On August 12, 1999, the U.S. Department of Commerce announced that Secretary William Daley would send to Congress legislation to close down the National Technical Information Service (NTIS). According to the departmental press release: "After extensive review and analysis it was determined that the core function of NTIS, providing government information for a fee, is no longer needed in this day of advanced electronic technology. Established in 1950, NTIS' core business - the sale of government documents in microfiche and on paper - is rapidly becoming less of the necessity it was as agencies and groups have begun to post their reports on the Internet for free."

In testimony before the House Science Committee, Subcommittee on Technology, Deputy Secretary of Commerce, Robert Mallett, stated on September 14, 1999, the case for closing NTIS.

NTIS was created in 1950 to operate as a clearinghouse within the U.S. Government for the collection and dissemination of technical, scientific, and engineering information of all kinds. However, the rapid growth of the Internet has fundamentally changed the way NTIS' customers acquire and use information. As the Department's Inspector General noted in March 1999, "Federal agencies are increasingly bypassing NTIS as a distribution channel, instead offering their publications directly to the public over the Internet."

It is not surprising then that -- largely because of these changes in the marketplace -- sales of publications from the traditional NTIS clearinghouse declined from almost 2.3 million units if Fiscal Year 1993 to 1.3 million units in Fiscal Year 1998. As a result, in the past two years, NTIS has lost millions of dollars, using nearly all of its retained earnings to stay afloat.

Mallett described recent efforts to keep NTIS viable by aggressively downsizing its workforce, resulting in the transfer of 46 staff members to other parts of the department, reducing office and warehouse space, and other cost cutting measures. He noted that NTIS had expanded its mission scope in recent years, venturing into business opportunities "on the perimeter of its statutory mission" and possibly competing against private sector businesses. In sum, the Commerce Department had decided to close down NTIS because its business model had become fatally flawed and the agency was a financial liability to the department.

As anyone who has followed NTIS is well aware, NTIS in recent years had transformed itself into a kind of general purpose service bureau for all agencies, willing to perform virtually any information service for which agencies and/or the public would pay. Today the NTIS repository contains almost three million technical reports, databases, and other publications. The agency makes its products available to the public in paper, microfiche, diskette, audiovisual, CD-ROM, and online formats. Sixty thousand customers receive over one million products a year from NTIS. NTIS operates FedWorld (URL: http://www.fedworld.gov), a kind of cyber-clearinghouse for federal government information on the Internet. The site began as an electronic bulletin board (EBB) service that enabled information seekers to dial into a single EBB and be routed to dozens of other federal EBBs. When the Internet emerged, FedWorld led the way for federal agencies in pioneering new information services on the Web. The FedWorld service has grown into a general purpose information utility not only for federal information but also for data from states and foreign countries. NTIS serves as web site host for a significant number of agencies, notably the Internal Revenue Service. The IRS web site is one of the most active in the nation, logging 30 million hits on April 15, 1999, alone. NTIS also operates a 24-hour-a-day fax-back system for paper copies of IRS forms and other documents.

By law, NTIS supports itself with revenues from sales of its products and services, and therein lies the key to its problems. The agency receives no congressionally appropriated funds. As sales from its traditional scientific, technical and engineering information (STEI) products declined, NTIS expanded its reach into all kinds of government information in an effort to remain institutionally viable as well as to fill gaps in the spectrum of federal information services. In the past decade, the agency's original concept began to fray as agencies developed their own robust publishing programs and lagged in sending materials to NTIS. The American Technology Preeminence Act of 1991 attempted a supply-side remedy by mandating that all STEI publications be sent to NTIS. But the law lacked enforcement teeth and compliance has been spotty. Although NTIS adds more than 50,000 new titles each year, some major STEI sources are not reaching its repository; some of the National Institutes of Health, for example, do not supply their publications to NTIS.

Demand has also fallen as the Internet revolution sweeps government, more agencies make their information holdings available free of charge, and NTIS' prices are perceived as falling on the high side. The department's August 12 press release cited the example of a recent Commerce report, The Emerging Digital Economy II, which could be downloaded from the departmental web site for free or purchased from NTIS for $27.

The result appears to have been a losing struggle for NTIS to remain self-supporting. At the very least, NTIS had lost the confidence of its parent department. Commerce's Inspector General informed Congress in early 1999 that the department doubted NTIS's ability to generate sufficient revenues to remain self-supporting. The loss of confidence was doubtless hastened by recent public debacles involving NTIS. In 1996, NTIS undertook a joint project with the Internal Revenue Service known as Cyberfile. The basic idea was that NTIS would serve as IRS's agent for receipt of electronic tax filings. Cyberfile was a well publicized failure. Blame fell on both agencies and NTIS was castigated for lacking the necessary in-house technical and management expertise to carry out Cyberfile. In 1999, NTIS again drew adverse media attention over an announced alliance with an Internet search service, Northern Lights, under an arrangement that would have charged FedWorld users for advanced search functions. Whether justified or not, the public perception arose that NTIS was charging for government information that ought to be free. The adverse media coverage embarrassed the Commerce Department and could have served as a motivating trigger for the move to eliminate NTIS.

The Other Side of the Coin
The Commerce Department decision to seek abolition of NTIS seems clearly to have been motivated by financial considerations to the exclusion of any other rationale. No serious thought was given to the valuable functions NTIS performs for the federal government and the public, and what would become of those functions if NTIS disappears. The department's reasoning appears to have been: in following its current business model NTIS has become a financial liability; ergo, close NTIS. Even the most casual observer can see that Commerce failed to explore another line of reasoning: NTIS has become a financial liability; ergo, seek a redefined mission based on a new business model that will ensure NTIS' financial security. Commerce chose not to explore the possibility that NTIS performs governmental functions so valuable that the department should seek congressional appropriations to support the functions.

The departmental announcement of Secretary Daley's decision stated that Commerce would give the NTIS collection to the Library of Congress, yet the House Science Committee hearing showed that Commerce had not consulted with the Library in advance of the announcement. Nor, testified Mallett, had the department consulted with users of NTIS' services. Contrary to the Deputy Secretary's testimony, Commerce did not undertake a serious and comprehensive review of its potential options. If it had, the department would have recognized that the Library of Congress could not possibly accession the NTIS collection without itself receiving millions of dollars in appropriations.

The deficiencies in Commerce's decision making processes were dramatized in the statement the Library of Congress submitted at the House hearing. Librarian James Billington asked that Congress consider "the full spectrum of operations involved in collecting, organizing, and supplying scientific and technical information."

If NTIS cannot continue in its present form, the federal government must examine which of its functions are sufficiently desirable and effective to merit continued federal support, and how and where such functions can best be sustained to ensure the uninterrupted acquisition, accessibility, and preservation of scientific and technical information. Each function now carried out by NTIS should be assessed as to whether the function is needed, whether it should be centralized (or centrally coordinated); whether multiple access points, public and private, should be encouraged; and whether the function should be self-sustaining or publicly supported in order to serve a larger public good. Only then can a responsible conclusion be reached as to which agency or agencies are best positioned to ensure future access to information in all formats, even to some low-demand but potentially important products.

NTIS occupies a stable position among the set of federal STEI agencies, which includes the National Library of Medicine, the National Aeronautics and Space Administration, the Defense Technical Information Center (DTIC), and many others. NTIS carries out valuable services both for STEI and non-STEI agencies. For example, DTIC releases 15,000 to 18,000 titles annually to the public through NTIS. All distribution and order fulfillment are handled by NTIS. DTIC is not equipped to discharge these tasks for itself, and the same is true for many other agencies. For many agencies, NTIS serves as a service bureau and/or fulfillment house. The National Center for Health Statistics, for example, has for decades distributed its electronic databases through NTIS, and has no desire to take the distribution in-house. NTIS is one element in the federal STEI system, and however imperfect that system may be, the effects of removing that element will ripple throughout the system with potentially deleterious effects to the public good. Removing a financial liability from Commerce could cause even greater liabilities for other agencies.

When accessioning titles into its collection, NTIS performs abstracting and indexing services on the titles. The agency recoups the cost of these bibliographic services through its product sales. In many cases, no one else performs these bibliographic controls over the government publications in question. The NTIS catalog database is thus a valuable product in itself. Private firms such as Dialog acquire the database and resell it to the public; the firms have nowhere else to turn for the database. Perhaps NTIS competes with some private firms, but for others it provides irreplaceable services. Perhaps some NTIS activities can be out-sourced if NTIS closes down, but many functions may simply cease if NTIS goes away, most especially the central clearinghouse function.

Conclusion
The Commerce Department espouses the belief that federal agencies should post their publications to the Internet for free distribution, rather than have to pay the fees NTIS charges. This naïve view ignores the fact that millions of users adamantly prefer printed copy, particularly when the document in question runs to several hundred pages. The success of Amazon.com, which uses the Internet primarily to sell print products, is only one of many witnesses to the continued vitality of print publications in the Internet age. Another witness is the fact that university presses are now posting full-text copies of books on the Internet and finding that this tactic actually increases print sales. The public wants information products in print and they will continue to want them in this form, notwithstanding wishful thinking at the Commerce Department.

Every net denizen knows that things disappear from web sites far more capriciously than they disappear from print. This problem Commerce would remedy with a legislative proposal requiring agencies to maintain STEI publications on their web sites for three years. It is doubtful that Commerce could get the proposal passed into law in the face of surefire opposition from other executive branch agencies such as the Departments of Defense and Health and Human Services. Even if it were enacted into statute, it is doubtful that such a law would have any more success in making an impact on agency behavior than the American Technology Preeminence Act did earlier in the decade. We already have laws requiring agencies to send STEI publications to NTIS and to send all publications to the depository libraries. Many large agencies observe these laws only in the breach; their behavior will not change just because Commerce wants off the hook for NTIS.

Perhaps only the Government Printing Office applauds the closing of NTIS, seeing an opportunity to eliminate a bitter rival and to incorporate into the depository library program the assets of NTIS. In all the brouhaha thus far, the White House and the Office of Management and Budget have been conspicuously silent. Commerce appears to be going it alone in the effort to eliminate NTIS and does not seem to have done its homework well.

Arrayed against Commerce are the STEI agencies who see the importance of NTIS in the STEI chain of production and distribution. The library community has also hammered Commerce, noting in congressional testimony that the proposal to close NTIS touches upon all aspects of federal information policy. The library community's central points were:

• First, NTIS should not be closed, nor its services transferred, until there is a thorough assessment of the full range of NTIS services, alternatives for providing each service, and the requirement that the program be self-supporting.
• Second, NTIS provides unique centralized services that are critically important to the ability of the public to locate and have access to the government's STI resources, including the tangible collection and current agency web-based publications.
• And third, technology has not yet solved two key challenges in moving towards greater dissemination of STI reports through the Internet: those challenges are centralized bibliographic access and permanent public access.

Will the Commerce Department succeed in closing NTIS? Personally, I doubt it very much. It is never easy to close down a government agency. In part, this is true because people's jobs are at stake, and members of Congress from the districts most affected by NTIS job losses were quick to condemn Commerce's decision. In part, it is true because of the discovery that agency's functions are valuable and important to preserve, and that is what has been happening to NTIS since the Commerce announcement. In all likelihood, Congress and the administration will end up providing appropriations for the continued functioning of NTIS.