ARCHIVED: Comments on PubSCIENCE Proposed Shutdown

There is no information service comparable to PubSCIENCE. PubSCIENCE addresses a different niche and audience than do the information services to which it has been compared. It provides a home for smaller publishers and gray literature to be made available to the scientific community and the general public. PubSCIENCE publishes abstracts for all the literature it cites, and the indexing and abstracting service is publicly available It is a unique and important service for the scientific community and the users of our Nation's libraries

For over 50 years, the Department of Energy's (DOE's) Office of Scientific and Technical Information (OSTI) has been collecting, preserving, and disseminating scientific and technical information for DOE. The origin of this information program was the Manhattan Project during World War II. From the beginning the fundamental purpose of this program was to ensure that research results were reported and made available to the agency, to researchers in the physical sciences community, and to the broader scientific community. As described on the "About PubSCIENCE" section of the web site, PubSCIENCE is a natural evolution of OSTI tools dating from the late 1940's, the Nuclear Science Abstracts and the Energy Science and Technology Database.1 DOE/OSTI distributed the Nuclear Science Abstractsto our Nation's federal depository libraries with no fee "in order to ensure maximum public access and dissemination of the results of research and development projects of interest to the federal government program. PubSCIENCE continues that tradition. In essence, PubSCIENCE is a modernization of Nuclear Science Abstracts and the Energy Science and Technology Database."2

OSTI's mission continues to this day to provide access to national and global STI for use by DOE, the scientific research community, academia, U.S. industry, and the public. PubSCIENCE is the culmination of the agency's lifetime tradition of providing scientific and technical information by bringing that information to the desktop. PubSCIENCE was created in 1999 to give physical scientists the capacity to search across the journal literature at no fee in response to the evolving opportunities presented by web-based technologies.

The Federal Government has a responsibility, recognized by the Paperwork Reduction Act and OMB Circular A-130, to disseminate information about its work products, services, and information sources for the public benefit.

Today, most basic research in physics, matter, chemistry, and energy is funded by the federal government. Government agencies such as DOE have an affirmative obligation to the public to make information collected through tax-payer dollars available for public review and use.

Among the federal agency responsibilities detailed in OMB Circular A-130, "Management of Federal Information Resources," (61 FR 6428-6453, 1/20/96) are that agencies:

  • Have a responsibility to provide information to the public consistent with their missions.  
  • PubSCIENCE continues an agency mission that has existed for over 50 years. Should disseminate information in a manner that achieves the best balance between the goals of maximizing the usefulness of the information and minimizing the costs to the government and to the public.
  • PubSCIENCE employs current web technology to create a centralized source of information that is easily accessed by agency researchers, the research community, and the general public.
It is important to note that the use of web technology also achieves other objectives delineated in OMB Circular A-130 - dissemination of information on "equitable and timely terms", "tak[ing] advantage of the fastest growing dissemination channel" (the Internet), and "help[ing] the public locate government information maintained by or for the agency."3

This Administration's E-Government initiatives encourage agencies to use new technologies to meet the Government's affirmative obligation to disseminate its information efficiently and effectively. PubSCIENCE utilizes the most appropriate current technology to disseminate government and government-funded research, together with other related research, in order to ensure that research results are reported and made available to the agency and to the broader scientific community. Thus, the use of web technology is simply one step in the evolution of this service and its effectiveness, and it adheres to clearly stated government objectives Indeed, the Department of Energy in early 2001 recognized the importance of PubSCIENCE as an outstanding resource, saying "The DOE Web Council selected the PubSCIENCE Web site as the featured site on the DOE National Library page for the February 2001 edition of, the Department's new homepage. This work was a valued example of DOE's commitment to build a rich Web site for the American public."4 It is difficult to understand why, with its commitment to broad public access, that DOE is now questioning the legitimacy of this important scientific tool.

There is a need for multiple channels for information and for a diversity of sources. PubSCIENCE provides information to DOE, other agencies, federal contract research and the research community in general.

The reasons given by DOE in its proposal to discontinue PubSCIENCE rely solely on an argument advanced by one segment of the private sector. We believe that the Government should not eliminate a valuable service that serves a real and legitimate public need because two private sector organizations have developed similar but not identical services. This is not an argument supported by any government policy and, indeed, it runs counter to recent government practices such as the Patent and Trademark Office databases Examiners' Automated Search Tool (EAST) and Web Enabled Search Tool (WEST), as well as the Security and Exchange Commission's EDGAR. The information contained in these government databases was collected from the private sector and each agency properly recognized its obligation to serve the public by providing no-fee access to these databases.

In fact, a closer examination of the services in question finds that the similarities between PubSCIENCE and the two data bases cited, Scirus and Infotrieve, are surface resemblances only. There are many very important differences that show beyond question that PubSCIENCE does not directly compete with Scours, Inforieve or other fee-based commercial services.

  • PubSCIENCE, Scirus, and Infotrieve are targeted toward somewhat different audiences; PubSCIENCE's audience is the scientific and research community, and the other resources are being marketed primarily to corporations and other large entities rather than the public;
  • PubSCIENCE is a relatively small resource focusing on the needs of researchers and scientists in energy-related fields; PubSCIENCE has agreements with 35 publishers. Scirus covers all areas of science journals published by a few large publishers and also makes available Web-based resources; Infotrieve also covers many more publishers than does PubSCIENCE but has agreements with only half of the publishers covered by PubSCIENCE.
  • PubSCIENCE contains abstracts and some links to the full text that do not exist in the other two resources.
  • PubSCIENCE has agreements with a number of smaller publishers whose literature is important to scientists and who are not included in the data bases of the two commercial resources. 
  • PubSCIENCE provides access, in addition to the published literature, to government produced information.  
In fact, a very small sample of three searches showed the following results:
  • Item 1: PubSCIENCE has the abstract; Scirus does not contain the citation at all; Infotrieve contains no abstract and charges $15.50 for the article.
  • Item 2: PubSCIENCE has the abstract; Scirus contains a similar abstract; Infotrieve has no abstract and charges $34.50 for the article.
  • Item 3: PubSCIENCE has the abstract and links to the full text on the Web at no charge; Scirus contains the abstract but no link, and charges $30 for the text that is available through PubSCIENCE without charge; and Infotrieve has the citation but no abstract or link, and charges $14.45 for the article.
PubSCIENCE has encouraged "a diversity of public and private sources for information based on government public information" as required by the Paperwork Reduction Act. The closing of PubSCIENCE will in fact diminish the required "diversity" of resources, leaving electronic access to these information resources only via commercial information providers.5
The discontinuation of PubSCIENCE will disable the indexing of a portion of the literature of the sciences (energy, matter, physics, etc.) relevant to private, not-for-profit, and public sector markets.

Indeed, a strong counterargument can be made that PubSCIENCE provides a critical service to other segments of the public and private sectors. For example, BioOne (a collaboration between scientific societies, academia and the commercial sector) has indicated that PubSCIENCE provides a valuable service to BioOne's 49 publishing partners, all of whom are not-for-profit organizations, by offering increased visibility for these journals on a very widely used site. This is of particular importance to the small to mid-sized publishers represented by BioOne, none of whom have titles that appear in ScienceDirect, nor are they covered by the Scirus search engine. PubSCIENCE provides these publishers with an effective, no-fee public awareness mechanism that helps these journals to remain competitive in their disciplines.

PubSCIENCE is a partnership with 35 publishers which provides information about resources on the publishers' servers. This voluntary partnership has enabled these publishers to provide site licenses to DOE and the federal scientific community. PubSCIENCE itself does not enter into commerce but has, in fact, enhanced commerce for these publishers of scientific literature. Moreover, PubSCIENCE does not discriminate by size of publisher, type of publication, or topic (within the range of its mission). Scientists applaud the fact that their research needs are taken seriously by PubSCIENCE in defining its relationships with publishers who can serve those needs. One reviewer has noted that "the contents of PubSCIENCE have been largely driven by the direct needs of the researchers of the DOE and the agreements reached with publishers."6 PubSCIENCE claims to include over 1300 peer-reviewed journal titles including forthcoming titles. Each journal is fully indexed, regardless of the subject matter .

PubSCIENCE is valuable to progress in the scientific and commercial sectors, including private, not-for-profit and public communities.

Implicit in DOE's stated reason for discontinuing this service -- that two similar services exist - is the idea that somehow the existence of PubSCIENCE creates a disadvantage for these other providers. But the reviewer cited above also noted that

Since there is little unique coverage in PubSCIENCE (approximately 90% of the journals are indexed in at least one other major research database), librarians and clientele of research libraries with access to databases such as Web of Science, Compendex, Inspec, Embase, or Chemical Abstracts may only find this resource useful if the research topic is energy-related.

On the other hand, university and college librarians, and possibly even public librarians, with restricted budgets will find that PubSCIENCE will supplement their collections quite well by providing access to peer-reviewed research and technical literature that may not be otherwise readily accessible. A title by title comparison with the Applied Science & Technology Abstracts, Biological and Agricultural Abstracts and General Science Index revealed that only 10% of the PubSCIENCE journals were also indexed in the Wilson databases. If the journal title lists from Chemical Abstracts Student Edition and BasicBIOSIS are included in this comparison, the percentage only increases to 16%. The strengths of PubSCIENCE lie in areas where the costs of databases are often prohibitive, e.g., chemistry and physics. This database is an excellent resource for smaller academic libraries where needs cannot often justify expenditure.7

PubSCIENCE is valuable to the scientific community and important for continued scientific inquiry, as well as for the continued competitiveness of our Nation. Eliminating PubSCIENCE will have a negative impact on scientific research, as it will impede the sharing of information in the scientific community.

The vitality of the research enterprise and therefore this country's scientific and commercial progress depends on easy access by scientists to the research results of others. The advancement of knowledge proceeds by trial and error and by the insight gleaned from review of the efforts of colleagues. Tools such as PubSCIENCE, Scirus, Infotrieve, and many fee-based commercial services that provide access to scientific information for which our Nation's libraries spend hundreds of millions of dollars each year combine to facilitate the process.

Over the last decade, the average annual increase in the price of science journals has risen significantly -- over 11% a year, with the price of a science journal in 2001 averaging almost $1,050. The result of these high prices is that fewer and fewer scientists have ready access to the scientific literature as libraries are forced to cancel subscriptions in an effort to balance their budgets, which rarely see increases even as high as 5%.

High prices are not the only issue facing the scientific community. Large commercial companies dominate scientific publishing and are acquiring additional content through acquisitions of individual journal titles and through mergers with other major publishers. As a result of these transactions, more and more of the research produced by scientists is under the control of fewer and fewer companies that limit access to that content via restrictive licensing terms and conditions.

Interestingly, these companies are not just purchasing other publishing companies. Over the past several years, publishers have purchased software and technology delivery systems. For example, Wolters Kluwer acquired Ovid Technologies (a platform for the delivery of electronic journals and databases) and Elsevier Science purchased Endeavor Information Systems (one of the leading vendors of library systems software). While this vertical consolidation of large content providers with dissemination technologies could result in more integrated information environments, it could also lead to the emergence of technologies that would favor some content or standards over others. As a consequence, for-profit providers can limit those journals indexed and thereby hinder access to some STI literature.

As noted previously, the key role of PubSCIENCE is the provision of no-fee access to federally-funded STI information resources to the scientific and research communities. If the Department does not continue to provide such services, the research community and the public have no assurance that no-fee access to federally funded research will continue. Infotrieve and Scirus are not comparable services to PubSCIENCE. Rather they are commercial services that are under no obligation to continue to provide no-fee services to the research community and the public.

Access to and preservation of information for future use by the scientific and research community is critically important. PubSCIENCE provides an important, unique service to these communities and should continue.

There is no question that all parts of our information structure - public and private - are essential to ensuring Jefferson's ideal of an educated populace and the Nation's desire to maintain its competitive position in the worlds of science and industry. Indeed, the President's Council of Advisors on Science and Technology (PCAST) has drafted a letter to the President noting concern with the level of federal investment in science and technology. The draft letter calls for improving "funding levels for physical sciences and certain areas of engineering." (see The Federal Government must maintain resources such as PubSCIENCE, which excels in providing a no-fee and robust access and delivery system to the scientific and research communities of this Nation.

Comments Submitted to the U.S. Department of Energy on September 7, 2002 by,

Emily Sheketoff
Executive Director
American Library Association Washington Office
(in collaboration with the Association of College and Research Libraries and the Government Documents Round Table)

Mary Alice Baish
Associate Washington Affairs Representative
American Association of Law Libraries

Prudence S. Adler
Associate Executive Director, Federal Relations & Information Policy
Association of Research Libraries

Mary Langman
Manager, Information Issues and Policy
Medical Library Association

Steven Aftergood
Project Director
Federation of American Scientists

Danielle Brian
Executive Director
Project on Government Oversight

Sean Moulton
Senior Policy Analyst
OMB Watch

Ted Smith
Silicon Valley Toxics Coalition/Campaign for Responsible Technology

Michael Sprinker
CIH, Director
International Chemical Workers Union Council/UFCW

  1. Accessed August 29, 2002.  
  2. Accessed August 29, 2002.  
  3. U.S. Office of Management and Budget. Circular A-130 Revised: Memorandum for heads of executive departments and establishments. Subject: Management of Federal Information Resources. Washington, DC: OMB, 1996.  
  4. Department of Energy,  
  5. U.S. Congress. Senate. Paperwork Reduction Act of 1995 [S.244]. Washington, DC: Government Printing Office, 1995.
  6. Maddux, Linda. Review of PubSCIENCE. College and Research Libraries News 62 (July/August 2001) 738-39.
  7. Ibid.
  8. pdf, accessed August 30, 2002.