2
Special Considerations in E-Government: Why Government Leads in Demand for Certain Information Technologies

Although government can generally build on the technologies and services emerging in the e-business marketplace, special requirements arise in certain areas, reflecting the nature of the government’s operating environment. This chapter briefly discusses several areas where government leads in demand for technologies used for e-government services.

UBIQUITY

Governments must provide services to all citizens. Unlike the private sector, they cannot opt to serve only the more profitable customers, nor can they select particular market segments in which to participate (though governments obviously do target specific populations to achieve their policy objectives). Nor can citizens choose which government they wish to deal with.

In the context of e-government, “services” should not be taken to imply only interactions resembling commercial transactions. In government, important services may involve provision of information, for example, or interactions with elected representatives, or commentary from citizens, businesses, and organizations on proposed regulations. Service is provided today primarily through in-person, telephone, and postal mail interactions; but e-mail, the Web, kiosks, and other forms of IT are providing new opportunities to expand the accessibility of services and make



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 45
Information Technology Research, Innovation, and E-Government 2 Special Considerations in E-Government: Why Government Leads in Demand for Certain Information Technologies Although government can generally build on the technologies and services emerging in the e-business marketplace, special requirements arise in certain areas, reflecting the nature of the government’s operating environment. This chapter briefly discusses several areas where government leads in demand for technologies used for e-government services. UBIQUITY Governments must provide services to all citizens. Unlike the private sector, they cannot opt to serve only the more profitable customers, nor can they select particular market segments in which to participate (though governments obviously do target specific populations to achieve their policy objectives). Nor can citizens choose which government they wish to deal with. In the context of e-government, “services” should not be taken to imply only interactions resembling commercial transactions. In government, important services may involve provision of information, for example, or interactions with elected representatives, or commentary from citizens, businesses, and organizations on proposed regulations. Service is provided today primarily through in-person, telephone, and postal mail interactions; but e-mail, the Web, kiosks, and other forms of IT are providing new opportunities to expand the accessibility of services and make

OCR for page 45
Information Technology Research, Innovation, and E-Government them available ubiquitously. Ubiquity has three dimensions: access for everyone, access everywhere, and access anytime. Access for Everyone The challenge of providing ubiquitous access to all citizens has several elements. First, it involves overcoming geographical and financial barriers. Much of the excitement associated with e-government is akin to that associated with e-commerce—online access from the home means not having to trek to a government office or find the time to place a telephone call during normal business hours (often only to be put on hold). But home access is constrained by such factors as the affordability of computer equipment and whether Internet or other telecommunications services are available and affordable in a given area. While the cost of computers has dropped significantly over the years and the penetration of computers and Internet access has reached roughly one-half of U.S. households, the hardware cost is still greater than that of many other consumer appliances; thus penetration lags significantly behind that of telephone service (which is nearly ubiquitous). Similarly, although dialup Internet access is available from a number of providers in most parts of the country through a local call, there are areas where dial-up access is either not available or only available from a single provider.1 Helping provide for those without home access, a growing number of libraries and other community institutions offer public Internet access, supported in part by several federal programs.2 Residential broadband Internet service provides much higher data rates that support faster download times and the use of richer multimedia content, but its availability is spotty. As this availability continues to grow, there will be concerns about whether those limited to dial-up access will be at a disadvantage similar to that experienced earlier by people without any form of Internet access. Even if affordable and available, today’s technologies and services present 1   Thomas A. Downes and Shane Greenstein. 1998. “Do Commercial ISPs Provide Universal Access?” Available online at <http://www.kellogg.nwu.edu/faculty/greenstein/images/htm/Research/tprcbook.pdf>. 2   The e-rate program, established under section 254 of the Telecommunications Act of 1996, uses a levy on interstate and international telecommunications providers to support discounts on eligible institutions’ purchases of telecommunications and Internet services plus internal networking, with discounts varying with location (e.g., high-cost, low-income). Community-based Internet access is also supported by the U.S. Department of Housing and Urban Development’s Neighborhood Networks program and the U.S. Department of Education’s Community Technology Center grants.

OCR for page 45
Information Technology Research, Innovation, and E-Government significant usability barriers for those less adept with computers and the Internet. The frustrations of trying to use present generations of systems and software are widely acknowledged to have the effect of discouraging use, especially among less-computer-literate segments of the population. Therefore, at least for the foreseeable future, many people are likely to opt not to adopt these technologies, though they still want access to government services. At a minimum, these gaps suggest that there will of necessity be a long transition period during which new IT-based services will complement but not replace more traditional forms of interaction with government. With respect to disability or language barriers, computerized services involve both advantages and disadvantages. They present additional barriers to those with vision or motor impairments, given the high degree of dependence on visual communication and the fine motor skills that characterize today’s keyboard and mouse-interaction paradigm. But the technologies may also provide the most effective tools to enhance access: online interactions can substitute for visits to a government office, thereby facilitating access to information and services for those with mobility impairments. Where adequate provisions have been made, it is possible to install modifications or use adaptive software (e.g., text-to-speech or voice recognition) to compensate for disabilities. Recent operating-systems releases and Web software standards contain provisions for this sort of access (such as hooks that allow screen readers or alternative input devices to work with them), and the World Wide Web Consortium’s Web Accessibility Initiative has developed both technology and usability guidelines. To maximize the extent to which such techniques are used, legislation was enacted in 1998 that mandates improved access both for federal employees and the public.3 In addition, rules developed by the Electronic and Information Technology Access Advisory Committee are in the process of being implemented by agencies.4 When implemented, they will ease, though not eliminate, access issues. Ubiquity also implies that important information be presented to citizens in a useful and meaningful manner. Simply offering access to government documents is not very effective without some means of conve- 3   Section 508 of the Workforce Investment Act of 1998, P.L. 105-220, which amended Section 504 of the Rehabilitation Act, 29 U.S.C. § 794d. The Americans with Disabilities Act of 1990 extended similar provisions beyond the federal government. 4   Government bodies working on accessibility include the General Services Administration’s Center for IT Accommodation and the U.S. Department of Defense’s Computer/Electronic Accommodations Program.

OCR for page 45
Information Technology Research, Innovation, and E-Government niently searching them and cataloging the results. Similarly, providing direct and unstructured access to the wealth of statistical data collected by the federal government does not necessarily constitute meaningful access. For example, one concern of the federal statistics community is that a low level of numerical/statistical literacy in the general population makes it especially important to provide users with interfaces that provide the necessary context and make it easier to interpret the information. It is certainly not feasible (nor affordable) for government to provide all of its information in a form that is understandable to everyone. But for key information and services, there will be expectations that segments of the population not be disenfranchised. One challenge in building systems that provide universal access is doing so without forcing everyone into a lowest-common-denominator situation. Access Everywhere, Anytime “Ubiquity” has another meaning in the information technology arena that is also relevant to e-government: it signifies software and infrastructure capabilities that enable individuals to connect and communicate in more natural ways. The vision, to date unrealized, is that applications from different organizations—including government agencies and offices—would work smoothly together and that users would be provided with effectively seamless access to information and services across various access devices and communications links. Moreover, users would be able to obtain information and services from wherever they are located, and they could use a variety of computing devices—not just desktop computers with wired Internet access. In this arena, both the private sector and government are only in the early stages of exploring potential applications and modes of operation. Use of IT in government applications requires availability and reliability. Even for common, nondemanding applications such as renewing a driver’s license, expectations will be high that computer servers and communications links are functioning most of the time. Particular applications such as emergency services and crisis management demand a very high level of reliability and availability. TRUSTWORTHINESS Services supplied by government have to satisfy high expectations of security and confidentiality. In this regard, government must provide guarantees that generally exceed those offered by the private sector. Individuals simply will not tolerate unauthorized or accidental disclosure of

OCR for page 45
Information Technology Research, Innovation, and E-Government personal information contained in tax filings, social security records, and the like. Conventional business practice is to treat trust requirements with a risk-management approach in which the costs of implementing security measures are weighed against the consequences—a calculation that indicates what (small) level of exposure may be acceptable, as zero risk is impossible to achieve. This approach is not easily implemented in a government setting, where the high level of scrutiny leads to a very low tolerance for failure and where stringent security requirements are the norm. Security in both classified and nonclassified systems has long been an area of emphasis in government information policy, as reflected in the Computer Security Act of 19875 and the security provisions of OMB Circular A-130.6 One essential ingredient of a trustworthy system is the ability to authenticate the identity of someone wishing to access it, whether that individual is a citizen seeking to retrieve personal records or engage in a transaction online, or a government employee authorized to open records.7 Identity theft has already proven to be a troublesome issue in the private sector, and it is easy to anticipate heightened concerns if it occurred with e-government services. The simplest computer-based authentication mechanisms, commonly used in Internet transactions today, rely on a shared secret, such as a password. But passwords have a number of weaknesses; for example, because they are chosen by users, they are often easily guessed. One way to strengthen such mechanisms is to make the secrets dynamic, such as by using information contained in recent transactions or communications, but this approach also has vulnerabilities. Public-key cryptography technology provides more robust authentication, though a major challenge is distributing the public keys in a secure manner. A public-key infrastructure (PKI), which encompasses technologies, processes, and policies, makes use of certificates that allow the identity of users to be authoritatively associated with their public key. One PKI model relies on trusted third parties, known as certificate authorities, 5   The Computer Security Act of 1987 (P.L. 100-235). 6   Office of Management and Budget (OMB). 2000. Management of Federal Information Resources. OMB Circular A-130. November 30. OMB, Executive Office of the President, Washington, D.C. Available online at <http://www.whitehouse.gov/omb/circulars/a130/a130trans4.html>. 7   The importance of adequate online authentication and lack of a good solution was apparent in the 1997 Social Security Administration’s stumbles in providing online access to Personal Earnings and Benefit Estimate Statements (PEBES).

OCR for page 45
Information Technology Research, Innovation, and E-Government who themselves use public-key cryptography to digitally sign certificates that bind the identity of subscribers to a public key. But full-featured PKI systems are not yet commonplace in either the government or the private sector. Obstacles to widespread use include the complexity of key management; the need to establish systems and procedures for issuance and revocation of certificates; the lack of compatibility among different implementations; and the cost, inconvenience, and system-performance penalties associated with PKI systems’ use. Design choices must balance such tensions as ease of use versus security and accountability versus privacy. Several federal agencies are currently working to develop or implement PKI systems, complemented by cross-agency PKI efforts by the Chief Information Officers Council.8 For instance, the Department of Defense is rolling out the Common Access Card, which uses PKI technology to issue certificates to military, dependents, civilian employees, and some contractors. Challenges in implementing PKI in government include justifying the cost premiums associated with being an early adopter, enabling the necessary level of interoperability among systems and across agencies, developing systems that scale up sufficiently, and funding and sustaining a cross-agency effort. Government pilots will also provide insight on important operational issues such as how frequently cards are lost or stolen and how well these circumstances can be handled through certificate revocation. In the development of authentication services, technology, policy, and management considerations are tightly intertwined. In the United States, there has been long-standing opposition to government-issued universal identifiers, as evidenced by the absence of a national identification card (though the issue resurfaced in 2001-2002 in the wake of the September 11, 2001, terrorist attacks).9 The reality is that each person establishes a number of different identities for interacting with businesses, organizations, and government bodies, and that each of these organizations limits the information that it discloses. For instance, a driver’s license is used for a different purpose and is associated with information different from the purpose and information related to a voter-registration card, and there is 8   General Accounting Office (GAO). 2001. Information Security: Advances and Remaining Challenges to Adoption of Public Key Infrastructure Technology. GAO-01-277. GAO, Washington, D.C., February. 9   For an assessment of technology and policy issues associated with national identification systems, see the forthcoming Computer Science and Telecommunications Board (CSTB) interim report on nationwide identity systems. The report will be available from the CSTB home page at <http://www.cstb.org>, as well as the National Academy Press Web page at <http://www.nap.edu>.

OCR for page 45
Information Technology Research, Innovation, and E-Government little need for exchange of data between the respective authorities. Reflecting concerns about privacy implications, record linkage by government has in fact long been limited by law. With electronic records, however, it becomes much easier to use identifying information (such as name, address, or social security number) to match records across multiple identities and different government programs. Although there will inevitably be pressures for e-government systems to mirror the complex web of authentication and identity mechanisms found in the paper-based world, trade-offs will be required. These pressures must be balanced against increased complexity that could reduce the utility of e-government systems and compromise their ease of use. The committee’s exploration of crisis management pointed to the need for dealing with exceptions with respect to who can access information and systems. In a crisis situation, access to a variety of private and/or government confidential records may be needed—such as quick access to the blueprints of a building that has collapsed in an earthquake or to medical records needed for emergency treatment—although it would be more tightly restricted under normal circumstances. With electronically protected records, one needs explicit mechanisms for handling exceptions in an emergency. Simply persuading someone holding needed information of the urgency of a situation, which is how such contingencies are typically addressed with paper-based records, might not suffice, as this request alone would not permit records to be unlocked. More generally, tools supporting adaptive policies will be needed in order to accommodate crisis situations, for both enabling additional access (to provide adequate response) and shutting down certain kinds of access (to avert new threats). Additional trust issues arise if software agents that purport to act on behalf of an individual are used. First, if a user grants authority to an agent to act on his or her behalf, what kind of trust can that person have that the agent will not abuse the authority? This issue requires both validating the code and the environment within which the code will operate with the user’s authority. Second, when allowing code to run in one’s own environment (e.g., downloading code that will do useful things on one’s own local machine), how does the person trust that the code will only do what it purports to do and not, for example, gain unwanted access to personal information stored on the computer? Two approaches have been developed for trusting code distributed over the Internet from servers to clients: (1) have the source authenticate the code before it is run (the approach used with Microsoft’s ActiveX controls), or (2) run the code in a special “sandbox” environment that limits what the code is able to do. Of course, there are potential weaknesses in both approaches—such as

OCR for page 45
Information Technology Research, Innovation, and E-Government misappropriated credentials in the case of ActiveX or bugs in the code implementing the sandbox in the case of Java. Thus, it remains an open question whether citizens will trust software agents to provide information and services in such critical areas as tax or health care or feel comfortable running mobile code on their own machines. Another dimension of the trust issue relates to the consequences of inadvertent information disclosure. Given the authority associated with government data, premature release of, say, a key economic indicator could have significant implications for financial markets. (This was the concern raised in 1998 when the Bureau of Labor Statistics inadvertently posted employment numbers to a bureau Web site in advance of the correct release date and time.) While such tensions arise in the private sector as well, the long history of public concerns over the potential ability of government to gather and correlate information about individuals means that the government will continue to play a key role in how these tensions are addressed. Research alone cannot resolve them, of course, but it can create technologies that enable greater flexibility. For example, technological tools can enable adaptive policies such as mutually trusted intermediate agents. ACCESS AND CONFIDENTIALITY Government collects some information on a strictly confidential basis (e.g., answers to statistical surveys), and other information is collected under the assumption that it will be carefully guarded. Both overall and specific statutory and regulatory restrictions control how such information is used or released publicly, reflecting very strong expectations that the promise of confidentiality will be upheld. Yet at the same time, government is charged with releasing information and making it uniformly available to all. Disclosure-limiting practices and technologies are required when individuals (or organizations) provide data under confidentiality pledges— promises that if and when the data are released by the government, they will be in a form that does not reveal the sources’ identities. In certain cases, disclosure can be limited to carefully selected individuals (e.g., those who have agreed to disclosure terms or who are sworn employees accessing the data for internal government purposes). By contrast, an organization conducting a marketing survey, for example, is not bound to the same requirements. Governments are also expected to protect the privacy of citizens—to include safeguards on what information is collected and how it is used. A basic privacy framework for the federal government is provided by the Privacy Act of 1974 (P.L. 93-579), which applies principles of fair informa-

OCR for page 45
Information Technology Research, Innovation, and E-Government tion practice to federal agencies. As illustrated by the concerns expressed over government Web sites that tracked user behavior by using cookies (small files created by Web browsers to preserve information across Web sessions, providing such capabilities as tracking users’ access across Web sites or correlating separate visits to a single Web site) and by the subsequent 2000 OMB memorandum laying down strict conditions for cookies’ use—government experiences significant pressures to maximize the privacy accorded to those using government systems and services. An ongoing challenge is providing citizens with assurances that these policies are being fulfilled. STRUCTURAL CONSTRAINTS Businesses are competing to meet customer needs better and are tailoring their market segmentation (identification of customer groups such as small businesses or students) accordingly. For electronic services, this usually means coevolving e-business technology and business models. Improvement in customer focus is often accomplished by changing processes and organizational structure to support business needs better— facilitating in particular the creation of portals for information and transactions that are aggregated around the needs of specific market segments. For example, businesses can reorganize to bypass intermediaries in relationships (so-called disintermediation). Government has been taking some initial steps in this direction as well, as evidenced by multiple government information portals and an increasing number of aggregate transaction portals. Difficult in any setting, modification of processes and organizational structure in government is often significantly constrained by legal and administrative strictures. For example, information-sharing barriers have been legislatively established to protect citizen privacy. A further complication is that the desired span of aggregation can include state and local governments as well as federal agencies. These factors raise the possibility that, rather than reorganizing itself, government might adopt technology to offer integrated user-oriented services. For example, one might create portals that aggregate information or services for the customer and dispatch the elements of the query or transaction to the appropriate set of government entities. Significant privacy challenges are also posed by portals for government information and transactions that are operated by third parties. Though a competitive market of third-party portals could result in improved access to government services, they will face the issue of how to provide legal and technical safeguards relating to confidentiality and integrity of information that citizens enjoy when interacting directly with

OCR for page 45
Information Technology Research, Innovation, and E-Government the government. For example, the privacy (and compliance) of a citizen is enhanced by laws that prevent certain government entities (such as the Bureau of the Census or the IRS) from sharing information with other government entities. This means that potentially no one but the citizen would be legally able to fuse the multiple data streams provided to the various government entities. The issue is what privacy safeguards can be provided when the citizen presents a broad range of data to a portal operated by a government entity or a third party, in order that the portal operator may appropriately transact on the citizen’s behalf with multiple separate agencies. IT IN SUPPORT OF GOVERNMENT FUNCTIONS This section provides a brief overview of two areas—crisis management and federal statistics—explored by the committee, and it also takes note of several other government application areas in which government leads in technology demand. This presentation is not intended to be comprehensive. Rather, it is meant to provide an illustration of the sorts of unique requirements that arise in the government sector. Crisis Management and Homeland Defense Crises include natural disasters (e.g., hurricanes, earthquakes, floods, and fires) and man-made disasters (e.g., industrial accidents, infrastructure failures, and terrorist attacks). Crises can put lives at risk and pose significant economic threats. The escalating costs (both human and economic) of natural disasters, reflecting in part the increased population living or working in areas at risk, underscore the importance of enhancing “crisis management” capabilities. The September 11, 2001, attacks have led to greater attention to preventing, detecting, and responding to terrorist attacks. Crisis management encompasses crisis response—the immediate activities in the wake of a disaster—as well as the longer-term “consequence management” activities associated with addressing disasters past, present, and future to improve planning, preparedness, mitigation, and recovery efforts. By its nature, crisis management is a very challenging process. It draws on the capabilities and resources of multiple organizations, including national, state, and local government, nongovernmental organizations, and the private sector (and, depending on the location and scale of the disaster, on international organizations and other national governments). Efforts may encompass many functional areas, including transportation, communications, public works and engineering, firefighting, law enforcement, mass care, health and medical support, urban search

OCR for page 45
Information Technology Research, Innovation, and E-Government and rescue, hazardous materials management, food supply, and energy supply. Information technology needs include telecommunications, information management, geographic information systems, remote sensing, and models and simulations. As explored in more detail in the committee’s earlier report on a workshop examining crisis management,10 government’s particular technological needs include these: Robust, high-performance communications infrastructure that can adapt to changing demands, manage traffic congestion, and permit priority overrides for emergency usage; Temporary infrastructure that can be quickly deployed when extensive damage has occurred; Data management and delivery mechanisms that can function even when the communications infrastructure is degraded and that support unanticipated information needs; Composition of information and communications systems in ad hoc situations, such as the establishment of an “instant bureaucracy” for crisis response and recovery (see Box 2.1); Improved interoperation of information systems, including that among systems operated by all levels of government and by private organizations (see Box 2.2); Support for effective decision making and coordination in the face of uncertainty and stress; Overcoming of language and other barriers to communicating with citizens; Enhanced means of warning populations at risk, especially providing information targeted to local circumstances faced by individuals or neighborhoods; and Adapting of e-commerce technology and practices to accommodate the exception-handling necessary in crisis situations. Federal Statistics Federal statistics on population characteristics, the economy, health, education, crime, and other factors play a key role in a wide range of policy, business, and individual decisions. The decennial censuses—along with related estimates that are produced during the intervening years— 10   Computer Science and Telecommunications Board (CSTB), National Research Council (NRC). 1999. Summary of a Workshop on Information Technology Research for Crisis Management. National Academy Press, Washington, D.C.

OCR for page 45
Information Technology Research, Innovation, and E-Government Box 2.1 “Instant Bureaucracy” Capability for Crisis Response and Recovery One way to assist affected citizens in gaining access to needed information and services in the wake of a crisis might be to provide the means to build, on the fly, a portal that links information and resources from federal, state, and local agencies; nongovernmental organizations; and businesses. In the event of a hurricane strike, for example, such a service might provide individuals with information on evacuation routes and available shelters, allow people to apply for federal disaster assistance, access relief services, file insurance claims, and so forth. At the same time, it would provide capabilities that support decision making and that coordinate actions and resources among federal, state, and local agencies and nongovernmental organizations. Such a service would be accessible via multiple channels, including the Internet, from kiosks (including ones quickly set up in or near a disaster area), and wireless devices. The technology for “instant bureaucracy” would take the form of a toolkit for rapid ad hoc assembly of a portal to support response to a crisis. The following would be components of the toolkit: Software tools for such things as authentication of users, access control, data integration, and query across multiple databases; Standards for data exchange among systems and agencies; and Personalization/persistence capabilities that would retain state between transactions, thereby permitting users to reuse previously captured information and resume interrupted sessions. drive the allocation of federal funding to state and local governments and the apportionment of legislative districts. Federal statistical data are used to adjust wages, retirement benefits, and other spending, and to provide insight into the status, well-being, and activities of the U.S. population. The surveys conducted to derive this information are extensive undertakings that involve the collection of detailed information, often from large numbers of respondents. Most executive branch departments are, in one way or another, involved in gathering and disseminating statistical information. About a dozen agencies have statistics as their principal line of work—the two largest are the Bureau of the Census in the U.S. Department of Commerce and the Bureau of Labor Statistics in the U.S. Department of Labor—and others collect statistics in conjunction with activities in particular areas. These federal statistical agencies are characterized not only by their mission of collecting statistical information but also by their independence and commitment to a set of principles and practices aimed at ensuring the quality and credibility of the information they provide. Thus, these agencies aim to live up to the trustworthiness expectations of citi-

OCR for page 45
Information Technology Research, Innovation, and E-Government Box 2.2 Enhanced Capabilities for Rapid Exchange of Information Across Levels of Government and with Nongovernment Organizations Under the rubric of “homeland defense,” a number of cases involve interest in building enhanced capabilities for exchanging information among diverse actors, including government agencies at all levels as well as private organizations. Early detection of biological attacks, for example, depends on rapid gathering of information assembled by doctors, hospitals, medical laboratories, and local and state health officials. Detecting misuse of student visa programs to gain entry to the United States for illegitimate purposes requires correlating student visa applications, immigration entry information, and college attendance records. In each of these cases, there is technical complexity in terms of needing to draw on multiple databases systems that are operated by distinct organizations and do not necessarily conform to any common standard. Also, sensitivities are associated with information exchange in each case—both medical and educational records have privacy expectations (and specific legal and procedural safeguards) associated with them. The following technologies would help enable these capabilities: Middleware agents that issue standing queries into local databases and send the results to a federal agency. For example, agents could periodically pull specified student enrollment information from university databases and transmit it to the appropriate federal agencies. The means for organizations to audit the information transferred to federal agencies, to provide assurance that privacy safeguards are being upheld. Tools for anomaly detection based on information drawn from multiple databases. zens so that people will continue to participate in statistical surveys, and to meet the expectations of decision makers who rely on the integrity of the statistical products they use in policy formulation. Confidentiality protection and trust take on particular importance in federal statistical data collection: information records are linked and data are combined across agencies (subject to statutory and regulatory constraints), and the resulting statistical products are made available to the public as well as to specialized users. Those providing the information want assurances that it is properly protected. The federal statistical system leads demand for IT in other areas as well. For example, because the data sources and users are extremely diverse and heterogeneous, one cannot rely on a simple system for retrieval and integration. Survey data collection technologies were another example of government leading demand for IT cited at the committee’s workshop; federal data collection efforts frequently exceed private sector efforts both in scope and complexity.

OCR for page 45
Information Technology Research, Innovation, and E-Government Military Applications Military applications have been a dominant focus of IT research for more than 40 years. The military was, and continues to be, an important customer of IT vendors, and military-funded research programs have provided a considerable fraction of academic research funding. The demanding information-management, analysis, and communications needs of national defense have been reflected in research programs addressing such areas as survivability, security, interoperation, life-cycle management, software engineering, embedding, networking, wearable computing, systems research, modeling and simulation, communications (including support for mobility), and high-performance computing. Progress made in these research programs not only has enhanced defense systems but also has stimulated advances in the commercial computing sector. Archiving Government is charged with preserving records of its activities. At the federal level, the National Archives and Records Administration (NARA) has overall responsibility for archiving, and it provides guidance and assistance to agencies on the creation, maintenance, use, and disposition of government records. Each federal agency is responsible for ensuring that its records are created and preserved in accordance with the Federal Records Act of 1950.11 A combination of pressures to reduce costs, enhance access, increase efficiency, and modernize aging technical infrastructures has contributed to a shift to government information resources being “born digital.”12 NARA and the agencies are challenged by the increasing amounts of material that either originates as or is converted to digital formats. Another agency whose central mission faces pressure as materials are increasingly born digital is the Library of Congress.13 As elsewhere in government, expectations on the part of both the 11   Federal Records Act (44 U.S.C. 2904), as described in GAO. 2000. National Archives: The Challenge of Electronic Records Management. GAO/T-GGD-00-24. GAO, Washington, D.C. Available online at <http://www.gao.gov/new.items/gg00024t.pdf>. 12   The answer to whether this shift will realize cost savings, especially in the short term, is not straightforward. Major reports still must be published in paper form, and all of the writing and editorial costs remain the same regardless of format. In addition, there are costs of preparing files for the Web and costs associated with maintenance and connectivity of Web servers. The cost of printing a report may decline as more people use its online version, but it could well turn out that the total cost of providing the report increases. 13   CSTB, NRC. 2000. LC21: A Digital Strategy for the Library of Congress. National Academy Press, Washington, D.C.

OCR for page 45
Information Technology Research, Innovation, and E-Government general public and specialized stakeholders for archiving are being increased by what is available on the Web in nongovernmental contexts. This shift, together with rapid technological change, challenge existing archival practices and the institutions responsible for providing longterm access to government information. In the past, standardized publication and filing systems were used to manage at least key documents, facilitating archiving. Today, the situation is more informal, with a considerable body of material residing either on internal systems or on agency Web sites. Efforts to establish governmentwide systems for such digital information have had limited success. Electronic documents are frequently changed over time, and methods for tracking changes or validating the authenticity and integrity of archived electronic documents require further development. The advent of digital material also raises many new questions about what is worth archiving, especially with respect to files with dynamic content. Web sites come and go (for example, there is no government source for the Clinton administration’s version of the White House Web site <www.whitehouse.gov>). With paper documents, one might choose to retain only final documents such as memoranda and not interim documents such as rough drafts and phone logs. With digital information, how does one select—or even distinguish among—materials for archiving? Technologically, it might be easier to keep all e-mail or all distinct draft documents rather than to be selective, but to make sense of this plethora of material (much of which might be superfluous) would place a greater burden on the archivist. Could techniques be developed to automatically select and archive appropriate content? Even when material is selected for archiving, a key challenge is that while some government systems make provisions for upward compatibility, obsolescence of the associated hardware and software and the fragility of physical storage media pose the risk that digital information will become inaccessible over time. Consensus does not exist at present on standard formats or standard metadata descriptors, although efforts to develop and adopt them are underway. The basic technical challenge is how governments can maintain digital material that is vital for their own operations, for public access, or for future research. The simplest approach is to create a physical artifact (paper or microfilm printout) of digital materials so that they can be preserved using traditional archival techniques. This does not appear to be an acceptable solution. For example, it takes away the very benefits that electronic documents can provide, such as easier search and retrieval. Furthermore, important types of information, such as databases or links among documents, do not lend themselves to paper representation. On

OCR for page 45
Information Technology Research, Innovation, and E-Government the other hand, preservation of digital materials is not a straightforward matter in light of media fragility and system and format obsolescence. Approaches to preserving digital material include migration of existing digital material to new software and hardware platforms as necessitated by new technology; development of emulators of obsolete hardware and software systems to permit continued access; and development and adoption of long-term standards and descriptive formats with longer life spans. Once digital material has been archived, another key challenge is to construct search and retrieval systems to facilitate access.