Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
APPENDIX C 262 APPENDIX C User Support Services The increasing dependence of multiple communities of users on networks requires that the Internet have the following characteristics: it must be accessible at a low cost, be user friendly, and have a capacity and features that enable effective and meaningful access to needed resources. Also needed (at a minimum) is the following support service infrastructure:1 â¢ Education and training, as well as programs and services, to assist users in utilizing the network; â¢ Outreach services to identify new communities of users and their distinct information needs; and â¢ Coordination between network providers and service organizations for an integrated approach to supplying user support services and access. The most basic of the needed services is a "white pages" capability that will permit users to find each other on the network. The government-funded Network Information Center has traditionally operated a user naming service. However, a centralized approach of the kind practiced in the past is no longer appropriate, because the network has grown, and a centralized approach does not scale well. At this time, the government should support the planning and deployment of a distributed user naming service. It could support the Internet Engineering Task Force (IETF), in conjunction with other standards-oriented groups, in the selection of a set of protocols for this purpose. In addition, the government (perhaps through the Advanced Research Projects Agency) should fund the development of a software
APPENDIX C 263 package that implements these protocols and is made freely available, as TCP itself was distributed initially. Prior attempts to accomplish this goal have failed, due to a lack of focus and priority. The need for a distributed user naming service is even more pressing now. Once such a framework is in place, companies, educational institutions, professional organizations, and other such groups should begin to assemble directories of their members, which can then be incorporated into the eventual distributed naming service.2 Although the overall framework must be centrally developed (for example, by the IETF with the support of the federal government), the data themselves should be collected and managed in a decentralized manner "close to the source," to ensure their accuracy and to allow for local access controls. The need for user support has expanded as the offerings of the Internet have become more diverse and complex and as the user population has grown. The Internet Society through the IETF has been improving user support through the development of a series of documents that help new users to join the Internet. For this kind of effort a centralized approach works well, and it should continue. Private enterprise and private-sector institutions have major roles to play in providing direct support to users. Answering questions, resolving service problems, and facilitating attachment are necessary services that will require substantial effort. The data network is not like a telephone. It is a complex and evolving technology, and it sometimes displays inexplicable and undesirable behavior. Assistance for users is thus critical. Sellers of individual products today provide good user support, but they cannot be expected to solve problems related to overall system integration. Today, institutional network users are supported through a growing body of for- profit system integrators and resellers, who provide support as their major product. The major problem in this area today is the lack of a mechanism for locating and evaluating these providers. In other words, we need a marketplace. This, of course, is what the information network is intended to provide. Individual users cannot hire a system integrator to solve their specific problems. Since most users attach to the network because of some specific professional or personal interest, such as work-related activities or a hobby, one approach to providing networking support for (some) individuals would be for special-interest groups, for example, professional societies, to contract to provide network support to their members. Existing network providers today offer support as an important part of their overall network service. As for-profit providers move to take over the actual operation of trunks and switches, these existing network providers can continue to provide service to their clients. Such an ap
APPENDIX C 264 proach would separate the business of network operations, which is becoming more clear-cut in its nature, from the still rather evolving task of effectively helping users. Customer service is a significant source of revenue for providers of computer products. The problems of supporting users of large-scale networks are more complex, and resolving these problems should present a significant business opportunity. What is important is that users clearly understand the cost and value of this service. Today, for example, cost competition for personal computers has eliminated any cost margin for support, and many users are first pleased at the very low cost of computers, and then frustrated when they cannot get parts to work. From the very beginning, network providers should be encouraged to price service and basic network access so that the cost of each is apparent, and so that a realistic level of support can be made available. NOTES 1. In the interest of assigning responsibility for such services, one suggestion made elsewhere was to involve the IITF in service delivery. "Draft Report of the Federal Internetworking Requirements Panel," prepared for the National Institute of Standards and Technology, January 14, 1994. 2. The harvesting of names and electronic mail addresses is becoming easier through technology, although there are drawbacks associated with these practices.