Making government services more accessible and accountable through the use of information technology was the dominant theme of the Federal Library and Information Center Committee's (FLICC) 2004 Technology Update, "The E-Government Act of 2002: A Progress Report," held on Sept. 1. More than 60 federal librarians and information professionals attended the all-day program at the Library of Congress in which a series of speakers described provisions of the 2002 E-Government Act and summarized the recommendations and continuing work of the Interagency Committee on Government Information (ICGI) established by the act.
Setting the stage was Patrice McDermott, deputy director of the American Library Association's Office of Government Relations. Stressing the important role that input from libraries played in the formulation of the law, she singled out requirements in three key areas: making government information more accessible both in the near and long term; "leveling the playing field," including gauging the public impact of moving to e-government services; and providing a more accountable government by helping people obtain information held by multiple agencies in an integrated fashion. McDermott concluded by urging the audience to "stay tuned." When all the recommendations are in, she asked: "Will the Office of Management and Budget ensure the implementation of the e-government policies? Will Congress do useful oversight? Will the public be consulted?"
Following McDermott's talk, Eliot Christian of the U.S. Geological Survey and architect of the Global Information Locator Service (GILS), reviewed the mission and activities of the ICGI, which has set up three work groups to carry out its charge—Categorization of Information, Electronic Records Policy and Web Content Management. Christian, as chair of the Categorization of Information work group, then introduced a series of speakers who highlighted topics considered by that group—the definition of government information, searchable identifiers, categorization and interoperable search—topics that are further addressed in two draft documents to be finalized and submitted to the Office of Management and Budget (OMB) by Dec. 17. They are "Requirements for Enabling the Identification, Categorization and Consistent Retrieval of Government Information" and "Recommendations for Search Interoperability."
T.C. Evans, deputy Superintendent of Documents, walked attendees through the topic "Defining What Government Information Is to Be Categorized." Emphasizing that a clear definition is critical "to enable citizens to obtain a predictable body of search results of similar granularity across varying communities of practice," Evans presented the following definition of government information:
"Any information product regardless of form or format that a U.S. federal agency discloses, publishes, disseminates or makes available to the public, as well as information produced for administrative or operational purposes that is of public interest or educational value."
Included is information created or exchanged within or between agencies or that may be subject to Freedom of Information Act requests. Not included is information that is restricted because of security or privacy concerns; information about the government such as television newscasts, or objects owned by or loaned to the government.
Moving beyond a definition of government information, James Erwin, director of Information Science and Technology at the Defense Technical Information Center, discussed the importance of searchable identifiers in retrieving that information, addressing the questions, "What are searchable identifiers?" "What are their requirements?" and "How can they be used to improve access to government information?" Erwin described searchable identifiers as uniquely identifying an information object and supporting persistent access both to the object and to information about the object. They must generate globally unique object names; be distributed, open, extensible and scalable; must support both tangible and intangible information objects; must work with existing identifiers such as the ISSN; and must be human readable. At the same time, they must be easy to use, support multiple interfaces, be compatible with existing and emerging standards, and be able to transition to future technologies in order to provide long-term persistent access.
In addition to persistent identifiers, consistent categorization of government information is key to providing access to that information, according to Richard Huffine, manager of the National Library Network of the U.S. Environmental Protection Agency. Such categorization must allow for multiple access points, provide context for the information provided and make it possible to select from large amounts of information and identify authoritative sources. The work group has made several specific recommendations, including developing categories, such as title and creator, based on the Dublin Core Metadata Initiative (www.dublincore.org), combined with the establishment of authority controls for key fields in order to promote consistency. Huffine noted that the work group is still grappling with a number of issues, including the granularity of information that should be categorized, ways to categorize older materials and a feasible time line for implementation.
Christian rounded out the morning session by presenting "The Case for Standards-based Interoperable Search." Noting that interoperable search is like plumbing—"when it is done right, you don't see it"—he observed that although searchers may not be aware of interoperability, it allows them to search across different systems despite system differences. He described the current situation as one of "stovepipes" of information provided by federal, state and local agencies, running on hardware and software from many different vendors. In general, each of these "stovepipes" needs to be searched separately. As a result, searchers run the risk of getting only partial, questionable results; analysts miss opportunities for synthesis; and government agencies miss opportunities to leverage each others' information sources. Theoretically, searching across all these at once might be done by gathering all the data into a single source run by one vendor. Christian noted that while this solution is conceptually simple, it is "unworkable on a practical level" and "unacceptable on a public policy level." On the other hand, using a standards-based interoperable search system, any agency can use any search technology that supports the standard.
The morning concluded with a lively question-and-answer session led by James King, chief librarian of the Naval Research Laboratory library. Among the issues raised were the need to include information produced for the government as well as by the government in the definition of government information, the potential archival role of agencies such as the Government Printing Office and the Library of Congress and the need to "push" agencies to treat e-records as official records.
Michael Kurtz, assistant archivist of the National Archives and Records Administration, opened the afternoon session by summarizing the work of the Electronic Records Policy work group. Included in its charge is the identification of common characteristics of temporary and permanent e-records and barriers to effective management of these records, as well as the development of tools and other recommendations for short- and long-term e-record creation and management. Summarizing the group's progress in each of these areas, Kurtz noted that it is on a fast track, with draft recommendations due to the ICGI by Sept. 30. The group has proposed eight metadata elements to describe e-government records and identified several crucial barriers to their management, including the failure of agencies to view e-records as critical assets related to their missions. This failure has resulted in poor implementation with little integration between records management and information technology activities. To address these issues, the group is working on an Electronic Records Management toolkit, which, if approved, might be available by the second quarter of 2005 and include tools in use or in development offering best practices, guidance, process models, lessons learned, training programs, tips and techniques and policies.
Concluding the forum was a presentation on the Web Content Management work group by Bev Godwin, director, and Sheila Campbell, senior content manager, of Firstgov. Godwin elicited chuckles from attendees when she displayed a screen labeled "The Las Vegas Effect" filled with more than 25 "required" links or logos previously mandated by various pieces of legislation or agency directives. Noting that the group's goal is to make U.S. government Web sites "the most citizen-centric and user-friendly in the world," she introduced Campbell, who provided details on seven policy recommendations in the ICGI's Recommended Policies and Guidelines for Federal Public Web sites now under consideration by OMB and expected to form the basis for guidelines to be issued by December. Included are recommendations to: ensure the authenticity, branding and timeliness of federal government public Web sites; present information written and organized from the audience's point of view; display information in a way that is easy to access and use; simplify and unify information across the government; require agencies to establish priorities and a schedule for posting content on their public Web sites; ensure agencies continue to comply with existing federal laws and regulations; and develop a continuing structure to guide Web content policies and requirements into the future.
