Blog Post

Maybe you don’t need a gig. Wireless might bridge the broadband gap

Not every community needs a fiber-to-the-home network. Not every home needs a gig to their doorstep. In our gigabit crazy era this statement might seem like a step backward, but take note that there’s more than one option for delivering the speed consumers and businesses need. Even Google is hinting that some form of wireless might become part of its goodie bag of services.

When people fixate on one technology to the exclusion of all else the people governing cities can make wrong choices that hurt or hinder communities’ ability to fully benefit from broadband. When investing in technology, users’ needs should dictate technology choices, not media hype. Two other recent broadband developments indicate some broadband decision makers should step back for a minute and re-assess their options.

Gigabit wireless goes high profile

RST, a new regional ISP, announced it had quietly built and acquired a 3,100-mile 100-gig fiber middle mile infrastructure throughout the state of North Carolina. However, it plans to deliver a 1-gigabit last mile service there and in South Carolina, mostly using Wi-Fi with fiber options available on demand. In Utah, home security and automation company Vivint threw its hat into the gigabit ring with plans to connect Utah homes wirelessly using gigabit Wi-Fi on rooftops to create a high-capacity mesh network built on customers’ rooftops.

The gigabit picture is still developing, and no rules are set in stone, so why not extensively evaluate all available options? Do consumers really care whether their connection comes through a wire or wirelessly as long as it’s fast enough to meet their needs, and guaranteed to be secure, reliable and affordable?

“We’ve come up with a unique approach in that we built a 100-gig fiber infrastructure, but we’re going with the new Wi-Fi standard – 802.11ac – to enable us in smaller communities and even big cities to reach out with 1 gig symmetrical speed,” stated RST Co-Founder Dan Limerick in an interview. “The major excuse big providers have given for not building out the entire U. S. with high-speed Internet access is the expense of it. With the technology we’re using, that’s a fallacy and we’re going demonstrate it very soon.”

Some communities already are benefitting by keeping their options open. Ottumwa, Iowa, for example, after doing considerable research is implementing a small pilot that will use a high-speed point-to-point wireless connection to link a seniors medical and assisted living facility into the project rather than build an extra two miles of fiber through an area with few potential additional customers. Furthermore, the project team may evaluate high-speed wireless (over 15 Mbps) along the fiber route to provide residential coverage.

Is less than a gig ok?

Some people are skeptical of the ability to deliver a gig wirelessly. “I know the limitations it has,” states Jesse Harris, Editor of FreeUTOPIA, which tracks broadband developments in Utah. “Even if Vivint has figured out a few tricks, there are some laws of physics they simply can’t avoid. Going with a mesh network could introduce a lot of latency, and unlicensed spectrum is subject to all kinds of interference that they can’t control.”

Matt Larsen, owner of wireless ISP Vistabeam, shares some skepticism of gigabit Wi-Fi but wants to see how Vivint and RST perform as they sign up lots of subscribers. He has seen point-to-point wireless consistently deliver reliable speeds in the 20-30 Mbps range and above, which is sufficient for many homes. “Gigabit speed is important for delivering the network capacity needed to ensure scores of individual subscribers can get these speeds simultaneously. “

Larsen believes, as does Limerick, another huge value wireless delivers is faster speed and lower costs to build broadband to under-served rural and urban communities. He said, “A WISP can deploy a five-mile network with line-of-site, point-to-point wireless in one or two days at a cost of between $4,000 and $5,000 using unlicensed spectrum. There are many variables in pricing for fiber buildouts, but a five-mile fiber run can cost at least tens of thousands of dollars.”

Every community is different in terms of demographics, broadband needs, geography, etc. The lesson people should take from RST and Vivint is that a thorough, objective evaluation of all available technologies is in their best short-term and long-term financial interests. For example, wireless last mile networks can meet peoples’ needs now, then become backup for fiber network later.

Craig Settles is a consultant who helps organizations develop broadband strategies, host of radio talk show Gigabit Nation and a broadband industry analyst. Follow him on Twitter (@cjsettles) or via his blog.

12 Responses to “Maybe you don’t need a gig. Wireless might bridge the broadband gap”

  1. Ed Dodds

    I think if Netflix keeps pushing the peer to peer possibilities which they may want for business model profitiability purposes now that there are officially tiered speed channels they may have to step into the urban wireless fray. Maybe not.

  2. Nicely done! I am trying to build up a database of instances of wireless substitution. Ideally, we could get our hands on consumer data, which would permit estimates of cross-price elasticities between wireless and wireline broadband connections. Economists recognize that it only takes a “critical share” of consumers at the margin to switch to all-wireless solutions to discipline the prices of cable modem providers. We may not be there today, but we should be there soon!

  3. Michael Elling

    Craig, you touched on 2 players in the article that (can) bring significant WAN side scale in terms of their mid-mile assets; Google and RST. The trick to last mile pricing is both understanding future demand and adjusting pricing ex ante to fill the pipes. In other words, it can become self-fulfilling as we saw with other “digitization” waves in long-distance voice, data and wireless in the 1980s and 1990s. Only the “last-mile” remains to be digitized and for pricing elasticity to become fully apparent.

    That requires the access provider understanding their marginal cost based on varying (iterative) supply and demand assumptions at every layer and boundary point. Basically most mid-mile govt grantees should be pricing bandwidth 90% below the current market price by taking a revenue share with (competitive) last mile providers. That will fill the mid-mile providers “pipes” more than they can imagine by creating generative and least-cost last mile solutions at the same time stimulating demand the 3 ways I outline in the above link. I liken their model today to selling an ocean of fiber one drop at a time. Doesn’t work. Need a rethink. Wireless (especially the smartphone) does that.

    Pricing which reflects marginal cost will also facilitate multiple, redundant wired/wireless topologies as commercial/institutional business systems gravitate to the digital world for HD collaboration, multi-media and real-time ecommerce, 4k content delivery, and the internet of things. And business ecosystems always demand backup/insurance for QoS and reliability as well as asynchronous pathways for safety and security (don’t want Google or NSA evesdropping, or malicious elements getting control of an asset or link). As Marc Zionts of Aicent says in the 28th minute of this podcast, it’s the coming “IT-ization” of the service provider business model.

  4. Re premises service, speed is only an issue because much of U.S. Internet infrastructure is deficient and at least decade behind where it should be. Speed became an issue because for many premises users, the quality of the customer experience using data, video and voice has been negative due to poor throughput (speed and latency) due to inadequate wireline infrastructure.

    Fiber to the premise both resolves the throughput problem while at the same time making speed itself less relevant– and by extension the gigabit discussion — since fiber offers enormous headroom for faster speeds.

    As noted in the article, wireless can work over short distances, but cannot substitute over the long run for fiber to the premise.

  5. Fiberman

    20 years ago we asked – who needs more than 1 meg – now we know that one meg isn’t enough. But it is not question about speed it is question about limitations – why there are limits in speed and why we have to wait internet services to open. People need instant internet with no limits and that’s why the only solution is fiber and the line must be symmetrical – in the long run fiber is the most economic solution to provide instant internet with no speed limitations in bandwith to every home. Only limit is the speed of the light.

  6. wispowner

    elfonblog, I’m looking forward to seeing your company roll out this fiber to every home. Please keep us all informed of your progress. Since you seem to know how easy it is, I’ll be waiting with bated breath.

  7. elfonblog

    “who needs a gig? (1-2-3 kick) who needs a gig? (promenade left, tapdance right) who needs a gig? (circle ’round and start again)”

    I’m getting sick of this narrative! It’s not about whether I need it or not, so stop using those words! Corporate sock puppets be trying to talk us out of pressuring them to give us what we deserve.

    Gigabit service costs less than what we’ve been paying for service 1/100th as fast. Since the economics works out that I can afford this amount, then by all means, give me my unlimited, symmetric gigabit service for $70/mo. It’s asinine to appeal to my sense of “fairness” and make me agree to pay the monopoly telcos their 950% price mark-up and stop being such a “greedy pig”. Poor telcos! They operate on a 95% profit margin. This means 5% of the revenue pays everything to maintain the network and provide the services, and the 95% remaining revenue goes to the money bin.

    Now, I know this article focuses on using wireless as a stopgap to reach rural areas where fiber doesn’t currently exist, but let’s stop debating whether wireless can replace fiber for use at fixed locations. NO. The economics of running fiber to homes is that it is cheaper in the long run. Wireless is more PROFITABLE in the short run. Wireless is ridiculously expensive in the short and long run. Wireless cannot be upgraded as easily and cheaply as the signal coming in over a fiber, nor ever have the range and capacity of fiber.

    So let’s terminate that corporate narrative once and for all. It’s not fair to fool or bully customers into blowing their broadband budget on the smallest amount they’ll settle for, in a monopoly market.

    I fully support the wireless broadband movement, but doubt it will ever be appropriate for anything other than temporary or fully mobile applications. Gigabit wifi is a welcome option, but it’s still going to be limited to short-range applications (homes, offices) unless the problems of mesh latency and site access are somehow solved.

    • The point here is not a corporate argument or a red herring for advancing an incumbent position. The argument is for not going into a broadband project with decisions already made before a thorough needs analysis is done.

      Every community is different and there are a host of variables that influence or directly impact what ultimate decisions make sense for the respective communities. If you only come at this from one narrow position made before any legwork is done, you could screw up the project’s potential benefit.

      Smart people will ask the obvious questions, the devil”s advocate questions, the seemingly unrelated questions to make sure this multi-million dollar decision is the right one for their community. And, if they’re really smart, they’ll do an effective pilot project to test those assumptions – wired, wireless, cable, some combination. Taking a month or two extra to work through options is a wise thing to do.

    • mattlarsen

      The answer is a combination of both wireless and fiber – wireless/fiber hybrid networks. Start out by getting fiber into a neighborhood and using fixed wireless to distribute the last mile/last 100 feet. That will scale up to and beyond cable speeds. Then use the cash flow from the wireless deployment to pay for the fiber upgrades over a few years. This gets people the service and competition that they want NOW, and fiber eventually and it can be done without government handouts and risky loans to telcos or municipalities that may not be able to sustain the business model.

      Fiber only is an economic disaster and takes a very long time to deploy. I have done both fiber and wireless, and I love fiber as much as anyone else, but it is very expensive and FTTH is very difficult to fund unless you have very patient investors.

      Hybrid fiber/wireless networks are a win/win.

      Matt Larsen

  8. Paco Loco

    Viva radio! TV White Spaces at 700 MHz. Is ideal for rural WISPs but the greedy US gov/ FCC Will eventually auction it all off to big cellcos.