Last update at http://inet.nttam.com : Thu Jun 8 10:56:50 1995 Community Networks and Small Internet Service Providers Richard Civille Executive Director Center for Civic Networking P.O. Box 5352 Washington, DC 20009 202-362-3831 rciville@civicnet.org Abstract This paper briefly examines emerging issues between community networks, small Internet service providers (ISPs) and regional bell operating companies. What are the areas of conflict between small ISPs and community networks, and opportunities for synergy and mutual benefit? What market, technology and regulatory changes could result in larger telecommunications industries absorbing the business processes of these smaller enterprises? The paper will expand on a perspective that small ISPs and community networks complement their best services to local markets by differentiating access and content roles, and the community network's unique role in working with the traditionally underserved. The long-term role of the small ISP is uncertain, in the face of the deployment of new telecommunications services such as frame relay. Background Community networks may now serve close to one million people around the country, although precise figures are unavailable. According to the National Public Telecomputing Network, its 53 affiliate systems have almost 380,000 users alone, making them the fourth largest consumer on-line service in the country. If the total number increases as a function of the annual growth rate of the Internet itself, which nearly doubles yearly, this number could be close to five million within three years. Because of the explosive interest in Internet services such as the world wide web, price drops in high speed modems, and most new personal computer purchases now being made by individual consumers, the growth rate of community networks can be expected to become very strong within the next 18 months. Begun in the late 1970s with the Community Memory Project in Berkeley, CA and evolving through such initiatives as the Santa Monica Public Electronic Network (PEN) and the Cleveland Free-Net, these systems have been characterized by free or low-cost dial-up public access, volunteer management, with technology support from colleges, city government or public libraries. More recently, as the number of these systems has begun to grow, these characteristics are beginning to change. Driven towards developing means of long-term financial sustainability, some community networks are exploring various charging or pricing mechanisms, becoming distinct organizations, and are collaborating more closely with local business groups. Part of the phenomena of these systems, reflected in their frequent partnering with colleges and universities, was the growth of the Internet. In the late 1980s it became increasingly clear to many that the Internet was becoming a public utility; that it was only a matter of time before it burst beyond the cloistered walls of the academic research community to the rest of the world. Community networks became local on-ramps to the Internet, an easy outreach service for an urban university to offer. Providing limited dial-up Internet access, typically electronic mail made community networks more useful, and valuable to the public. However, Internet access had never been the core purpose of community networks. Indeed the Santa Monica PEN project resisted Internet connectivity in its efforts to retain a local character. Community networks are grounded on providing computer-based communication tools to local organizations and individuals, and local information content. Thus whereas the Internet by its nature tends to support processes that benefit virtual communities of interest, community networks by their nature tend to support processes that help people of differing viewpoints, qualities and values better coexist where they actually live. Divergence of Service Functions Community networks are built on current technology, market conditions and regulatory regimes. Modem banks and enough telephone lines to support an increasing local user base have always been a basic service function for a community network. Often, community networks offered the only way for the general public to access the Internet in any form, or to acquire government or other local information, electronically. Until recently, such dial-up service to the Internet was not offered by any other local provider. This is rapidly changing with the commercialization of the Internet and the growth of small, commercial ISPs. A small business can generally be viable with a minimum of about 1,000 accounts that pay between $25-35 a month. On this scale, such small businesses have been growing at a steady rate around the country. For a community network organization dedicated to providing local information content, managing growing banks of modems and telephone lines requires a disproportionate percentage of resources. Small ISPs however, are in the business of providing direct Internet access. Because the growth of small ISPs has been so recent, and because many community networks adhere to a free public access service philosophy, there is a growing issue of how both types of enterprises co-exist in the same local markets, particularly when one offers some levels of Internet access for free, while the other must charge. This becomes even more complex, of course, when non-profit community networks begin to charge below market rates for service. Case Study: Aztec AzTec, a Free-Net affiliate of the National Public Telecomputing Network in Tempe, AZ follows the classic model of many community networks established over the past ten years. It is supported through the University of Arizona, is managed by volunteers, and offers direct dial-up access to both local information content as well as basic Internet services such as electronic mail and gopher. Established in August, 1994, AzTec currently has about 7,000 subscribers. Tempe is in a large valley with a number of other cities with a total population within a local calling area of just over two million. There are 22 commercial ISPs for this market, with all but three of them being small, local businesses. Since the inception of these businesses, only one has so far failed. A synergy appears to be emerging between AzTec, and the local ISPs. Because AzTec is free to the user, it is attractive to many who have no working understanding of the Internet. According to AzTec board member Joe Askins, ISPs "view us as a feeder to get people a little pregnant on the Internet". AzTec provides text-based electronic mail, LYNX, no ftp, no telnet, no SLIP or PPP. Once many users discover the limitations of not having the full suite of Internet tools -- often becoming frustrated with the difficulty of downloading files -- local commercial services quickly become attractive and affordable. Of course, all of AzTec's local information content, much of it produced with well designed web pages, continues to be available through local commercial ISPs. Indeed, a good number of the local ISPs include pointers to AzTec for their users. In this scenario then, the free public access community network serves as pre-competitive market development for the commercial ISPs. As well, by highlighting links to AzTec, commercial providers add additional value to their services at no additional cost, by providing convenient access to useful local information content. This relationship suggests that AzTec will not need to make continual investments in increasing phone lines and modem banks because much of the access requirements will be peeled off by the commercial sector. However, the commercial sector, focusing on direct access, does not specialize in developing useful local information content. The two roles, access and content, would appear to be complementary. For the truly needy who cannot otherwise afford either a personal computer or a monthly service charge for commercial access, AzTec volunteers are acquiring used computers and installing them as public terminals in libraries, community centers and other facilities. Entry of Telephone Companies The total market for local Internet access has been estimated at close to one billion dollars a year, and growing. Until this year, the regional Bell operating companies (RBOCs) have not addressed this market, nor have the long distance services such as MCI or Sprint. All have expressed interest and several have either announced plans or appear close to doing so. These anticipated services are generally called "Internet dial-tone". The RBOCs cannot provide untarrifed access to the Internet at large, but can provide Internet protocol (IP) services such as SLIP or PPP. This service must then in turn, be connected to the Internet by way of an internet service provider such as PSI. The carrier then, can establish an IP gateway, through the local exchange, in which the user can connect to the Internet directly. A long distance carrier such as MCI has an especially good advantage in rural markets, if it can offer direct IP as a function of long distance, thus by-passing the local exchange carrier (LEC). Customers may choose their long distance carrier, and the LEC must provide an attachment to their local switch. A long distance carrier, through a simple touch-tone code generated by a modem dial-up sequence could provide a residential customer direct IP service, charging whatever it felt the market would bear. On the other hand, the local exchange carrier can provide direct IP through a frame relay service. This service essentially provides customers a single telephone number in which to access a specially configured private, or enterprise-wide high speed network. Both of these developments have implications within the growing market for both community networks and small commercial ISPs. Case-Study: CRISNY The Capitol Regional Information Service of New York (CRISNY) is an independent community network project operating with the University of New York, Albany serving as a sponsor and fiscal agent. The service will become available to the public in the fourth quarter of 1995. CRISNY, unlike AzTec will not be a free service but will charge a low subscription fee mixed with subsidies through business services such as web page design. Moreover, unlike AzTec which is operated entirely by volunteers, CRISNY already has a full-time paid staff of four. Like AzTec, CRISNY sees its primary role as a developer and provider of local information content. Unlike AzTec, CRISNY will not maintain dial-up service and will not require investments in modem banks and telephone lines. A negotiation with NYNEX is nearing completion that would provide CRISNY users direct access from within the local exchange through a frame relay service. NYNEX has indicated plans to roll-out their Internet dial-tone service in the fourth quarter. This relationship will not likely support the business interests of local ISPs, unlike AzTec which often suggests that users acquire accounts with such services. Internet Dial-Tone and the Changing Market In the current regulatory regime, Internet dial-tone would not be tarriffed because there is technically no protocol conversion involved in the service. It is simply a pipeline not involving any data processing or other value-added. For example, NYNEX can only provide TCP/IP service, it cannot itself provide access to the global Internet -- an information service. This is why CRISNY must contract to NYSERNET, PSI or another provider for direct Internet access from their terminal server, while NYNEX provides the dial-up capacity to that server, that would otherwise have to be available through a modem pool. Community networks have long offered dial-up access to their services, which were then often connected through university-based terminal servers, to the Internet. This was for a time the only way the general public could acquire any form of Internet access. Times changed, and a growing commercial market for local access emerged, particularly attractive to entrepreneurs interested in starting up a small business. These niches were based on serving small local markets with limited capacity. Neither small commercial internet providers, and certainly not community networks, are capable of serving a mass market, they are geared to local markets. However, managing large modem pools is not practical at a certain scale, even in many local markets. In many parts of the country, the market has reached the scale, where such service is becoming attractive to telephone and cable companies. If regulations change and LATA boundaries dissolve (the fairly arbitrary geographic local calling areas for many states), a corporation with local market dominance such as NYNEX could be well positioned to use frame relay technology to provide Internet dial-tone across its territory. This could quickly kill the small business market for IP access, and eliminate this growing source of local jobs and revenues. It is not clear that such a change would have an equivalent negative impact on community networks that focus on providing useful local information content. However what this suggests concerning competition or collaboration with local newspapers is an important issue in its own right, though beyond the scope of this paper to address in detail. Maintaining Level Playing Fields for Local Competition The issue here is how best to provide for access capacity in a manner that does not simply encourage unregulated monopoly and market dominance that stifles innovation and competition. In the case of AzTec, an informal relationship has grown between a community network specializing in local information content and 22 small businesses in the region that provide Internet access services. These services address a growing market presently not served by either the larger telecommunications industries or newspaper publishers. Whether this rapidly growing and fertile ground for innovation and invention in the information economy is going to be absorbed and vertically integrated into existing industries that can scale to larger consumer markets remains to be seen. In the case of CRISNY, and its partnership with NYNEX to provide frame relay access to its services, there appears to be no option for synergy that would assist in the development of small, local ISPs. It would however appear at the present time, that there is growing demand for both community networks and small scale Internet service providers across the country. There also appears sufficient danger that both types of enterprises may be absorbed into larger industries. It remains an open question for debate, whether these enterprises are simply serving a limited function during a curious historical moment. Can a combination of free-market economics and policies promoting competition provide a level playing field to promote the continued vitality, innovation and invention that these two types of enterprises bring to both the information economy and revitalization of local communities?