ADMINISTRATIVE NOTES Newsletter of the Federal Depository Library Program --------------------------------------------------------------------- November 15, 1999 GP 3.16/3-2:20/17 (Vol. 20, no. 17) --------------------------------------------------------------------- Michael F. DiMario Public Printer Remarks Before the Conference on Government Information Issues in the 21st Century On the Foundations of Federal Public Access Policy Southern Illinois University Carbondale, Illinois September 17, 1999 Good morning. I want to thank Jeanne Simon for inviting me to speak. Jeanne and I have kept in close touch during her tenure as Chair of the National Commission on Libraries and Information Science. Her advice and guidance have been very helpful to me and the GPO in our work with Congress, the Administration, and the library and government information communities. She is doing a great job as Chair of the Commission and I am very happy that the President appointed her for a second term. I’m sure all of us here today send a "get well" message to her and wish her the best of luck. As this conference is on "Government Information Issues in the 21st Century," I thought it would be useful to go over the historical, constitutional, statutory, and administrative foundations of our current policy for providing access to Government information, and then comment briefly on how those foundations are impacted by technology and the issues it’s raising in the Information Age. Gutenberg’s Invention With the Year 2000 approaching, a number of retrospectives have been published about the events of the previous millennium. Many of these name the most significant events of the era. Which would you say are the most important? Nuclear energy? Columbus’s voyage to America? Vaccinations? The automobile? Life magazine is reporting that the most significant event of the past 1,000 years was Johann Gutenberg’s development, in 1455, of movable type. (Cited in Graphic Communications World, July 12, 1999.) This was not simply a means of producing ink-on-paper more economically. Gutenberg invented the concept of easily transferring language to a medium for widespread dissemination. The technologies we have today for accomplishing the same end—computers, e-mail, online systems—are indebted to this invention. With movable type, printing flourished. In 1455, there were fewer than 30,000 books in the world, all hand-made. By 1500, there were more than 9 million. (Graphic Communications World.) With increased printing came the spread of ideas, and with that came the Reformation, the Age of Discovery, and the Enlightenment. The transmission of ideas and information through printing eventually ushered in the revolutionary notions of democracy and self-government. During the ratification of the Constitution, William Cushing, Chief Justice of Massachusetts, wrote in a letter to John Adams, that "[t]he propagating of literature and knowledge by printing or otherwise tends to illuminate men’s minds and to establish them in principles of freedom…" (Letters of William Cushing and John Adams, 27 Massachusetts Law Quarterly 1942, p. 14, cited in David Mitchell Ivester, "The Constitutional Right to Know," 4 Hastings Constitutional Law Quarterly 1977, p. 121.) One of the most important results of Gutenberg’s invention was that it fundamentally altered the relationship of people to their governments. This change was brought about by the concept of providing public access to information by and about government. We consider this notion absolutely basic to our form of government. But what are its bases in our constitutional and statutory history, and what do those foundations imply for the direction of public access in the 21st century? An examination of the roots of Federal public access policy can point to both the positive and troubling consequences of the use of electronic information technology from the standpoint of public access. Historical Background Public Access and Self-Government The foundations of the "right to know"—the right of public access to government information—lie in the basic principle that self-government requires the public to be informed of the activities of the government. All of us here this morning are familiar with James Madison’s famous dictum about a "popular government without popular information," and that "a people who mean to be their own Governors must arm themselves with the Power which knowledge gives." (Letter from James Madison to W.T. Barry, Aug. 4, 1822, in The Complete Madison, S. Padover ed., 1953, p. 337, cited in Ivester, p. 123.) There is a requirement in the Constitution that Congress publish a journal of its proceedings. But some observers have argued that the right to know is inherent in the structure of self-government itself, and because of this, it can be assumed to be as much a part of the Articles I and II of the Constitution as it is the First Amendment to the Bill of Rights. (Ivester, p. 117, citing Parks, "The Open Government Principle: Applying the Right to Know Under the Constitution, 26 George Washington Law Review 1957.) Thus, according to these observers, even though the right to know isn’t a specifically enumerated right, it doesn’t have to be. Instead, the right to know is deeply embedded in our system of government as established by the Constitution, limited only in cases where there is a compelling state interest, such as national security, privacy, to protect the integrity of investigations, and so on. Foundations of Public Access in England The roots of the principle of public access to government information extend directly to the early years of printing. With the development of movable type there was an explosive period of growth in the printing trade, leading to the formation in England and other countries of printing guilds. By the 17th century in England, two developments—a period of religious and political ferment and the proliferation of non-guild printers—led to a move in Parliament to license printing presses. This action was backed by the Stationers Guild, which sought exclusive rights to printing work, and by those in Parliament who opposed growth in the publishing of ideas they disagreed with. The licensure proposal was opposed by the poet John Milton, who had published a non-licensed tract arguing for a more liberal view of divorce—namely, his own. In the now-famous essay Areopagitica— published, incidentally, without a license in 1644—Milton argued, "Give me the liberty to know, to utter, and to argue freely according to conscience, above all liberties…" (John Milton, Areopagitica: A Speech for the Liberty of Unlicensed Printing, to the Parliament of England, 1644, cited in Ivester, p. 125.) Milton’s essay has endured through the ages as a classic defense of freedom of the press. The efforts by the Stationer’s Guild to circumscribe the activities of non-guild printers ultimately failed. With the growth of printing in England, both licensed and unlicensed, there was an expanding interest in the open discussion of governmental matters. Sounds a lot like today’s Internet, doesn’t it? At the same time, there was a gathering movement to gain greater access to governmental information, including proceedings in Parliament. This was fueled in part by the new interest in government and in part by the printers who were responsible for producing this information until Her Majesty’s Stationery Office was established in 1776. Colonial Experience Perhaps in reaction to the attempts to license printing in England, some of the early American colonists imported fairly liberalized notions of public access to government information and codified them into law. For example, the Massachusetts Body of Liberties, produced in 1641, stated that, "Every Inhabitant of the Country shall have free libertie to search and veewe any Rooles, Records, or Regesters of any Court or office…" (1 The Laws and Liberties of Massachusetts, 1641-1691, J. Cushing ed., 1976, cited in Ivester, p. 127.) But other colonial governments from the beginning closely watched the activities of their printers. Printing first reached America in 1639 when a press was established for the use of Harvard College. Most early printing was official crown printing—performed by "publick printers"— and it was closely controlled. (Government Printing Office, 100 GPO Years, 1861-1961, GPO: Washington, 1961, p. 1.) Despite the provision for public access in Massachusetts, censorship was established there by 1664, and in 1671 the Governor of Virginia said: But, I thank God, there are no schools nor printing…for learning has brought disobedience, and heresy, and sects into the world; and printing has divulged them, and libels against the best government. God keep us from both. (GPO, p. 2, see also Robert E. Kling, Jr., The Government Printing Office, Praeger: New York, 1970, p. 6, citing The Statutes at Large, being a collection of all the laws of Virginia, from the first session of the legislature, in the year 1619, W. Waller, ed., New York, 1823, vol. 2, p. 517.) As in England, however, the growth of printing—which produced newspapers, tracts, pamphlets, and other materials—spurred in the colonies a growing demand for independence of the press as well as access to government information. Moreover, critical examination of the government was gaining popular approval in the colonies. In 1735, in a case that was closely watched, printer John Peter Zenger won acquittal from a charge of libel brought because of his criticism of colonial government. The case helped solidify the argument for freedom of the press, interpreted broadly as the freedom not simply to print about the government, but to acquire information from the government to examine and debate. (see Ivester, pp. 127-28.) Public Access and the New Government By the time of the American Revolution, the concept of public access to government information was widespread throughout the colonies. The 1776 Pennsylvania Constitution, for example, included the phrase: "The printing presses shall be free to every person who undertakes to examine the proceedings of the legislature, or any part of government." (Pennsylvania Constitution of 1776, § 35, in 3 The Federal and State Constitutions, Colonial Charters, and Other Organic Laws, F. Thorpe, ed., 1909, cited in Ivester, pp. 129-130.) This concept was adopted by the new national government. The Continental Congress passed a resolution "…that the Journals…except such parts as have been or shall be ordered to be kept secret, be printed…weekly…" (cited in GPO, p. 4.) This formulation was to be used repeatedly thereafter. Article 9 of the Articles of Confederation said, "The Congress of the United States…shall publish the journal of their proceedings monthly, except such parts thereof, relating to treaties, alliances or military operations, as, in their judgment, require secrecy…" (cited in Merrill Jensen, The Articles of Confederation, Madison: University of Wisconsin Press, 1940, p. 269.) During the Constitutional Convention of 1787, the issue of public access to government information was specifically discussed, and the debates were over the extent to which the government could impose secrecy. Many delegates held that secrecy was not objectionable where there was a compelling interest, such as in foreign relations or war. But in general, open access to government information was held to be the rule, and secrecy the exception, rather than the other way around. Patrick Henry said, "The liberties of a people never were, nor ever will be, secure, when the transactions of their rulers may be concealed from them…" (4 Debates in the Several State Conventions on the Adoption of the Federal Constitution, J. Elliot, ed., 1901, p. 170, cited in Ivester, p. 131.) Like the resolution of the Continental Congress and the Articles of Confederation, Article I, section 5, of the Constitution requires that "Each House shall keep a Journal of its Proceedings, and from time to time publish the same, excepting such Parts as may in the Judgment require Secrecy…" Notice that the timing of publication had gone from "weekly" under the Continental Congress and "monthly" under the Articles of Confederation to "from time to time" under the Constitution. In forming a strong national government to replace the weak government under the Articles, the delegates to the Constitutional Convention were more willing than their predecessors to accept limitations on the right to know. In fact, delegate Oliver Ellsworth of Connecticut objected to this clause altogether, arguing that "The legislature will not fail to publish their proceedings from time to time. The people will call for it, if it should be improperly omitted." But in response, James Wilson of Pennsylvania said, "the people have a right to know what their agents are doing or have done, and it should not be in the option of the legislature to conceal their proceedings." (Elliot, p. 408, cited in Ivester, p. 132.) This statement today places public access to government information on a firm constitutional footing. Public Printing at the Federal Level In Madison’s statement on public access, made many years after the Constitution was written, he talked about the necessity of "popular information" to a "popular government," as well as need to have a "means of acquiring" this information. How did the early national government respond to the requirement that "the people have a right to know what their agents are doing or have done?" How did it provide the public with the "means of acquiring" government information? How did it enable the press and the public to stay informed of its actions? The answer to that question lies in the development of government printing and a system for distributing government documents. Until 1860, Congress—which then accounted for the majority of Federal printing work—fulfilled its printing needs through available private printers, operating in much the same way as the earlier British Parliament and the colonial governments. But as at least one observer has noted, "the National Government was not well served by this arrangement." (Harold C. Relyea, "Title 44, United States Code—Public Printing and Documents: A Brief Historical Overview," Congressional Research Service Report No. 79-36 GOV, February 23, 1979, p. 6.) Work was performed sporadically, and was often left uncompleted if it was undertaken at all. There are many examples of incomplete or nonexistent records from the earliest Congresses: in one, Congress was not provided with a complete set of the statutes of the United States until 1845. (Relyea, p. 6.) Other observers have remarked on problems with accuracy, appearance, and form, which varied widely and generating continuing complaints in Congress. (Kling, p. 10.) Moreover, these early systems of public printing in the United States were costly and subject to scandals, fraud, and corruption. Beginning with the First Congress, printing was contracted out to the lowest bidder. After the Printing Act of 1819 (3 Stat. 538), the Senate and House of Representatives elected their own printers, and contracts were let to each based on a fixed-rate basis. Technology improvements soon lowered the costs of printing dramatically, and the fixed-rate system proved to be inordinately profitable. Congress re-established the system of low-bid contracting in 1846 (9 Stat. 113), but it was so badly mismanaged that it proved more costly that its predecessor. The fixed-rate system was returned in 1852 (10 Stat. 35). Public outcry over abuses connected with Federal printing led Congress in 1846 (9 Stat. 113) to establish the Joint Committee on Printing, which was given the statutory authority to adopt measures "deemed necessary to remedy neglect, delay, or waste" in Government printing. This step toward gaining control of Federal printing was supplemented in 1852 (10 Stat. 35) with establishment of a single Superintendent of Public Printing, who was charged with overseeing the activities of printers for the House and Senate. But Congress wouldn’t go farther. Proposals for centralizing the Government’s printing capability had emerged as early as 1819, when a joint committee of Congress recommended establishing a "national printing office…which should execute the printing of Congress while in session, and that of the various departments of Government during the recess." No action was taken, however. A similar idea was advanced in 1840, accompanied by an estimate that a centralized office would reduce Federal printing costs by half, but this proposal also failed. By the late 1850’s, however, Congress was willing to revisit the issue. The renewal of the fixed-rate system for printing generated enormous profits for contract printers. The profits were converted into political contributions, which were given in exchange for preference in the award of printing contracts, or worse. One report found that: Politicians who had no practical knowledge of printing succeeded in securing the place of printer, and farmed out the work to practical printers at a percentage of the receipts. The dominant party elected the printer with a positive understanding that he would devote specified sums out of his profits for partisan purposes. The record shows that much of the printing of the executive departments was given out secretly and at extravagant figures. In some cases six times a fair rate was paid for certain jobs and the plunder thus secured was systematically distributed for partisan purposes. (Relyea, p. 9, citing Laurence F. Schmeckebier, The Government Printing Office: Its History, Activities, and Organization, Baltimore: Johns Hopkins Press, 1925, p. 7.) A number of congressional investigations ensued, culminating in 1860 in legislation to centralize all Federal printing in GPO (12 Stat. 117). When it opened in 1861, GPO was directed to execute "the printing and binding authorized by the Senate and the House of Representatives, the executive and judicial departments, and the Court of Claims." The benefits of the new office were realized almost immediately. In GPO’s first year of operation, the Superintendent of Public Printing [the precursor of the Public Printer] reported an annual savings of $60,000 over pre-GPO costs; in the second year, a savings of more than $200,000 was reported. Savings more than doubled again in the third year, to more than $580,000 by September 1863 (GPO, pp. 38-39.) Further centralization of Government printing followed in the succeeding years, and in 1895 all Federal printing and distribution systems were consolidated in GPO with the General Printing Act. This Act, in the words of one observer, marked "the institutional realization of a panoply of reforms pertaining to public printing policy." (Relyea, pp. 11-12.) The Act brought under GPO control other Federal printing plants then in existence and provided for the production of virtually all other Federal printing at GPO itself. Significantly, the Act also relocated the Office of the Superintendent of Public Documents, formerly at the Interior Department, within GPO, administratively completing—nearly a century later—the "means of acquiring" government information that Madison had envisioned. As a GPO official, the Superintendent was made responsible for documents sales to the public, cataloging and indexing Government publications, and the distribution of documents to depository libraries nationwide. By placing the Superintendent of Documents within GPO, Congress created a system that facilitated the efficient selection of publications for public distribution from the comprehensive body of documents printed by GPO. Depository Library Program Before the Superintendent of Documents was connected with the GPO, the early national government struggled with the issue of distributing government documents to the public. Initially, special acts were passed periodically to distribute particular documents to State legislatures, governors, colleges and universities incorporated in each state, and historical societies. (see Joint Committee on Printing, Government Depository Libraries: The Present Law Governing Designated Depository Libraries, Joint Committee Print, 97th Congress, 1st Session, GPO: Washington, 1981, pp. 3-4.) These weren’t for archival purposes, since it is clear that the distribution of documents to governmental as well as educational bodies was for the ongoing use of these institutions. In 1813, however, the 13th Congress enacted a resolution (3 Stat. 140) providing for distribution of documents on a regular basis to these institutions for that Congress and "every future Congress," passing what was in effect America’s first Freedom of Information Act. This system of distribution, performed variously by the Librarian of Congress, the Clerk of the House, and the Secretary of State, remained relatively undisturbed until 1857, when the responsibility was transferred to the Secretary of the Interior, who was directed to distribute the documents to colleges, public libraries, athenaeums, literary and scientific institutions, and boards of trade or public associations designated by the Representatives. (11 Stat. 253; JCP, Government Depository Libraries, p. 4.) Subsequent laws required the Interior Secretary to receive, arrange, and distribute documents, to keep accurate statistics on the receipt and distribution of all documents, and to distribute maps and charts as well as books. Congress also provided Senators with the power to designate libraries, and the Interior Secretary was empowered to designate libraries and to remove depository status from libraries which weren’t "suitable" depositories. (11 Stat. 368, 379) In 1869, the post of Superintendent of Public Documents was established within the Interior Department, and his distribution responsibilities were broadened to include executive branch documents. (15 Stat. 292; Kling, pp. 111- 112.) Yet there were significant problems with document distribution in these early years, and until the 1895 General Printing Act was passed, the distribution and storage of government documents was, in the words of an observer, a "hopelessly haphazard operation." Here is a general characterization of the situation: Copies of documents were ordered with little regard for public or official interest. Of the 420 official depositories [by 1895], some were overwhelmed by mountains of government publications, while others received no regular distribution at all. Copies ordered for congressional use accumulated in Members’ offices until no storage space remained, at which time Senators and Representatives dispatched them to home libraries to crowd shelves often already jammed with other documents that had been obtained by direct distribution. Not wishing to offend Members of Congress, successive public printers stored leftover publications until natural deterioration solved the warehousing problem. In addition, no standard system for titling government documents existed. Consequently, practical cataloging was virtually impossible. (Kling, pp. 110-11.) The Printing Act was directed at resolving these problems by codifying all of the laws concerning the receipt, care, and distribution of government documents. Order was imposed on the system of documents distribution: proper storage was organized, excess distribution was curtailed, standardized cataloging was implemented, and duplication and waste were eliminated. Other major reforms followed: an act of 1913 made depository status permanent, to end the practice of removing depository status when redistricting occurred or a new Member was elected (38 Stat. 75); and in 1922 depository libraries were authorized to designate the publications they wished to receive, rather than receiving every government document that was produced (42 Stat. 436). These reforms continued through the 1962 Depository Library Act and remain in effect today. Current Issues Following Gutenberg’s invention, the great hurdle of the 17th and 18th centuries was to gain acceptance of the concept of public access to government information, and to give it a fundamental, constitutional basis. In 19th century America, the hurdle was to design a statutory and administrative system that made public access a working reality. As we approach the 21st century, the challenge that confronts us is how to preserve the principle of public access in the age of electronic information technologies. Continued Need for Printing With the growing use of electronics, there is the temptation to say that the government no longer needs a printing capability. I think this temptation should be resisted. Last year, GPO produced more than $650 million in printing services, sold millions of documents, and distributed millions more to depository libraries. (GPO Annual Report, Fiscal Year 1998.) Tax forms, press releases, passports, legislative documents, informational pamphlets and books, regulations, statutes, statistical data, and more—in printed form these documents continue to represent a major avenue of communication between the government and the public. Ink-on-paper today is still the most egalitarian of information formats. It is accessible, transportable, and economical. The increased dissemination of government information in electronic formats should indeed be pursued: improving our economy, sharpening our trade competitiveness, expanding our research and development capabilities, promoting educational opportunities, and a vast range of other public objectives depend on it. But protections must be provided for those who do not have access to computers and the other technologies necessary to make electronic access meaningful. If we are not careful about maintaining policies to provide for the efficient and equitable access of all citizens to government information, we run the risk of turning into a Nation of information-haves and information have-nots. I read the other day that white, high income Americans continue to predominate in the online world, according to a recently-released Commerce Department study. (Commerce Department, "Falling Through the Net: Defining the Digital Divide,"1999, as reported in the Washington Post, July 9, 1999.) Even the President has said: [t]here is a growing digital divide between those who have access to the digital economy and the Internet and those who don’t, and that divide exists along the lines of education, income, region, and race…If we want to unlock the potential of our workers, we have to close that gap. (Ibid.) Public spending on Internet access in libraries is making a real difference in closing this gap, yet the "digital divide" remains a problem. Until the gap is erased, printing remains an effective safeguard for ensuring that those without access to computers can still use government information, and for guaranteeing both the authenticity of official government information as well as permanence. Simplifying Public Access In the electronic information era, there is a strong need to provide systems that help people find the information they want. In recent years, in response to a variety of social, governmental, and economic issues, Congress has passed hundreds of laws requiring public disclosure. These were not the Freedom on Information Act or government-in-the-sunshine laws, but statutes requiring the dissemination of information on a wide variety of Federal programs. Such statutes were a common feature of Federal legislation in the 1960’s and 1970’s. A few years ago, while we were conducting our study on the electronic transition of the depository library program, the Congressional Research Service compiled a list of the various statutes requiring or authorizing the dissemination of Government information to the public. Approximately 400 statutes were identified. (Jane Bortnick Griffith and Harold Relyea, Congressional Research Service, "Compilation of Statutes Authorizing Dissemination of Government Information to the Public," January 29, 1996.) But when you layer on top of that the presence of the Internet and the World Wide Web, which are now used by each of the Federal entities covered by those 400 statutes to perform their required dissemination functions, what results is a tremendously decentralized system of information dissemination. In 1997, the General Accounting Office identified more than 4,300 web sites in the Federal Government. (GAO, "Internet and Electronic Dial-Up Bulletin Boards: Information Reported by Federal Organizations" GAO/GGD-97-86, June 16, 1997). The mathematics of this situation are astounding. In fact, we now have highly disorganized and confusing system of public access. GPO and NTIS are about the only entities where there is some degree of coordination of information dissemination. Throughout the rest of the government, as shown by NCLIS’s recent report "Assessment of Electronic Government Information Products," "responsibility for electronic publishing within agencies is decentralized, diffuse, and unclear," and "there is an overall lack of Government information policy guiding electronic publishing, dissemination, permanent public access, or information life cycle management." ("Assessment of Electronic Government Information Products," report of the National Commission on Libraries and Information Science commissioned by the Government Printing Office, March 30, 1999.) Ominously, the report concluded that "the concept of permanent public access is not well understood" among Federal agencies. Now, a reasonable argument can be made that a democratic society is better served by multiple points of access to government information than by a single or limited number of resources. Censorship is less easily imposed in such an information environment. But providing for the effective coordination of government information, and reducing and eliminating the confusion and frustration of finding that information, shouldn’t summon up images of Big Brother, nor should it spell the end of multiple points of public access. As information dissemination grows more decentralized because of technology, the need for organizing public access to government information—through bibliographic services, cataloging and indexing, locator and Pathway services—grows stronger. Decentralization undercuts the goal of public access, and in today’s information-dependent economy and society, information delayed is information denied. We need to combat the confusion, frustration, and delays imposed by decentralization with improved services to help the public find what they want quickly and inexpensively. "Access America" We’ve been approached by the Office of Management and Budget to participate in the new "Access America" initiative, which was recently discussed at the annual American Association of Law Libraries conference in Washington, DC. There are a number of purported "one- stop" shops on the web for government information, and our site, GPO Access, tries to be one of them, although none of them really is—although some are a great deal closer to this ideal than others. We agree in concept with OMB’s efforts, but I wonder whether this isn’t simply another layer of web services which in the end will only be a temporary fix to the real problem of decentralized points of access. There’s a question about whether this will be a content-oriented site or a directory, like GILS. And I question whether the host agency for this undertaking—the General Services Administration—is the appropriate agency to take this on. To my knowledge, the only government-wide program whose express purpose is to act on the principle of providing effective, comprehensive, and equitable access to information produced by all three branches of the government is the depository library program. In this age of a multi-trillion dollar government, this program performs on an annual budget of a little less than $30 million. To tackle the problem of finding government information, why not invest in the agency whose mission—whose traditional and historic partnership with the library community—is to achieve that goal? Draft Legislation Recently, in response to draft legislation prepared by the Inter-Association Working Group in the library community, we proposed legislation that would improve upon and expand our authorities under chapter 41 of Title 44—the legislation that created GPO Access. Among other things, it contains an authorization of appropriations, in effect calling on Congress to make a serious funding commitment to an online information system that has been successful beyond all projections. While GPO Access is now one of the government’s most popular and heavily used websites—more than 20 million documents are downloaded each month, and nearly 150,000 titles are available—it continues to be hamstrung by the current funding system. We continue to be required to fund this system out of savings achieved through the reduction in printed matter. A real commitment to GPO Access would enable immediate, full implementation of our electronic collection plan. From there we can move on to other objectives designed to simplify and improve public access to government information. I think that, with GPO Access, we have already proven that if we are given the tools, we will do the job. NTIS Following the Commerce Department’s recent announcement that it plans to close NTIS and transfer its archive function to the Library of Congress, we made a number of inquiries about whether this was an appropriate disposition of this function and suggested GPO as a potential alternative. I testified about this at a hearing earlier this week before the House Science Committee’s Subcommittee on Technology. This is not the first time such a consolidation has been proposed. A consolidation of NTIS with GPO was nearly achieved following negotiations with the Commerce Department in the early 1980’s, and later that decade, when the Department was unable to find a home for NTIS under its privatization initiative, GPO offered to take the agency on. NTIS’s 3 million-title collection—and that’s what it is, a working collection and not an "archive"—is a tremendously valuable resource for the public. Its shortcoming has always been that it was not made available to the public through depository libraries, due to the NTIS position that its publications are "cooperative" publications, which necessarily must be sold to be self-sustaining. Consequently, they have been exempt by law from the requirement to be distributed to depositories, and this collection today remains the largest source of fugitive documents in the government. As you know, GPO has a small pilot project with NTIS to make image files from some of these documents available to a limited number of libraries. Under this project, we reluctantly agreed to NTIS’s condition that the files not be made available for reuse or redissemination. A consolidation of NTIS with GPO would make all of the collection available for dissemination to depository libraries, with no restrictions on reuse or redissemination. It would eliminate the fugitive document problem almost entirely, at one stroke. We would fund this dissemination through GPO’s salaries and expenses appropriation for the library program, and we have made it clear to all that this would require an increase in our appropriation. The increase, however, would be fully justified by opening up the NTIS collection for the first time in its 50-year history to public use through the libraries. Transferring NTIS to GPO would also consolidate the government’s two largest publications sales programs. With the Commerce Department’s announcement of the NTIS closing, there has been a lot of discussion about the inappropriateness of the government selling its information products. If that were the only way the government made information available, I would agree. But I think that if the NTIS collection can be made available for the free use of the public through depository libraries, there is nothing wrong with also offering copies for sale for those who want them, at a reasonable price that recovers the marginal cost of reproduction, handling, and postage. There are possibly a number of efficiencies that could result from consolidating the GPO and NTIS sales programs. We would eliminate costly and unnecessary competition, under which we have split revenues for similar products, such as the IRS CD-ROM. GPO’s larger program, which last year sold some 18 million copies, could potentially offer a number of economies of scale to the NTIS operation, which only sells about a million copies annually, possibly lowering NTIS prices. Finally, a consolidation of these operations would help to eliminate public confusion about where to go to obtain government information. This would have tremendous value in making public access simpler and more user-friendly. Permanent Access Beyond these issues, we need to have a system for providing permanent public access to government information. In this age of electronic documents, which are put up and then removed from the web at will, ensuring continuing, permanent access to government information is emerging as a tremendous challenge. The same is true for agencies that are terminated and go out of business. Their information resources need to be made permanently available for active, ongoing use. This is a function that will involve a variety of institutions: government agencies, GPO, the National Archives, and the library community. What we are talking about is not simply creating archives of information, but permanently accessible and available collections, similar to the concept of permanent collections in regional depository libraries. We have already had a great deal of success in establishing partnerships with libraries to house specific agency collections. The government’s role in assuring permanent public access to government information is completely appropriate. The government can’t continually adopt new technology for making current information available to the public while at the same time abdicating its responsibility to provide for the safekeeping and accessibility of information that’s no longer current or "hot." The principle of public access to government information, and its long history in this country at the constitutional, statutory, and administrative levels, provides a rich set of precedents for today’s information disseminators to live up to. As technology continues to change, government rules, as well as the rules and roles of those in the library community, must continue to evolve. But what must stay constant is a continuing adherence to the principle of public access and its centrality to our system of informed self-government.