Submitted by acohill on Wed, 10/27/2010 - 09:18
Netflix had an outage of several hours that prevented their customers from accessing any streaming content. This article discusses whether Netflix is spending enough on infrastructure, but what has also emerged is that Netflix customers using the company's streaming services are now consuming 20% of all the bandwidth in the U.S. during peak evening hours. As I and many others have been predicting for years, video in all its forms is now driving use of the Internet.
Economic developers who dismiss the Netflix bandwidth trend as "not our problem" do so at their peril (and at their community's peril), as business use of videoconferencing is already routine. A year ago, Design Nine staff were rarely asked to make use of videoconferencing by our clients, but over the past several months, it has become commonplace for us to augment on-site meetings with video-enabled distance meetings. Internally, we coordinate the activities of staff in remote locations via video instead of the phone, and some days my videoconference use is probably higher than my use of the phone.
Meanwhile, communities that ignore the bandwidth tsunami and let their local businesses try to buy overpriced bandwidth via fifty year old copper infrastructure are putting their existing businesses at a competitive disadvantage (essentially inviting them to leave for a community with better infrastructure) and driving away new businesses. Design Nine's bandwidth costs have doubled in the past year, largely because of lack of competition. We have good service, but costs are increasing because in our business park, there is essentially a monopoly. I've spoken to one business owner down the street who has looked seriously at moving his entire business to Northern Virginia because of the lack of competitive fiber services here.
Submitted by acohill on Wed, 10/27/2010 - 07:55
A few months ago, a competitive telecom provider ran fiber down the main road near my home. Yesterday I figured out why; a crew was running a fiber drop to the bank branch on the corner. All over America, it is the dawn of Fiber 2.0. Fiber 1.0 took place in the late nineties, when an enormous amount of capital was spent on fiber too far in advance of the marketplace for demand. Along with the rest of the dot-com ventures, Fiber 1.0 was a bust. But today, the market for bandwidth continues to grow along a nice smooth curve, with the demand doubling every two years, and we have fifteen years of data to back this up. While the incumbents are busy trying to convince us they can meet this demand with 1950s copper cable plant, smaller telecom firms are busy spreading bits of fiber through communities to cherry pick the more profitable business customers. These companies tend to have no interest in full fiber build outs, and instead just want to lock up a portion of the local business market. Fiber is still costly enough (mostly for the labor to put it in) that once a customer like a local bank is captured by one of these smaller fiber firms, no other provider will gamble on the expense of building a second fiber cable to the same location: the first fiber provider in essentially creates a small monopoly. Nice work if you can finance it.
The tragedy is that as communities are chopped up among two or three small telecom firms, a Balkanization occurs. In effect, the incumbent duopoly (the phone company and the cable company) is replaced with a cartel--a handful of providers who have a vested interest in limited competition. Prices come down a little, but then freeze at the new cartel price point. The end result is that it becomes more challenging financially to build a single community-owned shared, high performance network; not impossible, but more difficult because key anchor tenants like schools and large businesses have already been cherry-picked with long term contracts.
Submitted by acohill on Wed, 10/27/2010 - 07:44
Fred Pilot makes an excellent observation in his excellent blog: he says that getting fiber to homes and businesses requires a change in attitude on the part of those homeowners and businesses--a shift away from passively accepting whatever an incumbent monopoly provider offers and moving to an ownership attitude. Fred talks specifically about using a coop structure, but the point is appropriate for any community-owned network governance structure, as ultimately, the residents and businesses of the community end up taking control of their own economic destiny by building a community-owned network business.
Taking control of a community's economic future requires work--more work than just quietly paying the cable or DSL bill every month and fuming over the high cost and the poor service. It means some community leaders have to learn new skills and take on the work of managing a new kind of community infrastructure. But most communities can do this; the world's best fiber network can be built in any community for far less than the cost of putting in a public water and sewer system. Communities routinely borrow tens of millions for water and sewer and successfully manage those complex systems and pay back the debt. This is still America, and we still have the ability and know how to have the world's best infrastructure. The real question is why are so many communities willing to accept third world telecom services?
Submitted by acohill on Wed, 10/20/2010 - 10:13
Stop the Cap! has an article about the incumbent fight to kill the nation's most successful open access network: Utopia. Utopia's open access network has thousands of subscribers and fifteen providers on the network, including three TV providers. I've actually had the opportunity to see the Utopia TV provider offerings, and the picture quality of an all digital TV channel delivered via fiber is incredible.
Has Utopia had problems? Some early financial issues developed because the first firm hired to manage the network made some decisions that have since been corrected. The management firm is long gone, and Utopia took network management and outside plant maintenance in-house almost two years ago, with excellent results. The incumbents in the area--Comcast and Qwest--have been invited repeatedly to come on the network and offer their service to their existing customers, but instead, the first seem to want to simply try to force the competition out.
The situation is unfortunate, but is being repeated in other areas of the country, where incumbents are choosing to try to force monopoly pricing onto communities instead of competing in the marketplace.
Disclaimer: The Utah Infrastructure Agency has been a client of Design Nine.
Submitted by acohill on Wed, 10/20/2010 - 09:24
The EPA and ICMA (International City/County Management Association) have issued a very useful and readable report on "smart growth" in rural communities. However, the word "broadband" does not appear once in the entire report, and there no mention at all of the need for access to affordable high performance broadband services. I used to say that broadband infrastructure is the current day equivalent of water and sewer with respect to economic development, but I have switched to "paved roads." Why? Many rural residents have wells and septic tanks, and will never have public water and sewer. But those residents want and need the same kind of network access that is available in urban areas. Rural communities could not prosper in the twentieth century without paved roads in most areas (yes, I know that many rural residents still live on unpaved roads), but could you imagine a twentieth century town thriving with an unpaved Main Street? Today, there are many jobs and businesses that can thrive on dirt roads if affordable high performance broadband is available. Along with the many other good recommendations in this report, broadband needs to be included as a basic economic development goal.
Submitted by acohill on Mon, 09/20/2010 - 07:16
Many parts of rural England, like many rural areas of the U.S., have "little" broadband speeds of just a few hundred kilobits, as opposed to "big" broadband delivered via fiber with a capacity of a hundred megabits or more. A speed test was recently conducted in Yorkshire, England. The goal was to download a 300 megabtye file by a "little" broadband connection and see if that was faster than sending it 120 kilometers by pigeon.
The pigeon won. They strapped a USB thumb drive to the pigeon and it flew the 120 kilometers in one quarter of the time needed to download the file. Silly? Sure, sort of. But it really shows why little broadband is not enough for rural America.
Submitted by acohill on Mon, 09/13/2010 - 13:47
WiredWest is a municipal broadband project that includes 47 towns working together to build and operate a last-mile, fiber-to-the-premises network for Western Massachusetts communities unserved and underserved by high-speed broadband. The WiredWest project covers 1,445 square miles; more than 27,000 households; 3,000 businesses; and dozens of community institutions.
This week WiredWest town delegates chose a preferred governance structure to be submitted for approval by individual towns. This critical project milestone keeps the WiredWest effort on track and positioned to serve residents and businesses once the Massachusetts Broadband Institute middle mile fiber project is ready.
Research on potential forms of governance was conducted by counsel and consultants with the assistance of WiredWest’s Steering Committee and delegates. Municipal counsel was provided with support from Berkshire Regional Planning Commission and Franklin County Council of Governments. David Shaw, of Kirton & McConkie assisted as project counsel for WiredWest. Shaw is one of the country’s most experienced attorneys in community broadband. Design Nine has provided overall guidance and planning services for the project, including feasibility studies, needs assessment, GIS mapping, financial modeling, business planning, and network design.
A public co-operative enables WiredWest to move forward legally, practically and financially. Work on other aspects of the project, including engineering, business planning and financing, is proceeding simultaneously over the next several months, to ensure WiredWest is positioned to secure financing and begin construction as soon as enough towns officially join the Co-operative.
Submitted by acohill on Mon, 08/23/2010 - 13:26
Pete Ashdown, writing in The Salt Lake Tribune, discusses the reasoning behind community-owned broadband, in the context of the Utopia project, one of the country's biggest community broadband efforts. Here is a key portion of the article:
These interests go against broad long-term goals that infrastructure serves — facilitating economic exchange and the general welfare. If every airline was required to build their own airport and every shipping company needed their own road, America would be on par with Somalia as an economic force.
Fiber optics technology has vast capacity that allows multiple service providers of Internet, television and telephone to provide service to homes and businesses. UTOPIA and other open fiber optic networks throughout the world have demonstrated that this model provides a level playing field for competition, which in turn drives down prices for the customer and motivates quality service.
If your home is connected with UTOPIA fiber, you can choose from a variety of providers. If you are connected with Qwest ADSL2, you can choose from Qwest. If you are connected with Comcast cable, you can choose from Comcast. If either of these two companies raises its rates unexpectedly or gives you lousy service, your options are slim to nil for switching.
Utopia and other open access projects like (e.g. nDanville, The Wired Road, Palm Coast FiberNET, and others) are driving down the cost of telecom for residents, businesses, and institutions in their service areas, and service providers--especially smaller ones--are signing up to offer services. Some pundits insist that the open access model is "unproven," but their recommendation is to stick with what has NOT worked--the traditional retail triple play model. Community-owned retail triple play creates a one time decrease in telecom costs but lacks the choice and competition among providers that provides steady decreases in the cost of services and a steady increase in the kind and type of services that go far beyond the triple play of voice, TV, and Internet.
Submitted by acohill on Fri, 08/13/2010 - 17:06
In this article that speculates about an Apple TV upgrade, there is an interesting tidbit that validates what I and others have been saying for a long time: HD content chews up bandwidth:
"...In turn, consumers would see a similar increase in streaming requirements. Xbox Live can stream 1080p video, but it requires 8-10meg broadband, which leaves an awful lot of people out in the cold. It has the option of downloading instead, but if you're out in the sticks on a 2meg stream you're looking at more than eight hours to download your film at 1080p. You'd best plan your Friday night viewing before leaving for work on Friday morning."
The discussion about the Apple TV is whether or not Apple will include the ability to show movies in HD 1080p format. The short answer is, "Not likely," because streaming 1080p movies and TV shows over the Internet requires a massive chunk of bandwidth--8-10 megabits. And that's REAL bandwidth, not the marketing happy-talk that always begins with "....up to..."
Notice that if you wanted to download that movie over your average 2 megabit connection, it would take more than 8 hours! And if you are on a cable modem connection with a few of your neighbors also trying to do the same, it would take a little longer, like never (ditto with a wireless connection).
The answer is simple: we need to switch to open fiber.
Submitted by acohill on Wed, 08/11/2010 - 08:44
I went on a three day hike on the Appalachian Trail with one of my kids who is off to college in a few weeks. We had a glorious time hiking one of the most remote and isolated portions of the entire AT, which also happens to be one of the most scenic (right here in southwest Virginia). I had no laptop with me, no broadband access, and cellphone coverage so sketchy that we only managed a couple of quick text messages to the wife assuring her we had not fallen off a mountain.
For three days, I was off the grid, and it was wonderful. And nothing bad happened because I was not checking email three times a day. I need to do that more often, and I predict that in the near future, more of us will be deliberately planning and arranging "off the grid" broadband-free vacations. In past years, our beach trips have slowly become something less and less appealing to me because I seem to have to spend a couple of hours every day working. People call and say, "I know you are on vacation, but...." It's the "but..." that I have learned to dread. "...but could you just take a quick look at this document I just emailed?"
It was not that long ago that we measured response time to memos and reports in days and weeks, and now we measure it in hours. Just recently, I had someone call me up at 11 AM, quite upset, that I had not responded to an email they sent at 9 AM. On that particular day, I was out in the field and had had neither the time nor a broadband connection. But somehow we have come to assume 24/7/365 instant access to the 'net.
It's not healthy, and we should all take a deep breath and smell the flowers along a trail through the woods more often....without our cellphones and laptops.
Submitted by acohill on Fri, 08/06/2010 - 15:43
WiMax, which was going to solve everyone's broadband problems three years ago, may already be dead, without ever really being deployed in any meaningful way. This short article suggests that Clearwire is thinking seriously about switching from WiMax to LTE for its primary wireless deployment technology.
LTE (Long Term Evolution) is, depending a couple of different variants, a 3G or 4G cellular wireless technology that has much higher bandwidth than current 3G cellular wireless systems. As is usual, the makers of LTE equipment are touting very high (read: exaggerated) upload and download speeds that are always predicated with the handy phrase "up to..." But like all wireless technologies, you have to share that bandwidth with everyone else using the same tower/access point, so your mileage will vary greatly. A convenient rule of thumb for evaluating actual wireless bandwidth is to take the maximum "up to" number and divide by ten. The result is more likely what most users will see in practice.
Will LTE be an improvement? Very likely. But it means replacing your current wireless device with a new one, among other problems. Mobile phone manufacturers like these changes in technology, as it forces millions of people to buy new phones. But companies like AT&T are not even done building out adequate 3G networks using the "old" 3G technologies. The transition will be slow. And we will still all want fiber. Nobody is going to be delivering HD movies on demand over an LTE wireless network, no matter what the vendors claim the "up to" bandwidth is.
Submitted by acohill on Fri, 08/06/2010 - 09:08
Here is an article that alleges that Boulder, Colorado's SmartGridCity project is in deep trouble. The article has a long laundry list of problems, but what jumped out at me is the list of so-called "partners." If you look at the SmartGridCity Partners page, you can see the root problem of this project is too many cooks. Just the administrative overhead of supporting this list of high priced consulting firms would sink any project. And the descriptions that accompany each partner reads like one of those buzz-phrase generators you find online. Here are a sampling of the buzz phrases:
So you have at least seven companies with seven proprietary and very likely incompatible technology "solutions" that are going to use taxpayer dollars to try to do a mash up of their stuff that will somehow save money. These kinds of efforts never work, in part because if you start with seven complex technologies, it is impossible to make them less complex by combining them. Fifty years of software development studies have shown this over and over again. It's not that different than Fred Brooks' mythical man month treatise, in which he showed that adding more workers to a software project already late just makes it later--in large part because adding more workers makes the development process more complex. The same principle is likely at work here. Adding more complex power management software to an already complex design makes it even more complex and, as study after study has shown, more error prone.
Here at Design Nine, we call ourselves "broadband architects" or "information architects." We work the way the traditional architect works--we do a clean, coherent high level design for our clients first, develop the financing and funding strategies needed to show the client how it will pay for itself, and then and only then do we go out to vendors.
My guess is that Smart Grid City ended up with seven or more "design" firms all trying to gain an advantage for their own stuff, and Boulder ended up with a mess. It's as if you wanted a house built, and instead of having an architect produce the design and supervise the construction, you told the plumber, the carpenters, the electrician, and the drywall guys to get together and come up with something. It's called "design by committee," and it is never pretty.
Submitted by acohill on Thu, 08/05/2010 - 08:17
A common theme, when discussing the financing of broadband, is to claim that the open access business model "has not been proven." True open access has only been around in the U.S. for about three years, and the opponents of open access are creating a double standard. Apparently, to "prove" open access works, communities that take that route have to be in the black within a year or two, and really, it would be better, apparently, if they were in the black on day one.
What's the double standard? The double standard is that private sector telecom companies take years to get into the black, and routinely run up billions in debt that won't be paid back for a long period of time. That's apparently okay--communities will be held to a higher standard, though.
There is a second, very sly argument running alongside the "prove it works" argument. It is this: "Open access is bad because it has not been proven." Okay...let's accept that premise for a moment and now let us ask the question that no one asks..."What business model should be used instead?" The business model that open access opponents want to use is the tired old monopoly triple play, which has, over the past fifteen years, failed utterly to get "big" broadband to U.S. residents and businesses.
So the open access opponents want everyone to use a "proven" approach: triple play, which has been proven not to work. Sly, no?
Open access projects like Utopia, nDanville, The Wired Road, and Palm Coast FiberNET are thriving--in those communities, increased competition has brought lower prices for telecom services, and the efforts are generating enough funds that all those communities continue to expand their networks to more customers.
Submitted by acohill on Tue, 08/03/2010 - 09:58
Via an Akamai report (registration is required to get the report), Latvia has moved up to fifth place in the average bandwidth rankings worldwide. The U.S. is down at number 22, with a net negative drop of about 1% in bandwidth over the last quarter and 2.5% drop in bandwidth over the past year. According to Akamai, the average broadband connection in the U.S. is about 3.8 megabits/second, which would reflect the fact that the cable companies dominate the broadband marketplace in the U.S.
It is worth noting that the FCC just set a new standard for the definition of "broadband," which is 4 megabits down and 1 megabit up. This reflects the continuing focus on broadband as an "entertainment" service (that's what some cable companies call it) rather than a business service. With more people and businesses trying to work out of the home, symmetric bandwidth has become essential to economic development. The continuing acceptance of a bigger pipe into homes and businesses and a much smaller pipe upstream reflects a lack of understanding about business and job needs for broadband services, which need the symmetric bandwidth.
Submitted by acohill on Fri, 07/23/2010 - 13:52
Via the Twitter feed BroadbandReport, a new study suggests that the U.S. ranks 23rd in the world in broadband deployment. Strategy Analytics, the company that developed the report, is using a new set of metrics that look at five different indices to come up with the ranking.
The U.S. will continue to rank low for years, as the market here is very different here than in the rest of the world. Most of other countries are much smaller, making the scale of the problem much smaller. South Korea, which ranks first in the study, is smaller than almost every state in the U.S. And most other countries take a government-heavy, top down approach to broadband, which works better in other places because there is really only federal government and local governments. Here in the U.S., we have a middle layer of government (states) that are largely autonomous. And I'm not arguing for a more extensive top-down approach in the U.S. Topography, geography, and local business conditions vary so widely in the U.S. that dictating a one size fits all approach is likely to have a lot of unintended consequences.
Submitted by acohill on Thu, 07/22/2010 - 10:20
I had a conversation earlier this week with a well-connected business person who is in the business of building data centers. The two top criteria his firm uses to identify communities in which to locate data centers is power and fiber. What he told me is that for the size of data center he typically builds (50,000 to 100,000 square feet) they are looking for power from two separate sub-stations, and that power from two separate grids is even better. Few communities get fed from two grids, but it is more likely to be able to get power to some industrial/commercial sites from two different sub-stations. Even though these data centers have backup generators on-site, the 15 to 30 megawatts these facility use make dual feeds desirable.
Power is something all communities have, and if extra capacity is needed, it can usually be added easily if the funds are available to support new transmission lines and/or new transformers. Note that the lead time on large power transformers is one year or more, so a strategy of "If someone wants extra power, we'll just build it," may not be leading with your best foot forward, as they may move your community off the short list in favor of communities that have already addressed power.
So that leaves fiber as a key discriminator in relocation decisions. Planning and building local fiber infrastructure can take six to 12 months at a minimum, so if you want to attract data centers, you want open access fiber assets already in place and ready to use.
Some data points: the massive Google data center in rural Washington state was placed there because fiber assets were already in place. The fact that cheap power was available was a secondary consideration. Danville, Virginia won a new, large data center recently because they had both fiber in place and reliable power. Fiber gives communities a competitive edge for business attraction.
Submitted by acohill on Wed, 07/14/2010 - 08:12
Via Jon Hunt's excellent Broadband Policy Watch Web site, Google has rolled out a Google Fiber for Communities Web site. There is not much new information; Google is still promising that they will select a community before the end of the year. Of interest is the focus on microtrenching. This is a technology that Design Nine has been using for several years. We particularly like the Teraspan products. We have used that very successfully in The Wired Road. The picture shows the slot you cut in the street, and then you drop the Teraspan duct which already has the fiber cable enclosed.
Submitted by acohill on Tue, 07/06/2010 - 09:13
Broadband Properties has published my article The Third Way for Broadband. This provides a concise description of how and why open access business models work for broadband networks. Note that the open access business model is NOT inherently one that requires a community-owned network. A private sector broadband provider, including an incumbent (e.g. Verizon, Comcast, etc.) could also adopt this model and do very well financially.
Submitted by acohill on Fri, 06/11/2010 - 15:43
The nDanville fiber network is almost three years old, and is beginning to get national recognition here. Design Nine has been working with the City of Danville on this effort since 2006. We did the early business and financial planning, vendor selection, and open access network design. More about nDanville is available on their Web site.
Submitted by acohill on Tue, 06/08/2010 - 15:03
Jeff Daily of App-Rising makes an important point in this article that I have been writing about for some time: "broadband" and "the Internet" are not the same thing. Broadband is the network, the transport system, the road. The Internet is just one of many services that can be transported over that road. Unfortunately, legislators don't always understand the distinction, and many incumbents are happy to feed the confusion to get state and Federal rules designed to prop up their monopolies.
Design Nine provides visionary broadband architecture and engineering services to our clients. We have over seventy years of staff experience with telecom and community broadband-more than any other company in the United States.
We have a full range of broadband and telecom planning, design, and project management services.
Free Fiber to the Home
Save NC Broadband
Blandin on Broadband
Intelligent Community Forum
FCC Broadband Blog
KGP Broadband Stimulus
Ars Technica Tech Policy
Bill St. Arnaud
Stop the Cap
Broadband Policy Watch
Lafayette Pro Fiber