The Mitchell-Yancey Broadband Consortium

Executive Summary

The Mountain Area Information Network (MAIN) is proposing a "last mile" wireless network for community-wide, broadband Internet access in two adjacent rural counties of western North Carolina: Yancey and Mitchell. Grant funds requested from the N.C. Rural Internet Access Authority total $635,000. This project is entitled the Mitchell-Yancey Broadband Consortium. These counties have an economic-distress ranking of Tier 1 (Yancey) and Tier 2 (Mitchell). Both counties have experienced staggering losses of manufacturing jobs due to plant closings over the last five years.

Indeed, these job losses resulted in Yanceyís unemployment rate jumping from 3.9 to 12.7 percent, the greatest increase in the state, according to the N.C. Rural Center ("Manufacturing Layoffs: Hard Times for Rural Factories, Workers and Communities," April 2002). Similarly, Spruce Pine, the business and manufacturing heart of Mitchell County, saw manufacturing job layoffs equaling 30 percent of the townís population, according to a recent report by U.S. Sen. John Edwards ("Economic Relief for North Carolina," Spring 2002). Mitchell and Yancey are also included among the state's 13 "most underserved" counties for telecommunications, as identified by the RIAA (Connectivity Incentive Grants RFP, Section 8.0).

The lack of telecommunications services in these counties was aggravated last fall when the local cable companies pulled the plug on their high-speed, cable-modem Internet services. We are not aware of any other entity -- for-profit or nonprofit -- planning to bring broadband Internet access to these rural counties.

The Mitchell-Yancey Broadband Consortium will be deployed in collaboration with the local governments in each county, using municipally-owned towers and buildings to ensure maximum county-wide coverage. The network will be built upon MAIN's existing infrastructure. MAIN has been operating as a low-cost, dial-up Internet service provider (ISP) in these counties for six years. In 1996, MAIN provided the first public-access Internet service at the public libraries in Mitchell and Yancey counties. MAIN currently co-locates its servers on local government premises in each county.

The wireless network will be deployed in two phases. Phase 1 will provide broadband connnections to the Internet for county buildings and agencies, including schools, libraries, public-safety offices, etc. A total of 25 government "anchor" tenants are planned in each county. A few of these sites will receive access in a "barter" exchange for co-location access to the towers. Aggregating this public-sector demand will help ensure long-term sustainability and help "radiate demand" for broadband -- via county employees, teachers, etc -- into the larger community. Phase 2 will provide broadband connections to the general public. The total project implementation timetable is 18 months.

A special feature of the Mitchell-Yancey Broadband Consortium will be a wireless "Internet Free Zone" in the three main population centers of the two-county area: Burnsville, Bakersville, and Spruce Pine. Each Internet Free Zone will provide free broadband wireless connections to any computer-user with a wireless modem within a several-hundred yard radius of these town centers. In addition to putting these towns "on the map" as high-tech tourist and small-business re-location destinations, the Internet Free Zones will make a bold statement to stimulate broadband demand in these rural, economically-distressed counties.

In summary, the Mitchell-Yancey Broadband Consortium mirrors the RIAAís legislative mandate to ensure broadband Internet access -- with minimum speeds of 128 kps to the home and 256 kps to businesses -- in rural, underserved counties of North Carolina.

A Solid Foundation for MYBC

The Mountain Area Information Network (MAIN) is a 501(c)3 tax-exempt nonprofit community network serving 14 counties of western North Carolina with low-cost, dial-up Internet acccess, plus an array of public-interest, community-network services. Thirteen of MAINís 14 counties meet RIAAís definition of eligible, underserved counties. This proposal is specific to two of those counties: Yancey and Mitchell. MAIN began organizing its community network in 1993. In the fall of 1995, MAIN was awarded a demonstration grant from the Technology Opportunities Program (then known as TIIAP) of the U.S. Department of Commerce to bridge the "digital divide" in 12 mountain counties, including Mitchell and Yancey. At that time, citizens and businesses in most of these counties had to call long-distance to access the Internet.

MAINís goal -- first and foremost -- was to provide affordable dial-up Internet access with a local phone call in these counties. Another major community network goal was using the Internet to strengthen the experience of community by bringing together "affinity groups" -- citizens with specific interests in common, such as crafts or geneaology. MAINís focus on community-building, however, was quite different from that of "virtual community" adherents such as Howard Rheingold, who viewed the Internet as a communications technology destined to make geographical space and place irrelevant. By contrast, MAINís concept of a community network places geography at the heart of the network. Our goal has always been to use Internet technologies to build community within a common geographic and cultural sphere: the mountain communities of western North Carolina.

The following passage from our successful 1995 grant proposal to the U.S. Department of Commerce illustrates our intent to use the Internet to build community within our region, and addresses why this goal is so important:

"The much-publicized Sunbelt economic boom has largely

by-passed WNC's native mountain families. While other similarly impoverished regions of the South have prospered with Sunbelt

growth, our mountainous terrain continues to be a barrier to sustain-

able economic development. . . . The exclusion of WNC from these

economic benefits is also tied to the region's historic isolation from

North Carolina's "power centers" of commerce, capital, government,

and health/education/cultural resources. . . . Mountain Area

Information Network (MAIN) is designed to confront and help over-

come this isolation by leveraging existing resources to bring the

National Information Infrastructure to the region. . . . Isolation from the

state's "power centers" is aggravated by an "interior isolation" among the

communities of WNC. For example -- due to the rugged terrain and size

of the region -- it takes five hours to drive from Alleghany County in

the northeast corner of Western North Carolina to Cherokee County

in the far west. . . . MAIN will serve as a demonstration model

for cross-jurisdictional collaboration and cost-sharing. It will also

demonstrate the use of the NII for creating cross-generational and

cross-cultural "bridges" and for using the Internet to build community

within the mountain region. . . ." (MAINís 1995 TIIAP Grant Narrative).

One example of our successful community-building via the Internet involves the growing crisis in mountain air quality, which has galvanized concern among citizens and community leaders in the years since MAIN was founded. This crisis led to the passage last month of the Clean Smokestacks Act. MAIN played a leading role in raising public awareness about this critical economic, public health and environmental problem in numerous ways. For example, we have used our community news and information "portal" -- which attracts more than 10,000 unique visitors per day -- to provide background information and breaking news about the air-quality problem, as well as the political struggle to pass the Clean Smokestacks Act. We also worked with the Canary Coalition -- an association of mountain-area public health and environmental organizations -- to build and host their website. We then featured the website and the coalitionís events in our homepage "Spotlight" and in our community calendar. Given our high-traffic homepage, we have the ability to drive traffic to nonprofit, public-interest websites which have vital information for our community.

Finally, we have given a permanent place on our homepage to the U.S. Forest Serviceís "Mountain Web Cam" -- which looks out on Cold Mountain in the Shining Rock Wilderness. This web-cam was designed to help citizens compare current air-quality and visibililty with historic norms, when the mountain regionís superb air-quality was celebrated in the slogan, "Land of the Sky." Likewise, in the spring, summer and fall seasons, we provide a highly visible link on the MAIN homepage for the "Ozone Report" from the N.C. Division of Air Quality website. This link is especially vital to citizens with respiratory problems, for whom ozone concentrations pose serious health risks. MAIN is the only website in western North Carolina where all these links can be found together. And we are the only non-commercial website with such high levels of daily traffic, by which public-interest content can find a ready local audience.

By combining affordable Internet access with the delivery of in-depth local news and information, MAIN has emerged as one of the first and most successful nonprofit "demand-aggregation" community networks in the United States. Today, MAIN is completely self-sustaining via revenue from more than 5,000 dial-up subscribers and more than 250 webhosting clients, both for-profit and nonprofit. A MAIN dial-up account is $12.50 a month, or $150 a year. About 300 low-income citizens with disabilities receive free or reduced-fee dial-up access from MAIN. MAINís network infrastructure for all 14 counties we serve consists of seven T-1 circuits and 490 dial-up lines. A fundamental part of our community network mission is to provide affordable Internet access to rural citizens and businesses, no matter how remote or economically-distressed their community might be.

MAINís History in Wireless

MAIN is especially well-suited to implement a "last-mile" wireless broadband network in Mitchell and Yancey counties. MAIN has more than six years of wide-area network (WAN) experience in rural western North Carolina, including a three-year old wireless WAN serving rural Polk County. In early 1999, Polk County community leaders approached MAIN about extending our community network to meet critical Internet-access needs in their county. Those needs included a 16-workstation computer lab at the Polk campus of Isothermal Community College, which was linked to the Internet by an inadequate 56 kps (kilobits per second) circuit from the local phone company. A faster circuit was simply too expensive, and no alternative provider was available. Seeking an alternative provider outside the county was cost-prohibitive, as most of Polk County is in the odd dilemma of residing in the LATA (local access and transport area) of upstate South Carolina. Likewise, the Polk County Public Library had completely inadequate Internet access, especially for public access.

Coincidentally, in 1997, MAIN employed a frequency-hopping spread spectrum (FHSS) wireless "air bridge" in remote Graham County to link the Graham County Public Library (which housed MAINís local POP) with the Robbinsville facility of Tri-County Community College, where the countyís first post-secondary computer-lab had just been installed. This wireless link enabled high-speed Internet access at both of these critical public-access facilities. The wireless link had the additional benefit of creating a cost-sharing partnership among the library, the community college and MAIN. This partnership was extremely critical for Graham County, which is arguably the most remote in North Carolina and, historically, one of the stateís poorest. This remoteness, not surprisingly, results in alarmingly expensive bandwidth due to the distance-sensitive "local loop" charges. By aggregating demand and sharing the costs, this partnership made the expensive bandwidth feasible.

Based on this experience, MAIN began exploring a similar wireless solution for our Polk County partners. Unfortunately, unlike Graham County, where the library and the college were only 600 feet apart with unobstructed line-of-sight, the ICC-Polk campus and Polk library are more than a mile apart and line-of-sight is obstructed by hills and ridges. Fortunately, one of the countyís tallest mountains, Tryon Peak, has a wealth of telecommunications towers, one of which, we discovered, is owned by one of the countyís municipalities, the Town of Tryon. With a guarantee of a free broadband Internet connection, the town granted MAIN access to its tower, which has unobstructed line-of-sight views of the college, the library, the town hall, and most of Polk County.

Access to the townís tower enabled MAIN to do the following: lease a landline T-1 circuit to the ICC-Polk campus, send a broadband radio signal to a relay on Tryon Peak, and thereby deliver broadband Internet access to the library and town hall. Most importantly, this wireless WAN enabled the creation of a cost-sharing partnership among the college, the library and MAIN, making feasible a very expensive T-1 circuit that no single partner could afford on their own. This wireless WAN and cost-sharing partnership continues today.

The technical and partnership-building experience gained in Graham and Polk counties gives MAIN a solid foundation on which to build the Mitchell-Yancey Broadband Consortium. In short, MAIN has the necessary technical experience and expertise to implement an expanded version of the Polk County wireless WAN in Mitchell and Yancey counties. In addition, MAIN has major advantages in Mitchell and Yancey which we did not have in Polk. For example, MAIN had no network or social infrastructure in Polk County prior to our being approached in 1999 by Polk community leaders. By contrast, MAIN has been operating in Mitchell and Yancey since 1996; and over the last six years, we have built substantial network and social infrastructure in these counties.

For example, with support from our TOP grant, MAIN purchased computers, leased phone lines, and provided the first public-access to the Internet in Mitchell and Yancey via our partnerships with the public libraries in these counties. MAIN has also offered various levels of free Internet training in these counties over the last six years. Today, MAIN serves 637 dial-up subscribers in Mitchell and 585 in Yancey, with a renewal rate of 90 percent. In addition, MAIN hosts 31 websites for citizens, businesses, and nonprofits in these two counties. Nonprofit websites include the Yancey County government, the Town of Burnsville, Yancey Historic Association, the Mitchell County Senior Center, the Penland School of Crafts, the Mitchell News-Journal, and Mitchell-Yancey Habitat for Humanity.

In addition, MAINís current network infrastructure in Mitchell and Yancey is quite robust. Each county has its own dedicated landline T-1 circuit to the Internet, which is linked to a digital 56 kps modem-server. (Of course, actual dial-up speed is ultimately determined by the subscriberís modem and the quality of of their business or residential phone line.) MAIN currently co-locates in Yancey County at the Burnsville Town Hall (where we helped the town activate a LAN) and in Mitchell at the countyís 911 call-center in Bakersville. Our technical analysis indicates that the addition of a second landline T-1 circuit will be required in each county during the second or third quarter of this project. A long-term prospect which we are exploring with the assistance of the N.C. Center for Geographic Information and Analysis (NC-CGIA) is a wireless "backhaul" to Asheville of all Internet traffic in Mitchell and Yancey, a move which would allow us to drop the more expensive landline T-1s in favor of cheaper backbone Internet access at our Asheville POP. Given the still-speculative nature of this option, it is not factored into our RIAA workplan or budget.

Combining MAINís existing network infrastructure with a wireless "last-mile" build-out using existing, locally-owned towers will create a solid and robust "end-to-end" broadband solution for these historically underserved counties. Moreover, what the RIAA learns from the Mitchell-Yancey Broadband Consortium will be invaluable in helping other rural areas of North Carolina overcome obstacles to obtaining affordable broadband Internet access .

 

The Technical Plan

At first glance, using line-of-sight wireless technology in the mountains may seem foolish. However, as we discovered in Polk County, wireless transmitters and relays located on mountaintop towers can provide extraordinary line-of-sight opportunities. The Mitchell-Yancey Broadband Consortium will be built across existing municipally-owned telecommunications towers in these two counties. We have access to a total of five towers, plus one water tank. One of these towers is a U.S. Forest Service fire tower. Staff from U.S. Sen. John Edwardsí office have already secured tentative approval for its use if we need it. Mitchell and Yancey counties together cover 533 square miles, and the towers provide multiple line-of-sight opportunities to optimize our coverage of the populated areas of the counties. Moreover, since several of these towers can see each other, we have multiple opportunities for building redundancy into the wireless WAN.

Our methodology for researching and establishing the feasibility of line-of-sight, wireless deployment was based on latitude-longitude coordinates of the towers and "geo-targets," i.e., the specific buildings and agencies (such as schools, libraries, etc) identified by the county governments and the "e-communities" steering committees as having the highest potential or greatest need for affordable broadband access. Using GIS tools and the street addresses of our more than 1,200 MAIN subscribers in Mitchell and Yancey, we also mapped the distribution of those subscribers in the two-county area. We then took the tower and geo-target coordinates, and the subscriber maps, to two GIS consultants: Neil Thomas of Research Data Inc. and Mark Phillips of Geometry Technologies (both are long-time MAIN supporters, and much of their time and expertise for this feasibility study was donated).

Neil Thomas used ArcView, a proprietary mapping software, to project the "footprint" of wireless coverage from the various tower locations. He then overlaid this footprint on a map of the geo-targets (the county-government "anchor" sites plus MAINís subscriber-base). Mark Phillips conducted a similar analysis using an open-source GIS software from the University of Minnesota called MapServer. Together, the analyses concluded that, theoretically, we can directly reach virtually all of the 50 priority anchor sites with a line-of-sight wireless signal.

This type of GIS analysis -- often called a "viewshed" study -- is routinely used by wireless firms in preliminary planning. In fact, Research Data Inc. has conducted several such "viewshed" studies for commercial wireless firms in the Southeast. Of course, a computer-generated analysis is no substitute for field-testing for line-of-sight.

We know from experience, for example, that tree-cover can obstruct an otherwise clear line-of-sight path. In such cases, several options exist for overcoming the impediment. First, an inexpensive but sturdy mast can be attached to the building in order to achieve clearance of the obstruction. Second, existing and emerging wireless technolgies are able to "punch through" walls and attics in the 1-2 mile range and through leaf-cover in the 3-5 mile range. This wireless technology is called "near line-of-sight." Wave-Rider is just one manufacturer of this more robust form of wireless.

Finally, as a last-resort, we can make a "dog-leg" relay from a building or facility that has both line-of-sight to the tower and line-of-sight to the otherwise obstructed target. Indeed, we are planning on this option for what we believe will be a handful of problem sites. A similar last-resort option is to negotiate access to a privately-owned cell tower which does have line-of-sight, or could provide one leg in a dog-leg path to a priority geo-target or remote community of subscribers. Our county government partners believe that free access could be negotiated, as the cell-phone companies will want to remain on good terms with the county staff and commissioners as they consider future cell-tower needs.

Our "viewshed" analysis also indicates that each of the 50 or so anchor sites are no more than 5-7 miles away from a tower, while some anchor sites are as close as a few hundred yards to a half-mile. These distances fall well within the range of the various unlicensed wireless technologies, both currently available and emerging, that we have examined and could employ. These wireless technologies include the following types: 900 MHz, 2.4 GHz, 5.2-5.3 GHz, and 5.8 GHz. The 5.8 Ghz technology will most likely be employed as our "backbone" or "backhaul" technology to ensure redundancy and to provide a robust, secure platform for the multipoint, client-side technology, which will likely come from one or more of the aforementioned lower-band, wireless technologies.

We have looked at several theoretical network designs based on the tower locations relative to the geo-targets in the two-county service area. Various designs are possible using omni-directional antennae, sector-antennae, or some combination of both. On-site analysis is essential before any detailed network design work can be undertaken. If the Mitchell-Yancey Broadband Consortium is funded by RIAA, we will issue a request for proposal (RFP) for a network engineering firm with a strong background in wireless.

Our research and planning for this grant proposal have already identified five firms which would be competitive candidates for this project. Four of these firms are in North Carolina, and two of those are in western North Carolina. One of these firms is already experienced with several of the Mitchell-Yancey towers. At least four of the firms appear to meet the RIAAís definition of "small and historically underutilized firms."

The firm we hire would be responsible for conducting on-site analysis of the towers and geo-targets and for designing a wireless WAN that is compatible with MAINís existing network and scaleable for future growth. The firm would also provide training to MAIN staff and volunteers in maintaining the network and in connecting subscribers to the network. The wireless WAN will be designed to enable MAIN staff to maintain, operate and upgrade the network.

Another important aspect of our plan is the customer-premise equipment (CPE). We believe our project will spur demand for broadband from both residential and business customers by having the RIAA grant cover the CPE costs for the 18-month grant period. Total grant funds earmarked for CPE costs is $166,850. The CPE units we are considering cost $500-$600 each. Though this hardware may sound pricey, it has important features that will save money and bandwidth in the long run. For example, this CPE technology is called "near line-of-sight," which means -- as we stated earlier -- it can deliver a solid signal through leaf-cover or into an attic. These are especially critical features given the topography and demographics of these rural mountain counties. The indoor version of this CPE is also attractive because it eliminates the need for outdoor mounting and cabling, making for a much cheaper and quicker installation. Most importantly, this higher-cost CPE allows "rate-management" to be set at the customer-site, which eliminates the need for rate-managing routers at each tower.

Efficient rate-management is a critical factor in assuring the highest quality of service. We know from experience that without effective rate-management protocols, one or two wireless subscribers can "soak" your bandwidth, thereby reducing overall performance across the WAN. Our technical analysis also finds that rate-management at the customer site will make the wireless WAN much easier to manage, thereby conserving precious network staff resources.

In short, by having the RIAA grant fund all the CPE costs for the 18-month grant period, we create a very strong incentive for residential and business customers to sign up for broadband BEFORE the grant period ends. After the grant funds are exhausted, we may have to recoup some portion of the CPE costs from the customer or find some other source of subsidy.

Another advantage of implementing rate-management on the client-side is that it allows for more efficient use of bandwidth, which in any rural network is the scarce and cost-prohibitive resource. The nonprofit Buffalo Wireless Internet Group (www.bwig.net) has been very satisfied with this approach, and is planning to convert its entire network to this client-side, rate-management technology. Similarly, the nonprofit Runestone Electric Association in Minnesota uses this approach to operate a wireless WAN which serves 250 customers -- business and residential -- over two T-1 lines (http://home.alexweb.net). Putting that many broadband customers on two T-1 lines sounds implausible, but the engineers at both BWIG and REA assert that the "intelligence" of the network -- plus natural load-balancing between daytime business-users and evening residential users -- makes it possible. Based on this research, we are planning our wireless WAN on the basis of one T-1 line for every 100 broadband customers.

Our current plans are to charge $30 a month for 128 kps for residential, and $50 a month for 256 kps for businesses. We will also offer faster speeds for "super-users" at rates still to be determined. This basic monthly fee-schedule -- and the broadband adoption rates on which our workplan and budget are based -- were determined in part by a survey of our dial-up subscribers in Mitchell and Yancey counties. That survey, which was answered by 408 respondents -- about 33 percent of our subscribers in Mitchell and Yancey -- reveals that 85 percent of the respondents consider "slowness" to be a "common" problem with their dial-up connection. Meanwhile, a whopping 85 percent stated that dial-up was an "inadequate" means of access for their "future needs."

These responses indicate a strong basis for developing widespread demand for broadband. However, given the high levels of unemployment and economic distress in these counties, it is essential that we keep these basic monthly rates at $30 and $50, respectively.

The projectís scaleability is assured by several aspects of the technical plan. First, the wireless WAN will be built upon Cisco 3640 core routers. These modular routers are scaleable to DS-3 (T-3) capacity, which is well within our five-year projection. MAIN built its demand-aggregation network in 1996 around 10 Cisco 2501 routers, and we have upgraded to more high-performance Cisco routers over the past six years. As a result, we have a very high level of confidence in the capability of the Cisco 3640 router and its scaleable capacity. Likewise, the capacity of the radio transmitters we are considering can be expanded by "firmware" and user-license upgrades to levels well within our five-year projections. Furthermore, we believe the geographic spread of the towers over this two-county area (and the fact that three of the key towers can see each other -- give us a mulititude of design opportunities to ensure scaleability well into the future.

The Internet Free Zones

Though MAINís broadband wireless WAN project is clearly a "last-mile" supply-side effort, it will also have a major "demand-side" aspect called the "Internet Free Zone." The IFZ is designed to spur demand for broadband by giving citizens a "free taste" of broadband Internet access within a 300-foot radius in three key town-centers. Those towns are: Burnsville (county seat of Yancey), Bakersville (county seat of Mitchell) and Spruce Pine (primary business center for Mitchell). The IFZs in these towns will be created by the deployment of "Wi-Fi," a fast-growing form of 802.11b wireless technology that is quick, easy and cheap to deploy. "Wi-Fi" is an abbreviation for "Wireless Fidelity."

Hereís what a recent Wall Street Journal report had to say about the growing popularity of this wireless technology: "Companies from Starbucks to chain hotels are making a push to set-up local Ďhot spotsí for Wi-Fi and charging a fee for access. . . . And technology enthusiasts are setting up competing, free community networks by sharing their home Internet connections with the neighborhood." ("Cordless Computers Get Affordable," WSJ, 26 June 2002, page D-1).

These Internet Free Zones will give local citizens and visitors to Mitchell and Yancey counties a free taste of broadband access and generate curiosity and interest about broadband in general. The IFZs will also create a "buzz" in the Southeastís tourism and economic development world that Burnsville, Spruce Pine and Bakersville have emerged as "cutting-edge," high-tech-friendly towns that are prime re-location destinations for lone-eagle entrepreneurs.

Given the relatively modest cost for setting up broadband "hot spots," the Internet Free Zone concept has the potential to spread quickly well beyond Mitchell and Yancey to other mountain communities, thereby spurring even more awareness of -- and demand for -- broadband Internet access. We believe the Internet Free Zone concept is a "canít miss" tool for promoting economic development and spurring demand for broadband in these underserved counties.

Project Budget and Sustainability

MAIN is seeking total RIAA funding of $635,000 for the Mitchell-Yancey Broadband Consortium over 18 months. The budget categories and totals for the requested funds from RIAA are:

Wireless WAN Design and Installation: $305,015

Customer Premise Equipment (CPE): $166,850

Connectivity: $50,500

Network Operations: $107,635

Total RIAA Funds Requested: $635,000

RIAA support for MYBC is critical. With local, state and federal governments running deficits, the likelihood of government support for this project is nil. As a nonprofit, grassroots-supported community network, MAIN is fortunate to be self-sufficient from revenue generated by our dial-up subscribers and hosting clients. However, we do not have access to the kind of capital and operating support required to launch a project on the scale of the Mitchell-Yancey Broadband Consortium. With the MYBC, the Connectivity Incentives Program of the N.C. Rural Internet Access Authority has an historic opportunity to build a sustainable rural broadband network that can serve as a powerful prototype for other underserved rural counties.

The projected long-term sustainability of the Mitchell-Yancey Broadband Consortium is based on several revenue-generating factors, plus an expected drop in bandwidth costs in year three of the project. We currently pay $950 a month for each T-1 line to Mitchell and to Yancey, with no installation fees. This is a rate we have negotiated with our long-time backbone supplier, ITC-Deltacom, which has also generously donated network-engineering consulting services over the years. We expect little or no change in this rate over the first two years of the project, and we thus have budgeted on this basis. However, in year three, we are budgeting based on a monthly T-1 charge of $500.

The basis for this projected reduction in bandwidth costs is the ramping-up of the Internet-backbone project spearheaded by U.S. Congressman Charles Taylor and his Education Research Consortium (ERC), which announced earlier this year a federal appropriation of almost $15 million for this effort. Itís conceivable that the ERC project could ramp-up sooner, but we chose to be conservative and assume that it will be operational some 30 months from the time of this writing. This projectís failure to become operational in the next 2-3 years does pose some risk, as our current projection of sustainability is tied to a reduction in backhaul costs for bandwidth.

On the revenue side, our sustainability is based on three key revenue-streams in the two-county area: 1. Anchor Tenants -- approximately 50 county government agencies and schools; 2. Super-Users -- 10 commercial and 5 nonprofit users with bandwidth needs from 512 kps to 1 mps; 3. Business/Residential Users -- 290 users by the end of the 18-month grant period. Our five-year budget projection shows a break-even situation for the first four years of the project. By year five, we should be generating net income of $33,765. All income from the Mitchell-Yancey Broadband Consortium will be used for maintenance and upgrade for the wireless WAN.

The MYBC will leverage in-kind contributions from MAIN and from the respective county governments in Mitchell and Yancey. Approximately 25 percent of MAINís annual operating revenue is devoted to the Mitchell-Yancey Broadband Consortium as in-kind matching funds. This calculation is based on the fact that dial-up subscribers and hosting clients in Mitchell and Yancey generate about 25 percent of MAINís annual operating revenue. In fiscal years 2003-04 and 2004-05, for example, this in-kind support totals $216,719 and $240,116, respectively (see MAIN pro forma budget, 2003-2007 attached).

Our county government partners are contributing in-kind support by waiving the co-location fees for access to municipally-owned towers and storage space for MAINís network POPs at the Burnsville Town Hall and the Mitchell County 911 call-center. This support totals approximately $57,000 for the 18-month grant term, based on a standard commercial co-location rate of $400 a month.

It should also be noted that the value of all intellectual property resulting from this project will reside in the public-domain. MAIN is firmly committed to open-source technology and to serving as a model for other nonprofit, public-interest ventures. Likewise, there are no plans for this project to convert to a for-profit venture. Finally, MAIN is in good financial standing, as certified by our attached budget documents and audit letter. All payables are current.

Key Project Personnel

Wally Bowen is executive director and founder of MAIN, which was conceived in 1993 and launched two years later with funding via a grant from the U.S. Department of Commerce. Bowen has been working full-time on public-interest, media-access projects since 1991, when he founded the nonprofit Citizens for Media Literacy. MAIN was launched as a demand-aggregation project of CML to promote citizenship and citizen-access to the media via the Internet. Bowen has spoken on the links between citizenship, community-building and media-access at the Aspen Institute, Harvardís Graduate School of Education, the National Association of Media Education, the New Mexico Media Literacy Project, and "Networks for People," the annual conference of the Technology Opportunities Program of the U.S. Department of Commerce. He is a strong advocate for using demand-aggregation strategies plus leveraging the value of public assets and public right-of-ways to bring affordable Internet access to rural areas.

Robbie Robinson is a 16-year veteran in network administration, and he has been MAINís network administrator since 1997. He is responsible for maintaining MAINís wide-area network, Internet connectivity, routers, remote-access servers, and Linux servers. He has helped MAIN transition from a single T-1 line and analog dial-up capability to a 14-county network with seven T-1s for Internet connectivity and 490 digital dial-up lines. Robinson is also MAINís point-person for network issues related to the Polk County wireless partnership with Isothermal Community College, Polk County Library and the Town of Tryon. In addition to extensive experience working with local telcos and inter-exchange carriers, he is also proficient in Cisco routing, Linux OS, remote-access server configuration and maintenance, C-programming, and scripting. He has also been instrumental in supporting and maintaining MAINís public-access terminals.

Trained as a mathematician, Mark Phillips is a 20-year veteran in information technology. He currently owns and operates Geometry Technologies

(www.geomtech.com), a web-applications consulting firm specializing in visual and GIS applications based in Asheville, N.C. Since moving to Asheville in 1999, Phillips has become a valuable member of MAINís volunteer Technical Advisory Committee (TAC). Before launching Geometry Technologies, he worked at the

University of Minnesota's Geometry Center, where he helped develop the

well-known Geometry Center website (www.geom.umn.edu), one of the earliest World Wide Web (1993) interactive websites. He also led a team of five programmers in a three-year effort to design and develop "Geomview", an interactive 3-D visualization package (http://www.geomview.org). A graduate of Davidson College, he also holds an M.S. degree from Rensselaer Polytechnic Institute and a doctorate in mathematics from the University of Maryland.

Neil Thomas is founder and owner of Research Data Inc., a GIS-consulting firm in Asheville, N.C. RDI provides geographic information analysis, computer mapping and research services for corporate and government planners, engineers, natural resource managers and other professionals who need to visualize geographic and demographic patterns and relationships. Thomas has conducted "viewshed" analyses for wireless firms in North Carolina and Wisconsin. He is a graduate of Warren Wilson College and holds a masterís degree in environmental management from Duke University. He is also a long-time MAIN subscriber and volunteer member of the community networkís Technical Advisory Committee (TAC).

Tim Sanders is a nationally-known wireless consultant, writer and regular contributor to Broadband Wireless magazine. He is the founder of The Final Mile: Fixed Wireless Consultants located in Asheville, N.C. (www.thefinalmile.net). His wireless ISP clients include PrairieInet, the largest fixed-band, unlicensed wireless provider in the United States (www.p-inet.com); Iron, an ISP in Santos, Brazil (www.iron.com.br); Z-Wave of Raleigh, N.C. (www.z-wave.net); Internet Blue Ridge of Rutherfordton, N.C. (www.bluestreakwireless.com); and SkyRunner.Net of Asheville, N.C. (www.skyrunner.net). His articles include "Rural Broadband Wireless: A Business Model That Works," "Wireless Start-ups in the Unlicensed Band," and "The Art of Thinking Small: How to Market Fixed Broadband Wireless Services in Overlooked Smaller Communities." Sanders is the former general manager of the Asheville-based SkyRunner.Net, where he was responsible for rolling out the companyís first multi-state ventures. END