BIG BROADBAND
for Economic Development
(or... Fiber, Wireless and Bandwidth for TeleWork)
By Wayne Caswell, CAZITech Consulting
(SPEAKER NOTES from this 11/3/2004 presentation at Austin InnoTech Conference & Expo, in conjunction with the CIO Summit. The abstract, 1/4-size slides, and a list of online resources is also available.)
Slide 1: Fiber, Wireless and Bandwidth … for TeleWork
Purpose:
Be Provocative and stimulate Questions, Debate & Innovative Thinking
Intro:
While Austin fights over toll roads and light rail, other high-growth regions are driving economic development with TeleWork programs and high-speed fiber optic and wireless networks. They know that technology innovation depends on sharing knowledge, and that telecommunications infrastructure is just as vital to future economic growth as transportation systems were in the past, where industries benefited by being near major highways, waterways, railroads, and airports.
Slide 2: Big Broadband … for Economic Development
Intro, CONT.
This session shows how telecommunications infrastructure supporting BIG BROADBAND enables TeleWork and encourages innovation.
Austin has already become a world center for wireless development – with UT’s Wireless Network & Computing Group, the Austin Wireless Alliance, and similar initiatives – but I will argue that the city should also invest in fiber optics as a public utility that complements wireless.
This is important since broadband services using DSL & Cable lack the capacity for important applications like video conferencing, which supports telepresence and the ability to hire the best talent wherever it is.
Presentation Origin
Much of the material from this presentation is pulled from a white paper I wrote earlier this year, “Reviving the FORGOTTEN Information Superhighway,” which is on my Web site and describes the need for a National Broadband Policy. The U.S. is the only Group of Seven country without a policy that encourages broadband deployment as a way to ensure competitiveness in the global information economy. My paper helps policy makers understand the issues and what’s at stake so they can balance the public good against powerful lobbyists and their special interests.
About CAZITech
Independent consultant firm serving Broadband, Wireless, and Home Networking markets with Strategic Planning, Market & Competitive Analysis, Market Development, and related services.
Slide 3: Topics
Agenda – These three overlapping circles highlight the four main points in my presentation – Access, Affordability, Speed, and the APPLICATIONS they enable.
Broadband Speed - measured in Megabits (or Gigabits) per second, is what enables many new applications.
Access – is measured as the percent of the population that subscribes to broadband services. The higher the penetration rate, the more incentive businesses have of offering new applications and services.
Affordability – the cost per month directly effects how many people subscribe.
Applications – Over dialup connections, they are limited to email, messaging, and browsing with simple graphics. Faster DSL & cable service adds digital music and support for more media-rich Web pages, but is that enough for video? We’ll explore that.
Logistics & Politics – I’ll end with a look at the logistical challenges of extending fiber deep into the network, ask who should own and manage that infrastructure, and offer some opinions about the politics and business models critical to market success.
Slide 4: Speed: and the Applications
Applications – We started with the three overlapping circles (Access, Affordability, and Speed) and how they play off of each other and discussed how lowering Cost and increasing Speed helps increase Access.
Larger – I showed how Broadband supports New App, including high-def video, so this chart makes each of the circles as larger – more Affordability, more Speed, and more Subscribers. With many more consumers able to access faster broadband, application developers have more reason to invest in:
Digital Entertainment,
e-Commerce,
TeleWork,
TeleMedicine,
Distance Learning, and
e-Government.
Trillions – The net benefit is worth TRILLIONS !!!
Slide 5: e-Commerce
e-Commerce – Let’s look at just one of those application sectors: e-Commerce.
B-to-C – The charts here, with 2002 data from Forrester Research, includes both B-to-B and B-to-C e-Commerce, shown by region. What is not shown is that B-to-B makes up most of the market today but B-to-C is arguably where the biggest opportunity lies.
Leadership – The little pie in 2002 was worth over $2 Trillion, and that’s projected to grow to nearly $13 Trillion by year-end 2006.
Notice that North America has the dominant share but that our growth rate is much lower than in Asia and Western Europe.
We are now the only G7 country without a national BB policy, we’re losing ground and our technology leadership, and our share of the global economy is shrinking. This is cause for grave concern and should be a wake-up call to policy makers and voters alike.
Slide 6: TeleWork
Traffic – Time ($80M/yr: ~$4,000 per telecommuter, $126M: ~$700/yr per commuter), Fuel ($20M/yr: $500/yr + $55/yr), Accidents (lives), Wear+Tear
Assumptions: 700K population, 200K workers, 10% telework, $60K/yr salary, $120K burdened, 50hr/wk, 45min each way, 5 min extra congestion
Environment – Ozone Action Days affect federal highway funding
Employer – Avg. 10% Productivity,
Recruit/Retain/Relocation, Office Space = $10,000, Fiber over-build = $800
(costs measured in $100's vs. benefits measured in $10,000's)
Employee – Family time,
Continuity of Operations – Distributed workforce is less likely to be crippled by Bad Weather or wiped out by Terror Attack. We must protect the physical security of our workers and continuity of key business and government operations.
The Interstate Highway System - I-35 was part of an interstate highway system introduced after WW-II (in 1955) by President Eisenhower as an investment in national security to facilitate quick shipment of goods and heavy equipment, and it has had untold economic impact. For today’s digital society, we need an Information Superhighway.
Landscape of Cities – Location/Location/Location, Electricity/Cars = Elevators/High-Rise so workers could interact. Now Cities/Countries risk losing skilled labor when engineering graduates move away to follow jobs.
Slide 7: Speed: for TelePresence (1)
Loudoun, VA Economic Summit – Congressman Frank Wolf (R-VA) funded the TeleWork Consortium and several Pilot programs that stressed TelePresence.
Without telepresence, telework is limited to specific job types.
Need BIG BROADBAND (up & down)
Loudoun Magazine (up-scale quarterly) – Lease was up
Group videoconference vs. Peer-to-Peer (CuSeeMe, NetMeeting, etc.) combining conferencing (non-verbal communications), instant messaging and white board.
TelePresence is an example of real-time collaboration. Video streaming is also real-time but one-way. The up-link speeds of cable & DSL are too slow for video conferencing (need 512 Kbps)
Rural Opportunities too
Slide 8: Speed: for TelePresence (2)
TelePresence also enables the apps shown here.
TeleMedicine – Rural, Disabled, Aging Baby Boomers (impact on Healthcare System). A Florida project serving kids with diabetes found that telemedicine eliminated 80% of their hospital visits and saved thousands of dollars per patient.
Signing & Surveillance
Lifelong Learning – 30 yrs at IBM, but no longer provided by employer, so how?
Collaborative Performance – U.Texas hosted Internet2 conference with demos of Miro String Quartet
Slide 9: Speed: and Application Requirements
Broadband is often referred to as a Big Pipe, so we'll use the pipe analogy to explain what apps need what speed.
Dial-up – 56 Kbps (if that) easily supports text applications (original PC = 300bps) and Web browsing with low-res images. Those simple apps are represented in my drawing as be-bees & marbles. Sending & receiving photos & music is painful, like sending many marbles.
1 Mbps – DSL and Cable modems support speeds up to 1 Mbps for faster graphics & music – essentially more of the same. In addition to carrying more be-bees & marbles, the bigger pipe can also carry a few new apps such as compressed video, shown here as golf balls. (For some, "always-on" is the biggest benefit of DSL or cable.)
100 Mbps – Some experts say we need at least 100 Mbps, while others say we need 1 Gbps, so we can get multiple streams of high-definition video – like several basket balls, a beach ball, bowling ball, etc.
Compression – With MPEG4 compression, HDTV consumes only about 4-5 Mbps instead of 20. I mention compression since basket balls and beach balls can be compressed by sucking out the air and then inflated again at the other end. But bowling balls don’t compress much, and since it takes time to compress and decompress (latency), that won’t help telepresence apps.
So how much speed is “really” needed? If we want to receive 2-3 simultaneous video streams at HD resolution with MPEG2 compression, then we’d need about 100 Mbps. With MPEG4, we might get that down to 30 Mbps.
UHDTV – If our infrastructure must last well into the future, then we should consider the needs of UHDTV (ultra-high-definition TV), which could require multiple Gbps.
Slide 10: Speed: Download The Matrix (7.6 GB)
Real-time – Streaming music or video requires QoS (buffer helps accommodate short N/W outages).
Mobility – Download onto mobile device requires more bandwidth due to impatience. (Faster than Real-time)
Dial-up? – Out of the question. Pony Express is faster.
DSL or Cable? – $40/mo for 1 Mbps, but FedEx is faster. (Business Class service is a bit faster but much more expensive: 5-10Mbps for $100-200/mo).
Fiber-to-the-Premises – FTTP provides Fast Ethernet and Gigabit Ethernet speeds, and often at the same cost as DSL (about $60/mo).
Pricing Dilemma – Value of Digital Movie (7 GB total) vs. value of Sensor Alert (100 bytes total) vs. voice (64 Kbps)
Leading to a comparison of the Technology Options...
Slide 11: Speed: and Deep Fiber
Solution = Deep Fiber – As we extend fiber closer to the user’s device, the legacy wiring gets Shorter and therefore Faster. But as we extend closer, we run into more worries over Rights-of-Way.
DSL – We can use existing telephone twisted pair wiring for the last mile or so. As we move fiber closer, we move from ADSL speeds to VDSL, with data rates approaching 24-52 Mbps in five years or so. That’s fast enough for one uncompressed HDTV stream or many compressed streams, but there is NO two-way 100 Mbps roadmap.
Cable – Advanced Hybrid-Fiber Coax systems use coax cabling for the last mile. When fiber extends to the curb instead of just to the neighborhood, then the speed of HFC networks increases, approaching the 100 Mbps target, so it seems that cable is an alternative to taking fiber all the way to the premises, and in fact that is what Time Warner plans to do. DOCSIS 2.0 = 800 Mbps (30Mbps * 53 channels).
Wireless – There’s a lot happening with Wireless Municipal Area Networks with technologies such as WiMAX, which reaches up to 30 miles and can support data rates of up to 280 Mbps per base station (but not both at the same time). Even newer point-to-point millimeter wave technologies can exceed 1 Gbps.
Austin has arguably the largest number of FREE wireless hotspots, but Philadelphia and San Francisco are example of cities with plans for wireless municipal networks covering the entire city.
Powerline – (not shown) The FCC has recently approved Broadband over Powerline
FTTP – Taking fiber to the premises is easiest in new neighborhoods with no existing infrastructure. What I show here is an approach that uses passive optical couplers, and you can see the data rates supported. Landowne is a development in Loudoun, VA that offers 100 Mbps performance at $60/mo (or $140/mo for voice, data and television).
Slide 12: Speed: and Home Networking
Vision – My specialty is Home Networking, and I have a vision that includes "Consumers with EASY access to services and Service Providers with EQUAL access to consumers, all without worry about incumbents that control the infrastructure," but I don’t see that coming from cable or phone companies and am excited by recent Public or Public/Private networks, as I’ll describe in the following charts.
Last Mile – Here’s the problem: The fiber loop and fiber backhaul network are very fast, and home networks are relatively fast, but the last mile connecting them is often way too slow.
FCC – Cable companies were first to upgrade their networks, spending billions in the process and now adding VoIP services to compete with phone companies. This was possible because they were never required to share their networks with competitors. Phone companies, on the other hand, did face such rules, so they were less likely to upgrade. Recently the FCC removed that requirement for new fiber networks, and as a result SBC, Bell South, and Verizon have accelerated plans to take fiber all the way to homes and to offer TV service.
Home Network – Once the digital broadband network reaches the home, it will need to connect the phones, stereos, TVs, PCs, and other devices inside. That’s the realm of the Home Network, and even though fiber is sometimes found in high-end new home construction, it’s not practical to retrofit in existing homes. The same goes Ethernet cabling, so the most common home networks are wireless or based on existing power lines, phone lines, or coax cabling.
Standards – All of the home networking standards shown here are headed toward speeds of over 100 Mbps.
RG – The Residential Gateway is a bridge between the different networks, often adding security and remotely provisioning various services.
Affordability – Next I'll discuss Public broadband vs. Private broadband and the implications on affordability and number of subscribers in the U.S. vs. abroad. In my vision, Competition drives low prices and rich features, making broadband attractive to more people.
Slide 13: Affordability: example
Affordability – Some boast that over 80% of the U.S. has access to broadband, but only about 20% actually subscribe since most Iconsumers think it’s too expensive and not fast enough to add compelling benefit.
S.Korea – Leads the world with over 80% of households subscribing to broadband and with speeds that are 10-times faster than here for just $25/mo, which includes voice, data, and television. Another $8/mo adds Wi-Fi hotspot access. This rapid growth and competitive pricing is largely because of aggressive government policies and incentives that have encouraged competition with a national objective of ubiquitous access to 155 Mbps to 5 Gbps by YE 2005. But note that it’s easier to run fiber in Japan and S.Korea since most of the population lives in high-rise apartments, but know that the U.S. is the only G7 country without a national BB policy.
Japan – 1 Gbps for $38/mo (voice, data, video)
USA – In comparison, I’ll use my own expenses to show how poorly the U.S. is able to compete on a global scale. I pay:
$100/mo to SBC for local phone service.
$45/mo to Time Warner for RoadRunner cable modem service.
(Yankee Gpe – 17% at $45 but 71% at $19.95)
$59/mo for digital cable TV (required for RoadRunner), but this also includes $10/mo to rent my DVR.
So, for inferior voice, data & video service, I pay over $200/mo.
Now add Mobile phone service (with long distance calls included) with three phone numbers in a family plan with lots of minutes, and I pay over $380/mo. That’s ridiculous!
Bundles – I’m an ideal candidate for bundled services if that can give me a single bill and support number while lowering my costs. Now I could move to another neighborhood, like Avery Ranch, where they already have FTTP, or I could move out of Austin altogether.
Business Models – Why don't we have more competition, better prices and faster networks? It's largely because our business models don't justify it, as we'll see in the next few charts.
Slide 14: Access: Fiber Cost vs. Take Rate (1)
Neither cable nor FTTP will reach my vision if it’s owned (and controlled) by a service provider, because installation is expensive and the payback period is long (much longer than private companies allow), and because it Limits Competition.
Intro – This chart shows the fixed cost of new fiber networks and the variable cost per subscriber. These costs are easier to justify in new (green field) installations, or where the carrier can get monopoly power, or (as I’ll discuss later) when there’s a New Business Model.
Take Rate – The average cost is lower with more subscribers, so justification needs as many customers as possible, essentially moving as far to the right as possible along the green curve. It can take several years to reach 50% penetration, and stockholders aren’t that patient. Reaching a high Take Rate is even more difficult when competition is present, so it’s often not justified to run fiber (except in new neighborhoods where the utilities have yet to be installed).
Slide 15: Access: Fiber Cost vs. Take Rate (2)
Triple Play – FTTP needs as many services as possible, and bandwidth alone won’t do it. Neither will HDTV. Network operators talk about the Triple Play of services since that moves monthly revenues up and into the “profitable” range.
Killer App – To move even higher requires a killer app, like telepresence, but this can be a Chicken & Egg dilemma, since the apps aren’t developed until bandwidth is available, and bandwidth isn’t installed until apps are available.
Slide 16: Access: Fiber Cost vs. Take Rate (3)
Monopoly – So BIG BROADBAND needs a high take rate of 35% or more, as well as a enough services to justify a higher monthly fee (RPU, revenue per user). BUT, without TelePresence (or some other killer app), it’s hard to get that high fee. And with competitors vying for the same customers, it’s hard to get the high take rate.
FCC – The FCC’s new rules giving telcos monopoly control of their fiber networks has the potential of keeping out other potential competitors by making it nearly impossible to capture a take rate high enough to justify their own infrastructure.
Public – Which brings me back to the concept of Public Broadband Networks.
Gartner Consulting – predicted that faster broadband deployment in Michigan would help create nearly 500,000 new jobs and $440 billion of additional economic output over the next 20 years.
Public Airport Model – Airlines don't build their own airports; cities do. Governments can absorb longer payback periods and are good at building roads, bridges, seaports, and airports. Like airports, public networks are funded with long term municipal bonds that are paid off by revenues from ISP access fees that also fund network operation. The result is more competition and more choice.
Monopoly Model – Imaging UPS building its own roads for its delivery service and charging a fee to competing truckers or to the public. Now imagine every trucking company being forced to build their own roads. That would be a terrible waste of duplicate effort and would not serve the public good. Incumbent monopolists often wage political battles, with their deep pockets and powerful lobbyists.
Lobbyists – Big Broadband is disruptive to many industries, so the list of opponents includes not just cable and phone companies, but also car manufacturers and oil companies (telework can take cars off of roads), brick & mortar universities (new competition from University of Phoenix Online), professors (often view the course materials as intellectual property and don't want it online), landlords (of older buildings downtown where it's hard to run fiber), etc.
Successes – But there several are examples of communities that have won political battles against the powerful lobbyists, with the result being More Competition, More Choice, Lower Prices, Better Features, and More Subscribers, leading to Accelerated Innovation AND Economic Development.
Voice – service can come from incumbents (Bell South, Quest, SBC), mobile phone companies (Cingular, Sprint), cable companies (COX, Time Warner), or new VoIP companies (AT&T CallVantage, Lingo, Net2Phone, Skipe, Vonage).
Television – can come from Time Warner or anyone else that bundles programming, or it can come over the air from satellite and terrestrial broadcast or across the fiber from on-demand services (CinemaNow, Intertainer, Movies.com, MovieLink, Netflix, Starz).
Music – Even though Road Runner bundles in the Rhapsody music service for free, consumers with Apple iPods use iTunes instead, and a public network opens music competition to BuyMusic.com, MSN Music, MusicNow, Napster 2.0, Virgin Digital, Walmart and many others.
Slide 17: Access: Individual Communities and their Adoption of New Technologies
This chart shows the economic impact of a public infrastructure decision affecting railroads.
Pre-Railroads – Let’s contrast Chicago and St.Louis during the emergence of the Railroads. Up until about 1870, the main means of transportation was through Rivers and Canals, and that drove the economies of both cities – Chicago along Lake Michigan and the Erie canal, and St.Louis on the Mississippi River.
Railroads – While St.Louis passed laws prohibiting railroad bridges over the river to protect precious water traffic, Chicago embraced the railroads, and that made all the difference.
Fiber – A similar comparison can be made between sister-cities in Iowa, one who installed fiber optics, and one who upgrade their cable system instead.
Slide 18: Access: Fiber Cost & Benefit (tale of two cities)
Cedar Falls – With a population of 36,000, Cedar Falls was the smaller of the two, but it’s the one who chose fiber optics.
Waterloo – Waterloo was nearly twice as large, with population of over 68,000.
Fiber – Here’s the comparison after Cedar Fall installed fiber and Waterloo went with coax. New construction in the smaller Cedar Falls nearly doubled that of Waterloo, and less taxes were needed to fund that growth.
Mayor – As you can see from the quote, Waterloo Mayor John Roof is now sold on fiber optics.
Slide 19: Politics: Who else is doing Fiber?
Avery Ranch – One of the most attractive amenities of this Austin community is its fiber infrastructure, which gives it a leg up over others around Austin.
Loudoun County – Last month I attended an Economic Summit in Loudoun County, VA, a suburb of Washington, D.C. located in the beautiful rolling hills of VA along the Potomac River. Loudoun is the fastest growing community in the country, and they just can’t build roads fast enough, so the daily commute is more than a pain. It’s agony. (‘Sound like Austin?)
UTOPIA – One of the presentations was about UTOPIA, a Public “slash” Private agency of 18 cities in Utah that is building fiber networks as public infrastructure and managing them like airports.
Texas Legislature – Several states have laws that prohibit municipalities from offering telecommunication services that compete with incumbents, and the US Supreme Court confirmed that states had the right to pass such laws. Texas is one such state, so I had the TX Assistant Attorney General send me a copy of the specific legislation. It was written 8 years ago, prior to the Internet and the Telecommunications Act of 1996, and was designed to protect telephone service. The Asst.Att.Genl. agrees that it does not seem to apply to digital broadband services, especially where the city sells no direct service to the public but instead provides right-of-way access to the information infrastructure for a fee in an open and nondiscriminatory manner.
Slide 20: Politics: UTOPIA (1)
Quest – The incumbent phone company, Quest, fought the project, hoping to preserve its monopoly, but Quest is now a partner in UTOPIA.
Comparison – The Blue line is UTOPIA’s base case, and the Black line is its projected Break Even, compared with the Green line, an average of Take Rate curves for similar communities across the country.
Break even is 1 in 3 homes choosing 1 of 3 basic services from dozens of providers by year 7, and that’s where the Real Benefit is.
Slide 21: Politics: UTOPIA (2)
The UTOPIA Take Rates are actually faster than in early fiber communities, largely because they enhanced their plans based on what was learned.
Slide 22: Politics: in the Market
POLITICS – A colleague of mine at IBM several years ago used an interesting analogy that I’d like to share with you since it describes the factors that determine market success. He said that only 20% of the “problem” was Hardware. 70% was Software. And…
90% – was Politics & Business Models. Companies must be able to make money with Big Broadband and the and enabled applications. Government should not determine winners and losers but can still offer vision and investment incentives as was done in South Korea.
Encouraging – I find several recent events encouraging.
Reform: The FCC has been a champion of regulatory reform and of removing laws that prevent competition.
Phone: New FCC rules now make it easier for mobile phone users to switch between service providers while keeping their phone number, and this concept is being extended to land-line and voice-over-IP service.
Cable: The cable companies have been exempted from old telco rules that would have required them to give competitors access to their network at competitive rates.
Telcos: And to encourage phone companies to install fiber, they no longer have to share access to these new networks, meaning they have new reason to invest, but also raising some concerns about monopoly power.
Standards: Bell South, SBC and Verizon recently announced plans to invest in FTTP and define industry standards that will help drive down deployment costs.
Capacity Glut: in PC & in long haul NW pressure last mile from both ends.
BUT – But who pays to install the networks, and who ultimately owns and controls them? Who protects them? Should it be service providers, or should this infrastructure be viewed more like the public highway system since it benefits all of society? Schools. Businesses. Health Care. The Disabled. The Environment. And our Future in the global information economy.
SUMMARY – Now that I’ve raised some provocative issues, including whether we even need fiber to homes, this is a good time to stop and let the debates begin. POWERFUL LOBBYISTS
Slide 23: Resources
See the links to online resources that I’ve placed on www.CAZITech.com.