HomeMy WebLinkAbout01 22 2013 EDA 7B Fiber Optic Network O .� PRIp
N � u 4646 Dakota. Street SE
Prior Lake. MN 55372
'�nvsso'��'
ECONOMIC DEVELOPMENT AUTHORITY AGENDA REPORT
MEETING DATE: January 22, 2073
AGENDA #: 7B
PREPARED BY: Dan Rogness, Community & Economic Development Director
AGENDA ITEM: FIBER OPTIC NETWORK DISCUSSION — DESIGN NINE, INC. REPORT
DISCUSSION: Introduction
Design Nine, Inc. was hired by the EDA to complete an independent evaluation
of the Broadband Advisory Committee's (BAC) fiber network feability report.
Andrew Cohill of Design Nine made a presentation to the EDA on December 17,
2012, and a written report was submitted by the end of December.
His tON
A Fiber Optic Network Feasibility report dated June 2012 was submitted to the
EDA for its review and deliberation by the BAC. Subsequently, the EDA began
to have meetings with the three local broadband service providers. The City
Council requested that a meeting be organized as a key next step in the fiber
deployment evaluation process. The Broadband Advisory Committee's recom-
mendation is for the city to get into the business of providing Triple Play services
in order to pay for the full fiber build-out cost. In doing so, the city would com-
pete against the incumbent service providers, including Integra, CenturyLink and
Mediacom. Another subsequent action by the EDA was to have an independent
analysis done of the report due to the large public investment necessary�to im-
plement the BAC's recommendation. ,
Conclusion �
The Design Nine report includes four recommendations:
1. Clarify the city's long-term community and economic development goals
in order to align the proposed fiber network investment with those goals.
2. Further evaluate the technical design to ensure that the nefinrork architec-
ture also supports the city's economic goals.
3. Reconsider investment in television and telephone services due to recent
technological changes, along with the added benefit of avoiding the cost
of local technical staff.
4. Evaluate the risk of receiving one or more lawsuits from incumbent pro-
viders by choosing the Hybrid Model; consider the impact of having a
65% local referendum in order to proceed.
ISSUES: Design Nine was hired by the EDA to complete an independent evaluation of the
BAC's fiber network feability report. Eric Lampland of Lookout Point Communi-
cations has provided his preliminary feedback on Andrew Cohill's report (see
attached comments from Lampland). The EDA now has hired finro consultants
with differing approaches on how to best pursue fiber-to-the-premise throughout
Prior Lake.
In addition, the City Council approved a resolution on December 10 authorizing a
fiber optic cooperative agreement between the city and Scott County. This ac-
tion begins the process to construct a fiber optic system serving city buildings
and four business parks. A request for proposals (RFP) is currently being writ-
ten in order to obtain information from potential broadband service providers,
including costs paid by those providers as partners working with the city and
county.
FINANCIAL Various multi-million dollar cost options have been provided to the EDA from
IMPACT: both reports depending on the model selected for full or partial build-out of the
fiber network. The estimated cost assocated with the cooperative agreement is
approximately $320,000 to be shared befinreen the city, county and broadband
service provider.
ALTERNATIVES: 1. Request that the BAC meet again in order to review the Design Nine re-
port and retum with any further changes to its recommendation.
2. Request that staff continue to schedule one-on-one meetings with Centu-
ryLink and Mediacom.
3. Delay further action until the city has received and evaluated responses
to the RFP related to the city-county cooperative agreement.
RECOMMENDED Staff recommends Alternative #3
MOTION:
ATTACHMENTS: 1. Design Nine Report
2. Response from Eric Lampland
2
., ... 1'�ii ii ��iui�l� Ao � �
+�...�-:. ,
�
t
_:
�--� .�.
��--�—
� �
xS�;qNa^X .:.. _..., � . '� .
. ':`� .',�..
. �S.f�
�
��p'.
;f ,
, ,;� � ���>
, .=:.%= � ��,. �" } ��x�` b
„ '� � °"$ „ ,,� ,,
�. > � , ��� �
��� „ , ,. � �,
re -� � �f
� �� ,; ru� a �� � i
4 � � /ku� � �M � :r
fx ,
3 • �a*r..
a ;' � � � ...:
� �,..�, j�� �
��.
�'' ' �
�
�
�,� , ,,
���, ... � � . ..
� �
��,:
' �� : ;,.
Prior Lake Broadband Review and Next Steps
DECEMBER,2OI2
.
- DESIGN NINE
� we build networks that perform
,
Contents
Overview of the Work I
Technical Design 3
Overview of a Modern Broadband Network .............................................................................................4
CoreNetwork .......................•--..................................................................................................................................5
Disuibution ................................................................................................................................................5
AccessNetwork ................................................................................................................................••-•-••••-•..............5
Colocation Facilities ...................................................................................................................................................5
Bacl�aul ........................................................................................................................................................................6
N and Phone Analysis 7
Cost Analysis 8
Legal and Organizational Analysis 9
OwnershipOptions ..........................................................................................................................................9
Municipal Ownership .................................................................................................................................................9
Non-profit .................................................................................................................................................................. I 0
Ad Hoc and Informal Partnerships :......................................................................................................................1 I
For profit businesses ................................................................................................................................................1 I
Business Model Analysis I 2
PaimCoast, Florida .........................................................................................................................................12
Danville,Virginia ............................................................................................................................................... I 4
FarmersTelecom .................................................................................................................................15
Lafayette, Louisiana .........................................................................................................................................16
Next Steps 17
Qverview of the Work
Design Nine was asked by the City of Prior Lake to review the F'iber Optic Network Feasibility
study completed by the Broadband Advisory Committee. We agree with the committee's
conclusion that a fiber network in Prior Lake is financially feasible, and our own independent
financial analysis suggests that it could, over time, not only generate enough revenue to pay for
the initial cost of construcrion but could also provide a modest stream of new revenue for the
City over the long term.
We offer the following summary of our recommendations:
� Before making a final decision, the City should idenrify the long term community
economic and development goals that would be supported by an investment in a fiber
network. For example, if a primary economic objective is to diversify the kind and
type of businesses in the downtown area, a more modest effort may meet that goal at
substantially less cost. On the other hand, if attracting more work from home and
business from home residents to the City is a strategic objecrive, then a full build out
may well be justified.
1 We would encourage further evaluation of the technical design to ensure that the =
network architecture also supports the City's economic objectives. If new job and
business development is a key goal for the City, a carrier class Active Ethernet net-
work capable of delivering very high performance business dass services may be a bet-
ter option.
� The market for TV and telephone services is undergoing major changes. The TV
market is moving away from traditional packages of TV service and towards a wide
variety of video streamed over the Internet rather than delivered via RF (on coaxial
cable or PON fiber). 1he City has the option of not investing in a TV head and in-
stead offering an IP TV offering purchased wholesale from an exisring IP TV pro-
vider. The same approach is possible with telephone service. Rather than purchasing a
telephone switch, the City could retail VoIP (Voice over IP) telephone services via a
wholesale contract with an exisring VoIP provider (there are dozens). By purchasing
TV and telephone wholesale, the City also avoids the very significant cost of technical
staff that would be needed to maintain a TV head end and a telephone switch.
Page I of 17
� The City may wish evaluate the risk of a lawsuit resulting from choosing the Hybrid
Retail model, along with the significant requirement of ineeting the 65°ib affirmative
referendum vote that would be required with the Hybrid Retail model. Other options
that suggest additional evaluation would be a simple dark fiber leasing model or an
open access model. In both of those business models, the City would NOT be offer-
ing any retail services, and a referendum would not be required, and the risk of a law-
suit is sharply reduced.
Page 2 of 17
Technical Design
1he proposed PON architecture delivers TV as a separate RF channel on the fiber; this
approach is functionally the same as the exisring cable system. Current DOCSIS-based cable
TV is nearing a dead end as comperirion from IP streaming services like Hulu, Netflix, and
many other emerging Internet-only video services are offering a wider array of oprions at lower
cost. For the first time in decades, cable TV companies are seeing negative growth, with
younger people uninterested in legacyTV cable offerings, and many other households forgoing
the high cost of cable TV service because of the poor economy. It is our opinion that the City
should carefully evaluate the proposed investment in a TV head end.
If the City moves ahead with the proposed hybrid retail model, an alternative for TV service
would be to sign a wholesale reselling agreement with an existing IP TV provider, rather than
purchasing a head end. Not only would the City then forgo the capital expense of several
million dollars, but would also forgo the requisite City staff required to administer the head end
over time. In an open access architecture, many of the expensive components of the retail
model are not required (e.g. PSTN gateway, TV head-end, subscriber services (e-mail, hosting,
etc.)) and are left to the private sector.
We would note that the PON architecture supports basic business services (mostly Internet)
but can not offer carrier class high end business services like Transparent LAN (e.g. private
VPN) services. This may be important if the City wishes to attract more high tech businesses
into the community. As an alternative, it may be useful to evaluate an active Ethernet
architecture, which can provide true Gigabit connectivity to both homes and businesses. If an
open access business model was adopted (i.e. no TV head end or phone switch required), a
carrier grade acrive Ethernet network would likely be less expensive while offering much more �
flexibility in providing business class services anywhere in the City. If the City wishes to attract
more work from and business from home residents, being able to deliver business class services
to any locarion could be important.
Page 3 of 17
OVERVIEW OF A MODERN BROADBAND NETWORK
CPE (Customer wireless � �
Premise Equipment) -�- � _ _ —C
�__--
Access fiber
Tower-mount
radio/amennas
Fiber or wireless core network
Neighborhood with backbone routers
cabinet or hut
Interne � # ��
Colo Facility
Core network
Distribution
Backhau routes fiber
�
�..
Colo has multiple service Colo comains
providers offering multiple network
services, with single point management
of access to all customers. servers. ,�
Access fiber
Neighborhood
cabinet or hut
Backhaul routes
� ..,
Connections to other networks via leased line or
middle mile fiber �pE Homes and businesses
Page 4 of 17
CORE NETWORK
The core network is often referred to as the "backbone" network. It is a high capacity route or
set of routes throughout a community or region that provides transport between towns,
neighborhoods, business districts, and other major facilities.
Ideally, the core network is designed as a redundant fiber ring, which provides both capacity
and gives the network the ability to continue operating even if the fiber is cut or damaged in
one location. A fully redundant ring can be expensive to construct, and so the "ring" feature
may be a long term design goal.
DISTRIBUTION NETVNORK
Distribution networks are connected to the core network, and provide primary network paths
through a town, neighborhood or business district. Distribution networks are generally part of
what is called "middle mile." Fiber-based distribution networks generally are built along most
streets and roads, and can be aerial fiber (mounted on utility poles) or underground fiber
(installed in underground duct or fiber cable that is buried directly without duct).
1he distribution network connects the core network (the network backbone) with the
individual connections within a neighborhood or business district that connect to home and
businesses. lhis portion of the network can be fiber-based or wireless, but fiber will be required
over the long term to support video services and other kinds of high bandwidth applications
like telemedicine, IP TV, business videoconferencing, and other emerging services.
ACCESS NETWORK
'Ihe access network is what is commonly called "the last mile," although "the first mile" might
be more appropriate, since customers should be a primary co�tsideration when designing a
network.
The access network is a direct fiber link between a fiber switch located within a neighborhood
or business district, or it may also be a direct point to point wireless link from a wireless access
point on a tower or building and the home or business. Network subscribers have to have
Customer Premises Equipment (CPE) to get a network connecrion, and this is simply a small
box that looks like a hub or switch. In a fiber network, the fiber cable is connected to one port,
and one or more copper Ethernet RJ45 ports allow users to connect computers, phones, and
TV set top boxes to it. In a wireless portion of the network, a small box with a radio and
possibly an external antenna is mounted on a side of the home or business with clear line of
sight to a nearby tower or building where the access radio is mounted.
COLOCATION FACILITIES
A colocation facility is a controlled environment (i.e. heated and air-condirioned) room with
Internet access through wired and/or wireless systems. 1he colocation facility will be a place
Page 5 of 17
where fiber, wireless, and copper-based network facilities meet. It will be equipped to house
high-end network equipment, servers, and other electronic gear. A variety of middle layer
network components and services can be located within the colo including, for example,
directory services, replicated content servers, rouring services, and other elements needed to
deliver new mulrimedia services to the home and small office from multiple, competing
providers.
Characteristics of a colocation facility include:
� A reliable source of AC electric power is required, with backup UPS (Uninterruptible
Power Supply) service available by an onsite generator.
� Controlled access to the facility (e.g. by electronic keycard) 24 hours/day, seven days a
week.
� Racks for locating network equipment and servers, and optionally locked cages for
equipment racks.
Functions of colocation facilities include:
� Hub for new broadband infrastructure development for the community.
� Location for a regional and community network exchange point for local service pro-
viders. Also called a peering point or inter-exchange point, this kind of facility can
reduce costs and increase performance in a win-win-win scenario (because it helps
keep local traffic local and reduces service provider costs, thereby reducing the price of
services). In Prior Lake, a modest colo facility would provide the meet point for pri-
vate fiber and any City-owned duct/fiber.
� Insertion point for multimedia services from multiple competing providers to reach
subscribers over single broadband medium (fiber, wireless, other).
� Community, campus, or building point of presence for new middle layer components
required to implement next generation Internet (directory services, caching, routin�.
� Focal point for technical resources and management of community infrastructure.
� Aggregation point for low cost access to gigabit scale network services.
BACKHAUL
Every community network requires one or more paths (routes) out of the community to carry
voice, video, and data traffic (in both directions). Backhaul network connections can be
purchased from a local incumbent telephone company. In some communities, there may be
other backhaul providers available. Backhaul connections are usually terminated at the
community-owned colocation facility. Any additional infrastructure constructed in Prior Lake
would need to meet service provider networks at a mutually convenient colocation facility or
meet point.
Page 6 of 17
TV and PhoneAnalysis
1he cost of purchasing a telephone (PSTN) switch ($491,217) and the accompanying cost and
risk of administering a referendum requiring 65% voter support to build the network of a can
be mirigated entirely by utilizing exisring wholesale VoIP services (in a retail, hybrid, or open
environment). If a ONT/CPE is chosen that has a SIP client and ATA built in, the VoIP traf�iic
can be priorirized on the local network and the SIP traffic can be sent directly to a phone
wholesaler. In this scenario, the City would still be the retail provider of services, but the VoIP
wholesaler would actually be handling the voice call switching and most of the technical
support needed to support dial tone.
On the TV side, channelized cable TV content is dying. The proposed TV head end
investment is old technology ($2,729,676 +$827,518 in Set Top Boxes). Not only is this
expensive, but City would have to negotiate content agreements; a costly and difi'icult process
that would require very specialized and expensive legal assistance.'Ihis cost could be avoided by
purchasing IP TV packages wholesale from an existing IP TV provider.
We would recommend a careful analysis of the take rate assumprions for TV prior to making a
decision. While a TV package will make an important revenue contriburion to the enterprise,
take rate is content oriented, and Dish Network and DirectTV will be hard to compete with
(and has very low content costs due to distribution size).
OTT (Over'Ihe Top) services (i.e. Internet based streaming video services like NetAix, Hulu,
NFL.com, HBO-Go, etc.) will provide a majority of content in the future, and the Parks survey
does not take into account that most of the OTT services are built into newer devices.
Additionally, the price of OTT boxes has come down significantly (Roku is now $49 and OTT
providers are offering them for free with a one year subscription).
Page 7 of I 7
Cost Analysis
The cost of purchasing a telephone switch and a TV head end could be avoided completely by a
doser evaluation of the open access model, in which private sector service providers would offer
those services. As noted elsewhere in this report, if the City adopted the open access model, it
would likely avoid the cost of a lawsuit that could result from the City selling retail services
directly to residents and businesses.
A more typical cost for OSSBSS (Operations Support SystemsBilling Support Systems)
software and hardware would be $100,000 to $200,000 as a one time license fee. Some OSS
software also includes small per subscriber fees (e.g. $1/month/subscriber) which covers all
software, upgrades, and ongoing technical support.
The City may wish to consider contracting out all outside plant (OSP) maintenance and
repairs, including emergency splicing. This approach eliminates the capital costs of bucket
trucks, tools, and equipment, and is generally less expensive for a network the size of Prior Lake
than the cost of dedicated full time City outside plant staff. As many as fottr full time staff
might be needed by the City to cover vacations and sick leave; a crew of two is generally
required to meet safety regulation when working on aerial or underground cable.
In the open access model, the City would be able to operate and maintain the network with as
few as four or five staff, assuming that outside plant maintenance is out-sourced to a qualified
private sector firm.
We would note that whether the City adopts the hybrid retail model or an open access model,
it is likely that the incumbent providers might engage in a price war. In other communities
where a new competitive network has been offered, the incumbents have made very sharp
reductions in pricing to retain customers. Price-cutting by the incumbents could depress the
take rate and revenue in the short term (i.e. the first two to three years of operation).
We would suggest a careful review of the build out schedule (what percentage of homes passed
over what period of time) and the expected take rate of services over time to ensure that these
projections are reasonable and that the projections do take into account aggressive marketing
from incumbents.
If the City adopts the hybrid retail model, we would recommend that a minimum of $500,000
be budgeted for legal and marketing expenses. If the incumbents sue the City, they will also
likely engage in a massive advertising campaign that would require a similar response from the
City to ensure that citizens and businesses had appropriate information on both sides of the
issue.
Page 8 of 17
Legal and Organizationa) Analysis
We would recommend that the City determine the ownership and governance of the proposed
enterprise prior to moving forwaxd with any implementation plans. We note that projects
which have not developed a solid business plan that answers these two important questions
prior detailed planning activities usually develop financial difficulties.
The City of Prior Lake should consider the following characteristics when making a decision
about the form of governance and ownership.
� Transparency - Does the governance structure provide adequate transparency about
decision making? Do stakeholders and interested parties have adequate ways to obtain
documents, financial reports, and related governance materials?
� Timeliness - How quickly can the governance enrity be legally formed? Time is of
the essence, as WiredWest must be able to leverage the funded MBI middle mile
backbone.
� Community Oversight - Does the entity have adequate community control and over-
sight? Do the communities have adequate representation in the governance structure
to ensure that assets are managed properly?
� Legislative Authority to Build/Operate - Does the governance entity have dear legis-
lative approval to build and operate a telecommunications network?
� Financing Options - Are there adequate financing options available to provide the
appropriate level of funding over time to meet the long term vision of WiredWest?
� Tax Liability - Does the governance entity incur tax obligations? >
OWNER O PTI O N S
Prior Lake has at least four oprions available to it.
MUNICIPAL OWNERSHIP
Many communities in the United States have municipal enriries that offer services to the
general public. The most common services are water and sewer, and are administered
operationally either as a department of the government or as an authority. Typical water and
sewer authorities are quasi-public entities that operate independently of direct local
government oversight but operate as a nonprofit. However, most towns in the WiredWest area
do not offer water or sewer services.
.Also common are municipal electric service operations. Several hundred communities in the
U.S. have municipal electric power, and some have moved into the telecommunications arena,
Page 9 of 17
largely because it is convenient to do so--the organization already has utility pole access,
eacperienced staff, and equipment like bucket trucks.
Government operated networks using the muni retail model attract legislation forbidding
localities from offering telecommunications services. Several states, including Pennsylvania,
Nebraska, South Carolina, and Virginia, have enacted legislation making municipal telecom
services illegal within the state shortly after a municipality or public service company started a
data service. 1he Virginia bill was overturned by the Federal Circuit court in a remarkably brief
decision that seems crystal clear:
I find that the broad and unambiguous language of � 253(aJ [the Federal Telecom
Deregulation Act� makes it clear that Congress did intend for cities to be "entities" within the
meaning ofthe TelecommunicationsAct. Therefore, � 15.2-1500(BJ [the Yirginia legislation
in guestion� is in direct conflict with federal law, and is void under the Supremacy Clause.
Section 253(a) is a concise mandate that no state may prohibit or hatie the e�'ect of
prohibiting the ability of any entity to provide any interstate or intrastate telecommunications
service."47 U.S. C.A. § 253(a) ......Simply put, it strains logic to interpi-et the term "any
entity' in � 253(a) to mean "any entity except for municipalities and other political
subdivisions ofstates." While it is true that such an interpretation is possible, the Supreme
Court bas cautioned that "[aJ statute can be unambiguous without addressing every
interliretive theory offered by tz party. �........ ?he federal statute, therefore, not only mandates
that no state statute "may prohibit" telecommunications competition, but also that no state
statute `inay have the e�ect ofprohibiting"telecommunications competition. 47 U.S.C.A. �
253(a).
While most communities that have been challehged by lawsuits have eventually won in court,
the legal battles usually add years and significant expense to such efforts. Lafayette, Louisiana,
as one example, spent substantial sums of money and nearly two years in court to defend the
right to build a community-owned network. The city eventually prevailed and now has an
outstanding network offering some of the lowest telecom service prices in the U.S., but the
effort was delayed for years by the lawsuit from the incumbents. Monticello, Minnesota also
eventually prevailed in its efforts, but only by enduring a lengthy court battle.
NON-PROFIT
'Ihere are various kinds of nonprofit businesses. The most common is the 501(c)(3), which is
limited to strictly charitable ef�orts. A 501(c)(3), according to IRS rules, must have a well-
defined charitable purpose targeted toward a specific need and/or a specific target population.
In other words, a 501(c)(3) cannot, according to IRS rules, operate as a nonprofit business that
provides services to the general public.
Many of the first community networking projects in the early and mid-nineties were formed as
501(c)(3) organizarions; it was common for these entities to offer dial-up Internet access to the
Page I 0 of 17
general public at a time when Internet service providers were still relatively uncommon. But by
2000, most of these organizations had closed their doors and/or disconrinued their Internet
access services because of IRS challenges.
Today (2012), we see new 501(c)(3) organizations repeating this approach by offering
broadband services either directly or indirectly (using an open access business model). It is our
view that eventually all these organizarions will receive letters from the IRS challenging their
status.
However, one or more nonprofit businesses may be useful as part of the overall effort. A 501c3
may be desirable as a mechanism to accept charitable donations, and more importantly, to apply
for certain kinds of grants. Once the funds have been received by the 501c3, and the donors
have received the tax credit, the nonprofit can, in turn, give or loan those funds to another
organization (e.g. an authority or coop chartered specifically to provide services).
AD HOCAND INFORMAL PARTNERSHIPS
Some local governments have deployed duct and/or dark fiber and have made ad hoc
arrangements to provide capacity to other institutions like K12 school systems or adjacent local
governments. In some cases, they have a policy for leasing duct or fiber to the private sectar
(the city of Sacramento, California has leased duct for over a decade) but have not developed a
comprehensive plan for management and expansion. Prior Lake is currently evaluating a
partnership with Scott County to extend the Scott County fiber to more areas of Prior Lake.
FOR PROFIT BUSINESSES
There are various types of for profit business organizations: individually owned businesses,
partnerships, general business corporations, and limited liability companies. The City could
issue an RFI asking for private sector proposals to extend fiber services to part or all of the City
over a reasonable period of time. It may be useful to do this as a first next step--that is, to
ensure that the private sector has had a chance to improve broadband offerings in the City
before the City makes a decision to move forward. If the City does issue an RFI and the
responses axe inadequate (or none are received), the City then has demonstrated there is
jusrificarion for City involvement.
Page I I of 17
Business Model Analysis
Other communiries across the United States are already actively pursuing new and innovative
public/private partnerships to improve the access and affordability of telecom services delivered
via broadband. Communities that have affordable broadband are enjoying a faster rate of
economic growth than communities that lack broadband, based on a CMLJ/MIT study
(Measuring the Economic Impact of Broadband Deployment, Sirbu and Gillett, 2006).
A Brookings Institution study (Crandall, Lehr, and Litan) in 2009 found that for every 1°�6
increase in the availability of broadband in a community, the level of employment increases
correspondingly by .39�0 annually. 'Ihe study also found that as the level of Internet users
increased in a community, there was a corresponding increase in economic growth, with a 109�6
increase in Internet use yielding a 1.3% increase in the economy.
A new digital divide is emerging, with fiber as a differentiator. Communities with affordable
broadband infrastructure and the ability (i.e. fiber) to expand capacity as demand grows over
the next seven to ten years should enjoy a measurable economic development advantage over
communities that lack such infrastructure.
A variety of business models are being explored, but most community-owned networks can be
classified as either Municipal Retail (similar to the Hybrid Retail model proposed for Prior
Lake) or Open Access.
Open access networks have several advantages that the City may wish to consider, including:
� Lower cost of build out because no TV head end or telephone switch is required.
� Lower cost of operation because all retail services are sold by private sector companies.
� Less risk of expensive lawsuits because the City would NOT be providing any retail
services.
Some additional informarion on some of these projects is provided below.
PALM COAST, FLORIDA
In 2008, the City of Palm Coast began exploring the potential of making existing City-owned
fiber assets available for business and commercial use. Exisring Palm Coast businesses were
expressing concern to City leaders about the high cost of Internet access and the limited
bandwidth available in the City. After a six month study of various business and financial
options, the City decided to focus on developing the network as a"carrier class" commercial
network capable of supporting virtually any level of business service that might be needed. �
As of early 2012, all four redundant fiber loops had been completed. The City invested in a
dedicated colocation facility with both shared rack space and private cages for service providers,
Page I 2 of I 7
and purchased "carrier class" network switches and routers to light up the fiber. Palm Coast
FiberNET was made available for service in May, 2010 (http://www ci.palm-coast:fl.us/
PalmCoastF'iberNET�, and had three service providers committed on day one.
Palm Coast F`iberNET provides service to City buildings and locarions, and successfiilly won a
bid to provide services to Flagler County Public Schools. The local hospital also uses the
network to connect hospital medical records and data services with several local health clinics
and medical offices. FiberNET was operating in the black operationally in year one, and
continues to do so as it enters its fourth year of operarion.
Palm Coast FiberNET is owned by the City of Palm Coast.
City enterprise funds were used to pay for the initial $2.5 million in fiber
construction, equipment, and the colocation facility.
FiberNET is operated as an open access network. Providers pay a
monthly fee per customer, based on connection size.
The City IT Department manages network operations, and private sector
contractors are used for outside plant maintenance and construction work.
FiberNet is an active Ethernet network that provides symmetric 100
megabit, Gigabit, and lOGigabit connections as standard. DWDM cir-
cuits can be provided upon request.
Page I 3 of 17
DANVILLE, VIRGINIA
The City of Danville, Virginia is operating an open access, open services network
(www.ndanville.net) focused on creating the right kind of economic development incentives
and accompanying infrastructure that will help retain existing businesses and help attract new
ones. Danville has a City-owned electric utility, and the growing fiber network is being
managed as part of the electric urility operations.
Using a multi-phase approach, the City first hooked up government offices and local schools in
2004, and in 2006 began planning for extending the high performance all fiber network to local
businesses and residents throughout the electric service area, which includes a large part of very
rural Pittsylvania county. The first businesses began to get hooked up in late 2007, and Danville
had fiber passing parcel in its business parks before the end of 2008. 1he City-County business
incubator was one of the first locations to receive the fiber services.
The network has been operating in the black and has generated enough revenue to make an
annual contriburion to the City's General Fund, and a portion of the nDanville revenue is being
used to expand the network. 1he City has completed the construction of Fiber To The Home
(FTTH) to some of its residential neighborhoods (a total of about 1600 premises), and has
been able to sign up an IP TV provider as well as Internet and telephone providers for the
residential customers.
The City is not selling any services to businesses or residents; all services are offered by private
sector service providers that use the network and pay the City for the use of the network via a
revenue sharing agreement. The availability of business class fiber has been a significant boost to
the downtown area of Danville, and the City has counted at least 150 new jobs within walking
distance of the Main Street commercial area of town.
nDanville is part of the City of Danville Utiliries Department.
The City of Danville Utilities Department has used a combination of loans and
revenue to fund the construcrion of the network. Revenue from key institutions
like the City and County schools have been a significant factor in the development
of the network
nDanville is an open access, open services network. All services provided to resi-
dents and businesses aze offered by private sector providers.
Network operarions are managed by the City. Some outside plant maintenance is '
performed by City utility crews, and some work is outsourced to qualified private
sector firms (e.g. splicing, some construction work).
nDanville is an acrive Ethernet fiber network, providing a 100 megabit symmetric
connection as the standard service. Gigabit and lOGigabit point to point connec-
tions are also available. nDanville has two colocation facilities available to busi-
nesses and providers, and the nDanville MSAP (Multimedia Services Access Point)
provides access to more than twenty-five local, regional, and national service provid-
ers.
Page I 4 of 17
FARMERS TELECOM COOP
Farmers Telecom Coop serves 17,000 subscribers in rural Georgia (www farmerstel.com), and
the customer-owned enterprise has begun executing on its plan to take fiber to every premise.
11ie telephone company chose an active Ethernet network design because an active network can
scale up more easily to meet future demand than an equivalent PON system, and because
network troubleshooring and diagnostics was deemed easier to perform, Customers receive a
100 megabit fiber connection capable of delivering a wide variety of services, including the
traditional triple play of voice, video, and Internet, but also video and movies on demand, HD
business videoconferencing, telemedicine and telehealth services, and a wide variety of other
business and residential services and applications.
The coop has an eight member board of trustees, and coop members elect
board members.
The coop is financing the network upgrade with internal funds and loans.
The coop is offering traditional triple play service bundles. 'Ihere is no
competirive service offerings.
'lhe coop handles both network operations and outside plant maintenance
with existing staff.
The network offers a standard 100 megabit symmetric fiber connecrion
that supports voice,TV, and Internet access, and Gigabit connections are
also available.
Page I 5 of 17
r
LAFAYETTE, LOUISIANA
Lafayette, Louisiana is perhaps one of the best known community broadband projects in the
United States. 1he City announced its intentions to go into the broadband business in 2004,
and was promptly sued by the incumbent cable provider. 'llze court case ground on slowly, and
it was not until the City had spent nearly $4 million on legal fees that the Louisiana Supreme
Court decided that the City had the right to compete directly with private sector telecom
companies.
Since then, thousands of customers have been connected and Lafayette is now famous for
having some of the lowest rates for Internet access in the United States, with a 50 megabit
symmetric package of Internet access for only $58/month. 1he network has now been
operational since early 2009.
Cox Communications, famous in Louisiana for regular rate increases, froze its rates in
Lafayette for several years following the city's initial announcement that it would offer
telecommunications services. Meanwhile Cox continued to raise its rates in other parts of the
state.'llie result was that even before Lafayette's system began operating it had saved its
residents and businesses nearly $4 million.
The network is owned and operated by the City of Lafayette and is part of
the Lafayette Utilities Department.
The City raised $110 million in funding to build the network.'Ihe long
term plan is to pass all 57,000 homes in the city.
Services are sold direcdy by the City in a traditional triple play retail
model.
The City Urilities Department operates the network and handles outside
plant maintenance.
LUSF'iber is an acrive Ethernet system with a standard 100 megabit
symmetric fiber connection. Gigabit connections are also available.
Page I 6 of 17
s
Next Steps
As the City moves forward, several options could be pursued.
� 11ie City could issue an RFI immediately that asks for private sector proposals for a
public/private partnership. It would be very useful to ask the private sector, including
the existing incumbent providers in Prior Lake, to step up and make an offer for a
publidprivate partnership prior to committing the City to a major expenditure. If no
practical proposals are received, then the City has demonstrated that there really is a
need for government to become involved. This could also provide some protecrion
against lawsuits later. '
� Evaluate the benefits of an active Ethernet network or a hybrid PON/Active network
to take advantage of the lower equipment costs for residential customers, but include
active Ethernet equipment so that any site on the network could subscribe to a dedi-
cated 1Gbps up to lOGbps services.
� The City could take an incremental approach to improving access to affordable, high
performance fiber in the community by building basic infrastructure only on an in-
cremental basis. By building just duct and dark fiber and leasing fiber pairs, the City
reserves the oprion to offer a lit open access or hybrid retail network later, but in the
short term could lower the cost to private sector providers of adding new services in
the City. Duct and dark fiber leasing is simple and easy to manage, and helps lower
provider costs.
�'Ihe City should consider "dig once" and joint trenching ordinances to leverage road,
sidewalk, and water/sewer improvements. With streets and/or ditches already open,
every project should be evaluated to see if telecom duct should be added at a small `
incremental cost. The City could also require all new development to have telecom
infrastructure (i.e. duct from the building to City right of way) just like water, sewer,
sidewalks, and other improvements are built by the developer but are handed over to
the City.
P�ge 17 of 17
Email comments from Eric Lampland, Lookout Point Communications,l/10/13
(comments on the Design Nine report for Prior Lake, Dec. 2012)
Andrew recommends Active Ethernet (AE) architecture -- we used a hybrid architecture of AE
and GPON for a number of reasons, but all of Andrew's concerns were covered. We focused AE
architecture on the business community and GPON on the residential community and we
added a wireless overlay. Most importantly, the hybrid structure we recommended was driven
by two key factors: cost/performance & manageability is better, and alignment with future
architectures is more secure (higher PON/AE speeds; Wave Division PON; alignment with the
standards bodies; things yet to come, etc.). Less informed arguments often get into "my
architecture is better than yours." It is not an either-or argument, it's both or all. Its use is best
suited in all cases. That is what is in your design.
Andrew recommends against an RF head-end -- we're not using one, not recommending
one. The recommended video architecture anticipates everything that Andrew points to and
implements an actual path to getting there. Video is a difficult market and I would be careful of
generalizations without a deeper understanding. On the BAC, Wayne loved our approach and
that is his business.
Andrew promotes Open Access and even provides ideas about private VoIP providers versus
Prior Lake implementing a soft-switch. We modeled several open access approaches and were
not able to find a sustainable model. The issue here is monetization. Who gets the money and
how do you pay off the bonds. A current client of mine uses a very good private VoIP
provider. The provider gets $16.95 per month for each customer my client charges
$29.95. Prior Lake's FTTH is likely to have approximately 4,000 phone customers. That is
$813,600 ep r year (4,OOOx16.95x12) that you would give to someone else versus a soft-switch
(roughly $400,000; a one-time cost) and for the other $413,660 I am sure you can find an
adequate staffer, even two! There is a whole complex of issues in Open Access. I really do
understand this as I have wanted to see this model work for years. We're not there yet, but as
you know, we looked long and hard at that approach during the BAC study.