Government Services Information Infrastructure Management
The growth and deployment of the Government Services Information Infrastructure (GSII), and its relationship with the national information infrastructure (NII), will require the federal government to drastically change the way it acquires and deploys telecommunications for support of the government's businesses. The GSII is an enlightened attempt by the Clinton/Gore Administration to form a "virtual government" crossing agency boundaries to interact more closely with industry and with the public, as well as with state and local government, to greatly improve the delivery of government services. The GSII is that portion of the NII used to link government and its services, enable virtual agency concepts, protect privacy, and support emergency preparedness needs. The GSII will have to be an integral component of the NII in order to do so. The GSII and other private sector efforts will have a significant impact on the design, development, and deployment of the NII, even if only through the procurement of such services.
This paper concludes that the federal government must adopt new mechanisms and new paradigms for the management of the GSII, including improved acquisition and operation of GSII components in order to maximize taxpayer benefits, to optimize the delivery of government services, to ensure that the GSII is an integral component of the NII, and to adopt industry standards rather than setting them. Government requirements and applications, as well as the available technologies to address those requirements, will continue to evolve; therefore so must the government's use of technologies and services. The requirements from federal government services and the users of these services logically form affinity groups that more accurately and effectively define these common requirements, that drive the adoption and use of industry standards, and that provide a significant technology marketplace to capture the vision of the NII both now and in the future.
It is critically important that the federal government adopt a management scheme that improves its ability to work with U.S. industry to ride the coming Third Wave,1 as opposed to being wiped out by it. A new management scheme is also needed to improve cooperation between the government and its partners (i.e., the private sector, academia, and state and local governments) and to improve the chances for a successful deployment of both the GSII and NII. This new management scheme must be built upon a modular and evolutionary approach for the GSII as well as for other large systems developments to greatly improve the successful use of information technology (IT) by the government, especially to provide service to the private sector and to the U.S. citizenry.
NOTE: This paper was commissioned by the Government Information Technology Services (GITS) subcommittee of the Committee on Applications and Technology of the Information Infrastructure Task Force (IITF). It was drafted for the GITS by John S. Cavallini and Robert J. Aiken of the Department of Energy. It was reviewed and endorsed for submission to the NII 2000 Forum by the GITS membership.
Today, the private sector is developing and deploying information infrastructure. At the same time, the government is concurrently developing and deploying information infrastructure, mostly through contracts with the private sector. Under the lead of and at the encouragement of the Clinton/Gore Administration, the federal government has placed increased emphasis on the development and deployment of an NII as a strategic priority. This emphasis results from the understanding that properly leveraged information and information technology are among the nation's most critical economic resources, for manufacturing industries as well as for more modern services industries for economic security and for national security.
The Clinton/Gore Administration has made a commitment to work with business, labor, academia, public interest groups, Congress, and both state and local government to ensure the development of an NII that enables all Americans to access information and communicate with each other using combinations of voice, data, images, or video at anytime, anywhere.2 This commitment was articulated very well by the National Performance Review (NPR) through its emphasis on using IT as a key element in creating a government that works better and costs less.3 The President and Vice President recognize the need to use IT to improve Americans' quality of life and to reinvigorate the economy. To this end, they outlined a three-part agenda for spreading IT's benefits to the federal government: (1) strengthen leadership in IT, (2) implement electronic government, and (3) establish support mechanisms for electronic government. Thirteen major IT areas were identified for accomplishing the three-part agenda:
Development of an NII is not an end goal in or of itself. Government requires an infrastructure to conduct its business more effectively and to deliver services to the American citizenry at lower cost to the taxpayers. A number of suitable national-scale applications or uses of the NII have been identified and documented by the IITF's Committee on Applications and Technology.4 These uses of the NII, in addition to nationwide humanistic applications such as health care and education, include the fundamental businesses or enterprises of the federal government such as law enforcement, electronic commerce (including benefits), basic research, environment, health care, and national security, and as such represent a significant set of driving requirements for NII deployment.
In recognizing this fact, the NPR concluded that the government use of IT and development of information infrastructure should be improved and better coordinated in order to effectively address government business requirements. The NPR made approximately 60 recommendations for action in this regard, including the development of a plan for a GSII to electronically deliver government services and to integrate electronic access to government-provided information and services. The GSII is that portion of the NII used to link government and its services, enable virtual agency concepts, protect privacy, and support emergency preparedness needs. It was also recognized that better integration and coordination were required not only across federal government
agencies, but also across federal and state governments, and local and tribal governments, to the private sector and to the public at large.
To achieve these goals, the Vice President established the GITS working group in December 1993. The mission of the GITS working group is to promote the improvement of agency performance through the use of IT, accelerate the deployment of advanced networking technologies, and in conjunction with the Office of Management and Budget (OMB) and General Services Administration (GSA), establish procurement and implementation policies, practices, and directives designed to improve productivity and reduce costs. The working group's responsibilities include implementing the NPR's recommendations, coordinating federal agency and White House initiatives for evolving the NII, and enhancing cross-agency IT collaborations as well as government and industry IT activities.
The GSII: Analysis
The GSII is that portion of the NII used to link government and its services, enable "virtual agency" concepts, protect privacy, and support emergency preparedness needs. The "virtual agency concept" allows services and functions to cross traditional organizational and geographic boundaries. For example, a U.S. citizen could, in this fashion, receive multiple services such as social security, food stamps, tax preparation assistance, information on the national park system, and so on through a single interaction with one agency.
The attributes of an effective GSII must include timely infusion of appropriate IT to meet requirements; coordination of infrastructure across government requirements and with the private sector; effectiveness, while also allowing for innovation and flexibility to meet specific needs; information surety to guard citizens' rights and privacy while also providing for national security needs; cost-effectiveness of IT acquisition and development while also encouraging a competitive technological environment; ubiquitous accessibility to all citizens, including those with disabilities; ease of use for locating government information or for service delivery; transparent (to the user) integration of the GSII portion of the NII with the rest of the Internet, NII, and the global information infrastructure; shared resource use across government entities; and the mechanisms, processes, and procedures to appropriately manage such a diverse infrastructure.
As opposed to the open system interconnect seven-layer model, information infrastructure is more often currently viewed as having three or four layers; a recent report of the National Research Council (Realizing the Information Future) proposes a standard open bearer service to promote, and perhaps maximize, interoperability across the NII.5 This is a significant feature that the GSII must adopt and integrate as it evolves. Nevertheless, merely focusing on the NII layer concept and other technical models can lead one to neglect two of the most important components of the GSII and the NII. They are people, both using and providing services over the NII, and the coordination mechanisms needed to effectively manage all GSII activities and efforts.
Standards are often cited as the only way to achieve interoperability across the NII or GSII. However, standards can often be too much of a good thing.6 The technology life cycle averages 12 to 18 months, with new hardware and software being introduced at a dizzying pace. The current standards processes have not been able to keep up. Competition among many vendors is crucial for reducing cost as well as ensuring a viable "technological gene pool" for both current and future requirements. Innovation requires new concepts. Flexibility is required to meet a wide and diverse set of special requirements and can enable the incremental evolution of the GSII and NII in a stepwise manner. Large systems development must be modular and better coordinated. Today's infrastructure has an enormous amount of different components composed of various technologies and standards. Therefore, a modular, seamless integration and evolution of the multicomponent GSII into the evolving NII will need to be based primarily on voluntary processes and proven interoperability solutions rather than on mandated standards. These and many more facts lead us to the conclusion that standards alone are not sufficient.
Furthermore, although the Internet protocol suite was originally developed as a military specification, it was not adopted as a federal procurement standard or an industry standard. Nevertheless, through versatility and capability to meet the research community's requirements, while supporting a productive range of interoperability, the Internet Transmission Control Protocol (TCP)/Internet Protocol (IP) became the most widely used
internetworking protocol. Therefore, common requirements and mutually beneficial, modular, evolutionary technical solution(s) will provide the best gauge of the range of interoperability that is ultimately needed.
Current federal coordination mechanisms are a result of the Brooks ADP Act of 1965 which was passed in an era characterized by mainframe computers when time-shared processing was the primary technique for managing scarce IT resources. Reflecting the centralized technology, the decisional and oversight responsibility for planning, acquiring, and managing IT was taken from the heads of the various executive agencies and vested in the administrator for general services. The theory was that investments in IT must be centrally managed and that only a dedicated cadre of IT professionals could perform the function.
Over the years, the original model eroded and technology became decentralized. Today, agencies plan and manage their own IT investments; the authority to contract for or to procure IT solutions is "delegated" to the agency heads. As a condition of the delegation, agencies are subject to an additional, centralized acquisition and management oversight. That oversight, however, is generally redundant in nature (i.e., oversight checkers outside an agency are checking internal oversight checkers) and neither set of checkers possesses the institutional expertise or the resources to adequately understand, let alone manage, the government's increasingly decentralized and diverse IT infrastructure. Ultimately, the centralized control model of the Brooks Act, which reduced both the responsibility and the authority of the heads of federal agencies, contributed to significant failures in IT management and seriously misspent resources.
The old centralized approach could achieve interoperability neither across the government infrastructure nor with the private sector, and it was even counterproductive in efforts to do so. In the future, an NII will undoubtedly incorporate multiple paradigms of internetworking, interoperability, and communications, thus making the task of coordination and interoperability even more difficult. Logically then, the question is how to promote interoperability. We propose starting with what the GSII will be used forthe business functions that can be considered GSII applicationsand with those people who share common interests in both using and providing the GSII. The applications should determine what standards and technologies are required and will provide interoperability among their own constituency as well as with other groups, if properly coordinated.
It should be noted, however, that GSA has taken steps to try to improve the government's acquisition of IT and to recognize this new paradigm of distributed IT functionality and management. Two notable examples are the Time Out Program that tries to get faltering large-systems efforts back on track and the Government-wide Acquisitions Contracts Program that empowers lead agencies to conduct multiagency contracts.
Applications and Affinity Groups
Government services and uses of the NII for law enforcement, benefits, and health care touch every community. Requirements for research, the environment, and national security go beyond the geographic requirements. The latter also require advances in the state-of-the-art, leading-edge IT capabilities. Hence, government applications provide significant financial leverage and incentive for NII deployment, as well as for new IT development. If one adds applications for education, generic electronic commerce, and energy delivery and management to this set of federal NII applications, the cost leveraging for deploying the NII becomes significant. For example, in energy, "[e]lectric utilities already serve over 95 percent of American homes (a percentage point above telephone companies)" with the "likely requirement that all these homes will need access to advanced telecommunications to manage energy" consumption.7 It is unlikely that telecommunications for energy demand management will replace conventional access mechanisms, especially for the last mile to residences. However, the energy utilities telecommunications infrastructure can augment, leverage, and enhance the cable and telecommunications industry infrastructure and facilitate access to the NII, and hence the GSII, for almost all Americans, including those in remote rural areas.
In addition to the impact the energy utilities may have on the NII, FTS 2000 and its successor will have a more direct impact on both the NII and GSII, since it will be a major vehicle for supplying telecommunications and services to the federal government. Today's FTS 2000 "currently provides intercity telecommunications for 1.7 million federal government users."8 Post-FTS 2000 is expected to go beyond the current FTS 2000 by delivering intercity telecommunications as well as value-added services such as electronic mail, key management,
teleconferencing, and more. This deployment will be an integral component of the GSII and will use industry-provided services almost exclusively, thereby leveraging the products of industry rather than competing with them. However, this reliance on and use of industry products and services will not lessen the impact that the federal government, via procurement vehicles, will have on the NII and GSII; therefore, the proper coordination of the deployment and management of the GSII with respect to the NII is even more critical.
The focus should be on achieving an effective, consultative, and collegial interagency process for managing the GSII. This will require creating a methodology for managing evolving IT policies, priorities, and programs that provides and coordinates the framework within which the mission, as well as the administrative, IT requirements, and activities of individual federal departments and agencies are conducted. The principal focus of the current top-level IT management is on what today are perceived as the common goalsmainly the administrative requirements for finance, personnel, facilities, and so onof the overall enterprise called the federal government. Implementation of a new enterprise model, based on an understanding of GSII requirements driven by these other mission applications as well as those that represent government services to its customers and their non-government requirements, is essential to making the government work better and cost less.
The Federal Internetworking Requirements Panel (FIRP), created by the National Institute of Standards and Technology at the request of the Department of Energy's Office of Scientific Computing and the high-performance computing and communications community, issued a report that recommends increased responsibility for this shared infrastructure, such as a GSII, for the mission areas of federal agencies or their logical affinity groups in as compatible a way as practicable with the common vision of the federal government.9 The term "affinity group" means government agencies, or functional interest groups therein, that share information electronically and have common IT requirements. The NPR emphasized the need for improved infrastructure for cross-agency groups. In addition, the FIRP report recommended the development of affinity groups to enhance cross-agency collaborations. Taken together, these factors make it reasonable to require that each agency explicitly ensure that GSII issues, including interoperability, are addressed not only within a given federal agency, but also with external affinity groups, industry, and the public, in which these agency mission or "enterprise" activities take place.
The federal research community is an excellent example of an affinity group. During the past decade, the federal research community has placed a very high priority on the application of advanced information technologies to enable advances in many science disciplines. As experimentation becomes too expensive, too unsafe, or too environmentally unsound, there is an increase in the importance and value of computational experiments, collaborative virtual reality, and distributed computing infrastructure technologies for remote access to one-of-a-kind facilities, shared data, and dispersed collaborations. Correspondingly, a sound and capable NII is needed, since the research community crosses many organizational boundaries, large geographical distances, and multiple capability needs. By virtue of its use of advanced capabilities, the research community or affinity group has mobilized to coordinate requirements and to cooperate in their solution.
Interagency, cooperative activities in the mid-1980s, prior to the start of the High Performance Computing and Communications Initiative (HPCCI), included studies to examine the need for common infrastructure. These studies resulted in the development of the National Research and Education Network component of the HPCCI,10 which proposed, and subsequently implemented, Internet technologies to support the internetworking needs of the research programs of the federal agencies. In one notable case, the Energy Sciences Network (ESnet) extended these common affinity group requirements to include other administrative requirements by implementing a multiprotocol network, installing gateways for multiple e-mail systems, and so on. And although technically different, the ESnet, the National Aeronautics and Space Administration's Science Internet, the NSFNET, and the ARPANET all were able to interoperate within an acceptable range for the research community as separate, but integral, modules in a network of networks. These activities and studies drew upon the expertise and the involvement of many academic and industrial researchers as well as organizations external to the federal agencies, constituting a large affinity group and adding a large user base to the Internet and establishing its technologies as a viable and productive technology paradigm.
This success notwithstanding, the research community of each agency also needs to interact electronically with other parts of its own agency (e.g., using ESnet solutions for administration and Information resource management), which are not normally a part of the research affinity group, as well as with its affinity group
colleagues. These requirements of the research community did not always converge within the group (although convergence was generally discussed and studied); however, convergence (and ultimately interoperability) was even more problematic outside of the research community.
Affinity groups do, however, represent a very powerful method for identifying common requirements, coordinating infrastructure needs, and promoting and maximizing interoperability of applications and services across agency boundaries, extending these ''functional" areas as virtual agencies. Affinity groups can also extend to industry and the private sector. Affinity groups could establish a common perspective for evaluating new technologies, eliminating unwarranted redundancies, interacting with various affinity subgroups or working groups, and sharing data, information, and knowledge about their enterprise or business area and how IT promotes effectiveness. By focusing on common requirements and solutions in this manner, affinity group activities can result in application-driven standardization for supporting important common functions, for setting priorities for new tasks, and for understanding the minimum capabilities needed to perform common business functions. They can also enhance the overall coordination for multiorganizational, distributed enterprises, as well as other attributes needed to maximize coordination for multiagency, distributed government services (i.e., the virtual agency) in the information future through the GSII.
Optimizing GSII and NII Compatibility
Federal IT management reform is needed to correct the current problems with regard to managing IT in the federal government to achieve a management scheme that works better. A new management scheme should set up mechanisms to aid agencies in carrying out their responsibilities, to evaluate agency IT investments via performance measures of programmatic results and products, and to promote both compatibility and interoperability across recognized affinity groups. We propose the establishment of a high-level "leadership council" that brings together recognized leaders of larger or more critical mission-driven affinity groups along with government and private sector IT services providers to:
It should be noted that administrative functions for personnel, finance, procurements, facilities management, and so on logically combine the traditional federal information resource management (IRM) organizations together into an affinity group. One could conclude that this grouping results in a federal government affinity group based on electronic commerce. Individual agencies would still require policy and oversight function for IRM or IT management activities; however, it is not envisioned that multiple large centralized IRM organizations would or could promote GSII interoperability goals (e.g., large redundant organizations both in the GSA and in the agencies) or adequately serve the GSII goals and objectives well into the information future.
Promoting interoperability, coordination, and information exchange across agencies and affinity groups is of paramount importance for achieving the goals of the NII and for creating an evolving and successful GSII. The best means of achieving this goal is through the use of the very same IT that will make the GSII successfulin other words, use of the technologies that we advocate for all agencies' businesses to further enhance the management and oversight of the GSII. As an example, a very important IT for furthering this goal is the World Wide Web (WWW), which is a product of both the federal and international high-energy physics community and also of the high-performance computing and communications research communities.
The GITS working group has endorsed the use of the WWW to create and maintain a "living" and "evolving" document accessible to all over the net.11 The GITS working group has the responsibility for implementing the NPR IT recommendations, one of which is to develop a GSII plan. Understanding the changing nature and the wide variety of GSII components, elements, and layers, the GITS decided to create a collaborative, on-line document on the WWW.12 The document consists of summarized reports of various affinity groups, agencies, panels, committees, and so on presenting the most current thinking with regard to GSII technology direction, issues, applications and management. It allows for interactive feedback and comment. All contributions will be referred to a GITS subgroup and/or be addressed by other expert groups.
On-line implementation plans, updates, and dialogue for the GSII, as well as its committees, activities, documents, and plans, will help to promote common understanding of issues and their status as well as to establish the foundation for the discussion of new ideas and/or requirements by government, industry, and the public on both a national and international basis.
Federal Versus Nonfederal Issues for the GSII
Some issues that need to be resolved for the deployment of the NII result from the differences between policies, regulations, and practices of the federal government versus those in the private sector. It is timely that the Congress and the Administration are now committed to telecommunications reform, as this will help lay the foundation for resolving some of the GSII versus NII issues in the future.
Key issues that need to be addressed for successful GSII and NII integration include procurement reform, key and certificate management infrastructure, electronic signature, intellectual property, common carrier status and open access for the network and information service providers, standards setting, and cost recovery for shared infrastructures.
First, federal IT management reform is needed to deal with Third Wave13 (i.e., truly information age) organizations and business so that the federal government can, in fact, achieve its NPR goal of creating a government that works better and costs less.14
Second, the necessary features and mechanisms for achieving a successful GSII should, at a minimum, include:
Third, improvement in government and private sector cooperation (e.g., via extended affinity groups) is required to achieve a more timely acquisitions process; a more responsive standards process; open-access interoperability between NII and GSII services providers; and revision of the Federal Advisory Committee Act (FACA) to provide relief for some activities with regard to the GSII and NII to encourage rather than to discourage such cooperation. The FACA can often discourage government activities from utilizing academia, industry, and other private sector organizations in developing priorities and goals for designing and implementing the GSII and ensuring that the GSII is an integral component of the NII.
Lastly, continued dialogue on the direction of development and deployment of the GSIIespecially relative to its superset, the NIIvia the WWW implementation of the GSII Plan, is needed to ensure convergence of these two very important national resources and to achieve the optimum "range of interoperability" and the maximum benefit that one could expect from such a complex and diverse infrastructure.
1. Toffler, Alvin. 1980. The Third Wave. William Morrow & Company, New York, pp. 233, 262, and 404–415.
2. Information Infrastructure Task Force. 1993. The National Information Infrastructure: Agenda for Action. Information Infrastructure Task Force, Washington, D.C., September 15.
3. Office of the Vice President. From Red Tape to Results: Creating a Government That Works Better and Costs Less. U.S. Government Printing Office, Washington, D.C., September.
4. Committee on Applications and Technology (CAT), Information Infrastructure Task Force Committee. 1994a. Putting the Information Infrastructure to Work. Information Infrastructure Task Force, U.S. Department of Commerce, Washington, D.C. Also, Committee on Applications and Technology (CAT), Information Infrastructure Task Force. 1994b. The Information Infrastructure: Reaching Society's Goals. U.S. Government Printing Office, Washington, D.C., September.
5. Computer Science and Telecommunications Board, National Research Council. 1994. Realizing the Information Future: The Internet and Beyond. National Academy Press, Washington, D.C., May.
6. Aiken, R.J., and J.S. Cavallini. 1994. "Standards: Too Much of a Good Thing?," ConnexionsThe Interoperability Report 8(8) and ACM StandardView 2(2).
7. U.S. Department of Energy. 1993. Positioning the Electric Utility to Build Information Infrastructure. DOE/ER-0638, Department of Energy, Washington, D.C., November.
8. Acquisition Working Group. 1994. Analysis of POST-FTS2000 Acquisition Alternatives. Interagency Management Council, Washington, D.C., September.
9. Federal Internetworking Requirements Panel. 1994. Report of the Federal Internetworking Requirements Panel. National Institute of Standards and Technology, Washington, D.C., May 31.
10. Federal Coordinating Council for Science, Engineering, and Technology, Office of Science and Technology Policy. 1991. Grand Challenges: High Performance Computing and Communications, The FY 92 U.S. Research and Development Program. Committee on Physical, Mathematical, and Engineering Sciences, Office of Science and Technology Policy, Washington, D.C., February 5.
11. GITS Working Group. 1994. Vision for Government Information Technology Services and the NII. GITS Working Group, Washington, D.C., July.
12. See http://www.er.doe.gov/production/osc/gsiiplan.
13. Toffler, The Third Wave, 1980.
14. Gore, Albert. 1993. Creating a Government That Works Better & Costs Less: Reengineering Through Information Technology. U.S. Government Printing Office, Washington, D.C., September 1993.