provided by government entities. In these cases, the government entity is acting effectively as a private contractor, and the goods and services it provides cannot be said to be entirely publicly financed. It becomes difficult, if not impossible, to distinguish between public and private goods, and the government entity is gradually converted from a public agency to a government-subsidized private contractor. Significant, direct funding of public agencies by private interests risks subverting public enterprise to private interests.
For all these reasons, the American Meteorological Society is opposed to efforts to recover the costs of publicly financed environmental data, above and beyond any reasonable excess costs associated with dissemination of the data.
c. National security
A primary obligation of governments is to protect its citizens from direct external threats to their health and well being, as for example during times of war. During such times and in response to such threats, considerations of public safety may warrant selective, temporary restrictions on the dissemination of environmental data.
d. Economic security
Some governments have attempted to impose selective usage fees or other restrictions on foreign users of publicly financed environmental data. The World Meteorological Organization, through its Resolution 40, explicitly states that nations or economic blocks may be justified in imposing restrictions on the reexport of environmental data for commercial purposes . As in the case of domestic restrictions on data distribution to private enterprises, it is difficult to prevent commercial enterprises from procuring the data indirectly through 'legitimate' users of the data, thereby encouraging limitations on the distribution of certain categories of data. It is the position of the Society that conditions on the reexport of publicly financed environmental data work against the public welfare and should be discouraged.
e. Intellectual property
Research results, books, databases, analyses, computer programs, and other fruits of intellectual activity that make use of publicly financed environmental data may often be considered to be owned by some combination of the individual or individuals who developed the products and those who funded them, whether private enterprises or public agencies. There is a large body of copyright and patent law designed to protect the intellectual property rights of such individuals, enterprises, and agencies. Such intellectual property, though its development may have depended critically on publicly financed data, often cannot be considered as publicly owned property, except under certain specific contractual arrangements where the intellectual activity was itself publicly financed.
There is much room for ambiguity between the extremes of privately held intellectual property and publicly owned data. Of particular concern are what are loosely termed "data compilations." For the present purpose, we will focus on compilations of publicly financed data. "Originality" is the primary requirement for the protection of intellectual property under U.S. copyright law. However, in a landmark case decided by the U.S. Supreme Court in 1991, the amount of labor or funding that may have gone into the creation of a data compilation does not by itself qualify the database as "original" for the purposes of copyright protection. Thus relatively unoriginal compilations of environmental data, such as arrangements of data in alphabetical order, do not qualify for copyright protection under U.S. law.
At the other extreme from U.S. copyright policy is the policy advanced by the European Union (EU) under its 1996 Database Directive, governing the legal regulation of noncopyrightable databases. This directive gives legal protection to databases, defined very broadly as any "collection of works, data or other independent materials arranged in a systematic or methodical way and capable of being individually accessed by electronic or other means." Of particular concern is the Directive's establishment of a "Sui Generis" right that protects not only the database itself but also, by extension, the content of the database. The way the Directive is formulated, it can be interpreted as allowing the acquisition of property rights even over long-established publicly financed datasets.
It is the position of the American Meteorological Society that the establishment of intellectual property rights must never compromise the availability of publicly financed data. While recognizing the intellectual property rights attendant on the development of new means of arranging and presenting data, such rights should never be formulated so as to restrict access to the publicly financed data on which they are based.
f. Environmental data acquired through public-private partnerships
In some cases, the acquisition and dissemination of environmental data are funded through a combination of public and private sources. The objectives of such enterprises are highly diverse, as are the reasons for securing multiple funding sources; we regard it as beyond the scope of this Statement to formulate a general policy concerning the free and open exchange of such data. The Society encourages the public institutions involved in such arrangements to explicitly address the issue of free and open exchange of environmental data in formulating public-private partnerships, so as to conform as much as possible to the spirit of this Statement.
The definitions laid out in this section pertain to the use of terms in the AMS Statement. The word "data" as used here is taken to mean numbers, graphical images, or descriptions of such numbers or images, stored in electronic or magnetic media, punch cards, or on paper.
Those data consisting of measurements of the atmosphere, hydrosphere, biosphere, cryosphere, or the solid earth. This definition includes data that consist of "raw" measurements as well as reductions of such measurements that reflect application of quality control measures or the calculation of derived quantities, and analyses of such data in the form of gridded data and graphical images. Descriptions of the data (metadata) and forecast values of environmental variables, made using statistical and/or numerical techniques, are also considered to be forms of environmental data.
A subset of environmental data as described above, procured exclusively for scientific research. Data that have been collected for the purposes of environmental monitoring and prediction but are also used in scientific research are not categorized as "research data" for the purposes of this Statement.
Environmental data critical to public welfare
A subset of environmental data as defined above that may be used to reduce exposure of the public to a threat to its welfare. Examples of such data include measurements that reveal a leak of a toxic substance or a life-threatening natural hazard such as a tornado.
Environmental data compilations
Those arrangements of environmental data designed to facilitate access to them. These include ordered arrangements of data and software whose main purpose is to facilitate access to data.
Paid for using proceeds from government-imposed taxes. Here no distinction is made on the basis of the size of the political entity imposing the taxes; proceeds from local, state, or national taxes are all considered as examples of public financing.
As the term is used here, a nondepletable and often nonexcludable product or service made freely available to the global community and regarded as beneficial or potentially beneficial to some or all members of the global public. A "nondepletable" good is one that cannot be diminished by repeated use, while a "nonexcludable" product is one that once having been supplied to some, cannot easily be denied to others.
The following references were used as guidance in preparing this Statement.
Hodge, G., 2000: After the EU Directive: The impact of European database protection on stakeholder groups. Bull. Amer. Soc. Inf. Sci., 26, 3. (http://www.asis.org/Bulletin/Mar-00/hodge.html)
Kaul, I., I. Grunberg, M. Stern, Eds., 1999: Global Public Goods: International Cooperation in the 21st Century, 1st ed., United Nations Development Programme, Oxford University Press, 584 pp.
National Research Council, 1995: On the Full and Open Exchange of Scientific Data. Nat. Acad. Press, 21 pp.
—, 1997: Bits of Power: Issues in Global Access to Scientific Data. Nat. Acad. Press, 235 pp.
—, 2001: Resolving Conflicts Arising from the Privatization of Environmental Data. Nat. Acad. Press, 99 pp.
Reichman, J.H., and P. F. Uhlir, 1999: Database protection at the crossroads: Recent developments and their impact on science and technology. Berk. Tech. Law J., 14, 2. (http://www.law.berkeley.edu/journals/btlj/articles/)
Stiglitz, J., P. Orszag, and J.M. Orszag, 2000: The Role of Government in a Digital Age. Report commissioned by the Computer and Communications Industry Association. (http://www.ccianet.org/govt_comp.php3)
Weiss, P. N., and P. Backlund, 1997: International information policy in conflict: Open and unrestricted access versus government commercialization. Borders in Cyberspace: Information Policy and the Global Information Infrastructure, B. Kahin and C. Nesson, Eds., MIT Press, 300 pp.
World Meteorological Organization, 1996: Exchanging Meteorological Data: Guidelines on Relationships in Commercial Meteorological Activities. WMO Policy and Practice. WMO Rep. 837, Geneva, 24 pp.