Background, Previous Disscussion
At the first NCLIS meeting on NTIS (January 19) discussion focused on the importance of having the Government continue the NTIS core functions and the possible organizational location of NTIS. It became clear that the notion of NTIS as a fully self-sustaining entity - although possible in periods of heavy report input and high report sales - is not viable as an ongoing operating approach for NTIS and not consistent with basic Government user charge policy. It was generally agreed that the core functions of NTIS would have to be continued and that appropriated funds should be made available for the funding of certain "public good" core functions. These conclusions were independent of the ultimate organizational location of NTIS. Therefore, moving NTIS organizationally does not eliminate the need for subsidizing core functions with appropriations but simply shifts the agency requesting the appropriation, possibly from the Executive to the Legislative branch.

The discussion at the last meeting touched on, but did not dwell on, the need to focus on the different stages of dealing with the NTIS issues. The NTIS issues need to be addressed in terms of the immediate short-term problem, the intermediate-term after the immediate issues are resolved and Congress has acted, and the longer-term where the new internet and digital technology will change the nature of the present NTIS operations. Discussion at the meeting touched briefly on the short-term and intermediate-term issues but did not deal with the longer-term issues, which will be raised by the adoption of the newer technologies. This statement focuses on these longer-term issues.

Short-Term
At the last NCLIS meeting Kent Smith forcefully and eloquently addressed the short-term issues. Action urgently needs to be taken to insure that NTIS does not decline - lose quality staff, lose additional report input from agencies, or lose existing business - before action is taken by the Congress on the Department of Commerce proposal. During this period the Department of Commerce may be inclined to make further cuts in NTIS resources to further reduce the risk of loss. This action by Commerce, or NTIS management pursuing what they believe to be their mandate from Commerce, could severely limit opportunities to continue the core functions of NTIS in any organizational environment as a viable entity.

Intermediate-Term
The 'public good" functions of NTIS that should be funded by appropriated funds wherever the NTIS function is located include: acquisition of reports, processing of bibliographic and indexing information, initial processing of document content for distribution and maintenance of the document archive. The remaining NTIS core functions: announcement distribution, any promotion or marketing, document distribution and order fulfillment should be funded totally from user charges for the services provided. Thus, in the intermediate-term all functions not funded by the appropriation would be variable costs, which would rise and fall with changes in user demand. With proper management this should eliminate future funding problems for NTIS. This in turn should reduce somewhat the entrepreneurial zeal of NTIS management and thereby reduce both potential public sector-private sector concerns and extensive marginally related NTIS service work for other agencies. The result would be a financially stable NTIS focused on its core functions.

How much would the appropriation have to be to cover the core processing tasks? In the NCLIS meeting on the 19th we discussed costs on the order of $100 per report or $5 million per year. This is a somewhat simplistic estimate since the reports already processed by the larger agency information systems (DOD, NASA and DOE) would require less NTIS processing and the reports from smaller agencies would require more. NTIS should be asked to provide estimates of these costs to formulate an estimate of the appropriation required and how it might change over time.

Long-Term
In the long term, the internet and digital technologies will change the economics of information distribution. We can envision an NTIS in which report content comes from major agencies in digital form and where content from smaller agencies is converted from print to electronic form by NTIS using appropriated funds. The internet could be used for most of the report distribution on a no-charge basis to US users. Since appropriated funds would pay for input processing, including mounting and maintaining a master web-site, the incremental costs of providing a single report to a user would be negligible and the service could be provided to the public without charge. NTIS would continue to charge for older reports where conversion and storage costs would be higher, for those users who prefer paper or microfiche to online access and for access from foreign users. The result would be an NTIS with: (1) wider distribution of current technical information at no substantial incremental cost, (2) a small distribution program for older reports and non-online formats based on user charges which fully recover incremental costs and (3) a fixed, modest, appropriation covering the core processing and maintenance costs.

Blanket conversion of the non-digital back file at an estimated one-time cost of $7 million is likely to be a questionable investment. The older material is less in demand and that demand is likely to be concentrated among relatively few titles. Digitizing perhaps one or two years of back file and then identifying older best sellers to digitize selectively would probably meet the need more cost effectively. Fulfillment of the older non-digitized content would be in print or microfiche on a user charge basis or, if the user preferred, even digitally on a user charge basis.

Bibliographic information and basic indexes would be available on the NTIS web site for free use. Since the content of the indexes would not be copyrighted, private sector intermediaries could integrate this information with their own search engines and finding tools. This would result in wider distribution of the information at no cost (or income) to NTIS. Full text search of the digital content on the NTIS web site might be a user charge service since it could involve more intensive use of computer resources.

Some of the larger contributing agencies (DOD, NASA and DOE) may provide their own users with access to their reports on their own agency servers. These agencies might find it cost effective to let the NTIS public user access the report content on the agency's server via the NTIS web site rather than replicate the content on the NTIS server. However, unlike the Department of Commerce proposal, NTIS would have the continuing responsibility for the maintenance of public access. Thus, if and when the report is removed from an individual agency's server or if public access to the agency's server is no longer permitted, NTIS would make the report available on its own server. Continued public availability would be assured.

The new internet dissemination would also largely eliminate the depository library fugitive document issue without additional appropriations. Depository library users could access the entire digitized NTIS collection as well as the index information on a depository library terminal connected to the NTIS web site without an NTIS charge.

Conclusions
The appropriation to fund NTIS public good operations is likely to be on the order of $5 million per year although NTIS staff should make a more careful assessment of this figure. Contrasted with an investment of $80 billion per year in Federal R&D, an annual investment of 0.006% of the R&D dollars to keep the taxpayer funded research results accessible to the public seems rather modest. Organizational location should be decided on some more substantive basis than which agency should make the appropriation request. The likely future opportunities for wider low cost distribution of research results using the internet would be seriously jeopardized by the Department of Commerce proposal which would eliminate an ongoing central index, central bibliographic control and the assurance of permanent public availability.

NCLIS should formulate recommendations to the Congress that reject the Department of Commerce proposal to close NTIS and transfer its report archive to the Library of Congress. NTIS should remain in the Department of Commerce, with sister agencies such as PTO, NOAA and Census, which have strong scientific and technical information dissemination responsibilities in their missions and the general Commerce mission of assisting US businesses. The Department of Commerce should request a limited ongoing appropriation to fund the central NTIS processing functions on an ongoing basis.

PFU 1/21/99