12 March 1992 /internet/legislative.actions/hearing.12mar92/roberts.testimony United States House of Representatives Committee on Science, Space and Technology Subcommittee on Science Hearing on National Science Foundation Network March 12, 1992 Statement of Michael M. Roberts, Vice President, EDUCOM Chairman Boucher and members of the Committee, I am pleased to present testimony today on behalf of the EDUCOM Networking and Telecommunications Task Force, a group of forty-eight universities with joint interests in the development of advanced computer networks to support research and education. NSFNET Success. Over the past five years, NSFNET has compiled one of the most remarkable success stories in the history of American science. In this short period of time, through a partnership of government, industry and higher education, an advanced production network with the highest level of bandwidth available anywhere in the world has been designed and deployed in the research and education community in the United States. At the same time, the network has been transformed from one serving a narrow group of supercomputer centers and federally supported research sites into one with connections to more than six hundred colleges and universities and over a thousand public and private research sites. The global Internet family of research and education networks, of which NSFNET is a part, is growing equally rapidly and now reaches more than three- quarters of a million computer systems in more than one hundred countries. On the campuses of the members of the EDUCOM networking task force alone, more than one million students, faculty and researchers have gained access to NSFNET and the Internet. Of special note is the fact that NSFNET now connects more than a thousand high schools and several hundred libraries as a result of an effort by NSF and the regional networks to reach all levels of education. In addition to the benefits within research and education, the success of NSFNET has materially aided the growth of a commercial market for Internet products and services which it is estimated will exceed four billion dollars in 1992, with growth at the rate of seventy-five percent a year. This progress - in advanced network services, in access provided to the research and education community, and in technology transferred to the private sector - far exceeds the levels planned five years ago and is a tribute to the commitment of the NSFNET partners and to the able leadership of the Foundation and its Networking Division. In the remainder of this statement, I would like to focus on three key issues - a new cooperative agreement for NSFNET, a commitment to a common infrastructure with participatory governance for the NREN, and the linkage of the NREN program to establishment of a broadband communications network for all Americans. We are submitting additional EDUCOM background material on NSFNET and the NREN for the record. NSFNET Competition for New Cooperative Agreement. Last November, the National Science Board approved a proposal by the Networking Division of NSF that it conduct a competition leading to new cooperative agreements for continuation of NSFNET for the period 1993-1996. In developing its proposal, NSF drew on studies and recommendations from a number of organizations within the networking community, including EDUCOM. We believe that the plan for new cooperative agreements is an excellent one that not only provides stability of network service but promises to continue the progress in technology that has been an important feature of the current agreement. However, we wish to comment on the aspect of the intended plan that deals with competition for the award within the private sector and a related issue, raised in recent press articles, that NSF should be obligated, in a spirit of fairness, to provide a level playing field for competitors. First, it should be understood that the new award will not be for standard commercial telecommunications services. It will be a cooperative agreement among partners in an effort to maintain and improve a leading edge, state of the art computer network which continues to meet the demands of the best science of which the United States is capable. That is the stated goal of the NREN legislation sponsored by your committee last year and of the program under development by the Administration. We should set our sights no lower. Second, with respect to levelness of playing field, it is not and has never been the responsibility of a federal agency to guarantee market entry for a private sector firm. Nor is it the responsibility of federal agencies to transfer technology to the private sector in a manner which guarantees market entry. What NSF has done very successfully in the current award, and proposes to do again, is establish the criteria under which firms, or joint ventures of firms with public sector participation, may join with the Foundation in designing, testing, and deploying advanced network technology. The greater the size of the funding and resource commitment that is made by private sector firms in competing for the award, the greater the likelihood of rapid progress in advancing network technology. It is important to distinguish between competition for the new cooperative agreement, and the creation of a competitive market in the private sector for network services based on NSFNET/NREN technology. Progress made during the course of the next agreement will support and strengthen an already impressive level of commercial network services that has developed in the last several years. Common Infrastructure for the NREN. At the present time, federal agencies participating in the HPCC program are preparing an NREN technical and management plan under the aegis of the Federal Networking Council (FNC). It is premature to comment on specifics of the plan until it has been released and reviewed. However, the view has been expressed by some that it is sufficient for the NREN to be just a family of largely independent networks, with connectivity and services under the control of individual agencies. This is clearly an unsatisfactory approach. It fragments the available federal resources. It confuses industry, which will be unsure of which agency approach will win out over others. It will reduce the rate at which the common infrastructure can be developed and implemented. Some federal officials have the opinion that full support for the NREN would jeopardize mission critical network applications such as real time satellite data collection and nuclear energy experiments. These special applications have never been part of the NREN as envisoned by the universities and constitute a small fraction of total agency use of computer networks for research and education. The universities, having made major investments in their campus networks and NSFNET connections, and having joined together to create and sustain the regional networks, believe that their federal agency partners in the NREN should make a similar commitment to a common networking infrastructure. Such a commitment must include agreement on mechanisms for participation in the creation and application of standards and policies for the network. When your committee continues its review of the Administration's NREN program later in the year, we believe this matter deserves further inquiry. NREN Linkage to National Information Infrastructure (NII). As the revolution in computer networking has gained momentum in recent years, the potential value of NREN technology is being recognized in areas beyond the original leading edge, Grand Challenge research objectives. Mr. Chairman, you and Representative Oxley have taken the Congressional initiative in the House with the introduction of HR2546, which calls for rapid deployment of broadband technology in the national communications infrastructure. In a related development, the Computer Systems Policy Project (CSPP) has called for a broader vision of the NREN and specifically recommended that the Administration "establish a technology and policy foundation for an information and communications infrastructure for the future." The FCC has also taken note of these developments in holding future network hearings last spring and issuing a Notice of Inquiry into Intelligent Networks last December. The university community believes that both the NREN and a broadband communications infrastructure for America are important, perhaps critical, national objectives. However, they are not the same, and neither should be treated as hostage or servant to the other. EDUCOM recommends that an explicit linkage be created between the two objectives. The NREN, guided by a government, industry and education partnership in developing and deploying advanced network technology, should be the means by which the country supports its research and education goals, and at the same time develops, tests, and transfers to the private sector its successes in technology. The NII, guided by new federal and state communications legislation, should be the means through which a revitalized communications industry, utilizing digital and fiber optic technology, brings the benefits of the Information Age to every American. Forging a new national communications policy which protects the public interest and enhances the private sector role in providing advanced communications services is a difficult challenge. We in higher education have a major stake in the creation of a broadband communications infrastructure to help us fulfill our educational mission. We are prepared to assist the Congress in achieving this vision, which will assure continued U.S. leadership in a world which is rapidly becoming a global information society. Summary. In conclusion, I would like to reiterate five key points: First, NSF leadership in the development and delivery of NSFNET has been outstanding. Research and education are being extraordinarily well served by the NSFNET program, the sucess of which is obvious from every measure of network connectivity and use. Second, the NSFNET program has positively stimulated the marketplace and there are a growing number of commercial providers of network services based on this technology where there were none only three or four years ago. Third, we in the research and education networking community believe that the planned "recompetition" for a new NSFNET cooperative agreement is an appropriate and manageable compromise between conflicting objectives. It increases participation from the private sector while continuing a successful management structure proven during the current agreement. Fourth, we urge the Congress to insure that the infrastructure and governance of the NREN reflect an effective partnership between the broadly based research and education community and the federal agency establishments so that standards and policies for the network will be formulated for the broadest applicability and greatest good. Fifth, we recommend that the National Information Infrastructure and implementation of broadband communications envisoned in HR2546 be explicitly linked to the NREN program, though in no way merged, so that the NREN may take early advantage of broadband infrastructure to support research and education goals while leading in the advancement of the technology.