Pew Internet survey flawed by badly outdated, retro perspective

With the relentless pace of Internet bandwidth demand growth to support multiple services including video, voice and Web-based services as well as a portable devices used in the home, there is near consensus that only fiber to the premises infrastructure will be able to accommodate the demand going forward.

That’s why I’m taken aback to continue to see surveys such as this one issued today by the Pew Research Center’s Internet & American Life Project that take a decidedly retrospective view of telecommunications services with their late 1990s distinction between narrowband (dialup) Internet connectivity and “high speed” broadband connections. 

Dialup service is obsolete and can no longer be considered a useful form of premises Internet connectivity. Had this survey been done in 2000 when the distinction between narrowband and broadband was still relevant, the distinction might have meant something. In 2013, it is a distinction without a difference. 

The other major contextual problem with a survey like this is it concentrates only on computer-based services such as Web browsing and email. That’s also a major flaw in the survey. The Internet now delivers video and voice services including applications such as online learning, videoconferencing and telemedicine – none of which are truly usable via a dialup service.

Why We Should Build a National Internet System Under the National Highway System - Eric Jaffe - The Atlantic Cities

Why We Should Build a National Internet System Under the National Highway System - Eric Jaffe - The Atlantic Cities: The National Broadband Plan of 2009, for instance, was mostly limited to policy recommendations and failed to encourage competition (which explains why incumbent providers like it so much). Proposed legislation requiring highway projects to install broadband conduit hasn't made it too far. The Obama Administration did issue an executive order last year calling for a "dig once" [PDF] policy to help promote broadband-highway coupling, but that still relies on private enterprise to do what it hasn't done to date: lay fiber everywhere.

So why not make the whole national internet system a public one, like the national highway system before it? At a time when elected officials are struggling to find a truly federal transportation goal, the concept might serve as a welcome rallying point. The government could sell some of its broadcast spectrum to foot the bill, but the user-pay model could probably work well, too — especially since people don't suffer the illusion that Internet access is free, unlike they do with roads.

"There is a really interesting parallel between transportation and broadband," says Lennett. "In the 20th century we needed to move cars, and in the 21st century we need to move bits.

Governments Should Focus on Infrastructure Despite False Statistics Peddled by NY Times and Others | community broadband networks

Governments Should Focus on Infrastructure Despite False Statistics Peddled by NY Times and Others | community broadband networks

Excellent commentary by Christopher Mitchell. A must read.

Satellite broadband: can it light up the UK's broadband blackspots? | News | TechRadar

Satellite broadband: can it light up the UK's broadband blackspots? | News | TechRadar: Due to the distance the signal travels, latencies never dropped below 700ms and hovered around the 800ms mark. Even with predictive caching that makes web browsing speedy, there's always that near-second delay traversing pages. It's not annoying enough to stop you browsing, but it just doesn't feel as snappy as a landline internet connection.

Despite new sooper dooper "Surfbeam" technology, latency remains sub par as this story shows and bandwidth is costly and rationed. This item appeared the same day as this ridiculous story on Google's O3b satellite venture that will supposedly provide 1 gigabit speeds via medium orbit satellites. And at latencies of less than 150 milliseconds, according to this IDG News Service account.

I'm not buying it. Satellite Internet sucks, period. It cannot support reliable voice or real time video connections or provide a high quality Internet connectivity user experience. Google should scuttle this misadventure and instead partner with community fiber projects instead of perpetuating this substandard Internet connection scheme to as a poor substitute to badly needed fiber to the premise infrastructure. 

California unlikely to subsidize community fiber Internet infrastructure over near term

The California Public Utilities Commission’s (CPUC) construction subsidy fund for Internet infrastructure won’t likely help offset the cost of building community owned fiber to the premise networks.

The CPUC’s California Advanced Services Fund (CASF) limits grant and loan subsidies to infrastructure projects that would serve either an “unserved area,” defined in CPUC Decision 12-02-015 as not served by any form of wireline or wireless facilities-based broadband such that Internet connectivity is available only through dial-up service or an “underserved” area defined as an “where broadband is available, but no facilities-based provider offers service meeting the benchmark speeds of at least three megabits per second (mbps) download and at least one mbps upload.” The CPUC retroactively revised the definition in 2012 resolutions T-17362 and T-17369 as areas “where broadband is available, but no wireline or wireless facilities-based provider offers service at advertised speeds of at least 6 mbps download and 1.5 mbps upload.”

Under either definition, both fixed and mobile wireless providers could block CASF funding of a community fiber project. And under the definition adopted in the 2012 resolutions, they wouldn’t even have to actually provide service to an area. They could merely claim they advertised service there at the specified 6/1.5 Mbs speeds.

Senate Bill 740, legislation re-authorizing the CASF that’s making its way to the desk of Gov. Jerry Brown incorporates by reference the definitions of unserved and undeserved areas in Decision 12-02-015.

The bill would also give incumbent wireline providers that have not built out their networks to serve all premises effective veto power over any community-based project to reach underserved households -- typically those in areas out of reach of DSL or cable Internet service or having access to slow DSL in areas where aging, poor quality copper cable plant (illustrated in the photo below) cannot support higher speeds. The bill bars funding of these projects “until after any existing facilities-based provider has an opportunity to demonstrate to the commission that it will, within a reasonable timeframe, upgrade existing service.” 

"Reasonable timeframe" isn’t defined in the bill and thus would likely be defined by incumbent telcos that told regulators and consumers since the early 2000s that they were building out their DSL service to reach them. (They’re still waiting more than a decade later, providing an operative definition of what's reasonable). The bill would also give incumbent telcos and cablecos the ability to stymie community fiber projects built by local governments simply by applying for CASF funding.

The rural "digital divide" isn't the same in UK, US

Millions miss-out as Britain's broadband divide reaches record levels - Yahoo! Finance UK: Telecoms regulator Ofcom warned the difference between the 'haves' and 'have-nots' would also get worse before it gets better as telecom and pay-TV giants focus investment on next generation "superfast" fibre networks in Britain's biggest towns.

Figures revealed by Ofcom today showed the average internet connection speed in urban areas is now 26.4Mb per second. In rural areas it's just 9.9Mb. Rural speeds have more than doubled since 2011 but households in the countryside now trail city dwellers by an unprecedented 16.5Mb per second.

Americans living in rural, quasi-rural and exurban locales and stuck with dialup or satellite or forced to make do with costly, data capped mobile broadband for their premises Internet service would find this account puzzling. For them, having access to nearly 10Mbs throughput would hardly be considered deprivation at the present time.

Mobile isn't premise service

AT&T’s latest home broadband service isn’t DSL or fiber. It’s LTE — Tech News and Analysis: The idea of using a mobile network to connect homes that have either no or slow broadband access is definitely an admirable idea, but it has its limitations. Today’s mobile networks simply aren’t designed for the intense data demands of are increasingly hungry home broadband appliances — they have limited capacity and that capacity must be shared with all of the other users on the network. That’s why the per-gigabyte costs of mobile broadband are so much higher.

An excellent point in this GigaOM article that puts this service in the proper perspective.  It's designed for mobile and not premises services. Offering a mobile service to a premises does not make it a premises service.

A few paragraphs down, the article notes:

In fact, AT&T is trying to eliminate the distinction between residential and mobile service entirely.

Nice try, but it won't work.  Mobile is mobile. Premises is premises.

Petitions or "broadband mapping," the end result is the same: Bumpkes

Mark dance sundridge broadband campaign pointless | This is Kent: Jane Hunter from the Westerham Town Partnership has campaigned at every opportunity to add names to the petition but she says she is not surprised to hear the votes will not influence the rollout.

"They have been stringing us along for no reason," she said. "They don't want people hassling them so they haven't told us the reality of the situation.

"And while we have been given figures by the council they never answer any of our questions on exactly what the criteria were for which areas would be part of the superfast project.

Across the Atlantic in the United States, petition drives similar to this one in the U.K. -- some dating back years and encouraged by incumbent providers and misguided demand aggregators -- proved equally pointless. Signatures on petitions can't overcome market failure and are just as futile as "mapping" broadband not spots will make them disappear. The result at the end of the effort is the same: bumpkes.

How BDUK bungled Britain’s next-gen broadband rollout | PC Pro blog

Interesting dispatch from the UK that portrays the dominant incumbent telecoms provider, BT, as favoring American AT&T U-Verse-style FTTC (Fiber To the Cabinet) over Fiber to the Premise (FTTP), apparently to avoid the higher cost of the latter network architecture.

The UK Government entity charged with overseeing that nation's Internet infrastructure program also allegedly ruled out fixed terrestrial wireless as a viable premises service option. That's consistent with the first point since fiber would have to be deployed to bring it very close to premises in order to achieve high wireless throughput.

The 2 key inaction risks facing community fiber projects

Creative risk taking is essential to success in any goal where the stakes are high. Thoughtless risks are destructive, of course, but perhaps even more wasteful is thoughtless caution which prompts inaction and promotes failure to seize opportunity.

  -- Gary Ryan Blair 

Communities contemplating fiber Internet infrastructure projects should keep in mind that there are risks -- negative impacts -- associated both with taking action as well as not taking action.  The latter risk -- termed inaction risk -- is perhaps one of the most threatening and pervasive risks.  For some regions and communities, that risk is being left permanently off the modern Internet grid and unable to realize the benefits it offers for government, public safety, health and education, economic development and transportation demand mitigation. 

Milo Medin, Google's vice president of access services, laid out two major underlying rationales explaining why communities needlessly run the risk of inaction in his address to the 2013 Broadband Communities Summit. 

1.  The unswerving belief despite more than a decade of market failure that incumbent legacy telephone and cable companies will upgrade and build out their infrastructures to serve all premises.  Here's what Medin had to say on that point:

Part of the reason the U.S. is falling behind is that most cities haven’t been intentional about their broadband infrastructure. Cities know they have to make sure the water system works and scales to support growth, the roads are maintained and built, garbage is collected properly. But often, they think broadband is something that the phone company or the cable company will take care of for them and they can ignore it, or that the FCC will make sure the appropriate incentives are put into place to drive competition and upgrades. Depending on those processes is how we got into the situation we’re in today.

2. The misguided belief that wireless services have obsoleted fiber networks. Medin explains:

Some argue that fiber networks are not really needed because of wireless network growth. As an engineer, quite honestly, this kind of talk makes my brain hurt. Wireless network growth is driven by fiber. All those base stations that smartphones connect to are increasingly connected by fiber because, as speeds go up, fiber is required to carry that kind of traffic. Copper just won’t do for modern wireless networks.

Cisco and others expect wireless data to grow by a factor of 50 in the next few years, and you’re not going to be able to solve that kind of growth by throwing more spectrum at it. You’re going to have to reduce the size of the cells, shrinking them, reducing the number of users that are being served by a given base station. And that means a lot more cell sites and a lot more fiber to feed those cell sites. In the limit, the future of mobile is going to look a lot like Wi-Fi: tons of small cell sites connected by a wireline network, connected by fiber – and that’s just physics, folks.

The full Broadband Communities article excerpting Medin's speech can be viewed here and here (pdf). 

The 3 big U.S. Internet infrastructure policy choices

The United States now has three major policy options on the build out of Internet infrastructure to serve all American homes, businesses and institutions:

1. Continuation of the status quo of investor-owned Internet infrastructure and associated private market failure that will leave significant numbers of premises lacking affordable Internet access over the long term and potentially permanently.
   
2. A well funded federal aid program including technical assistance grants for community fiber to the premise network construction projects, funded by existing programs such as the U.S. Department of Agriculture's Rural Utilities Service, a program jointly administered by multiple agencies or by a newly created, dedicated agency.  In addition, federal preemption of state laws barring local governments from constructing, owning or operating Internet infrastructure.
   
3. De-privatization of all Internet infrastructure (either immediately or over a period of years) combined with a fast track federal construction project to build out fiber to serve all U.S. premises, similar to the 1950s interstate highway project.

Please add your comments.  Which do you favor and why?

The U.S. Needs A Federal-Aid Highway Act For Affordable Broadband -- Now - Forbes

The U.S. Needs A Federal-Aid Highway Act For Affordable Broadband -- Now - Forbes

Digital media veteran Gary Myer urges a massive Internet stimulus program that goes far beyond the $4.5 billion allocated for Internet infrastructure in the American Investment and Recovery Act of 2009.  (Most of that money went toward middle mile infrastructure that typically left residences and small businesses off the net).

Myer as well as some of the commentators on his Forbes piece point out getting fiber to every U.S. doorstep not only would create a lot of jobs since a large majority of the cost is labor.  It would also make the U.S. network more valuable since more would be connected to it, replacing the current dysfunctional, hodge podge of disparate legacy cable television and telephone company networks whose high cost business models fail outside of densely populated areas.

Myer also puts to rest the fanciful, wishful thinking that cell phone networks obviate the need for premises Internet connections.  Those networks are designed for lower bandwidth mobile voice and data and lack the capacity and reliability to serve as primary premises connections. Those bandwidth caps on mobile service exist for a reason.

Fiber to the Home Council : Blogs : Telcos Saving Serious Money by Upgrading to FTTH, Survey Finds

Fiber to the Home Council : Blogs : Telcos Saving Serious Money by Upgrading to FTTH, Survey Finds: (WASHINGTON) – Small and medium-sized telephone companies that have upgraded their networks to all-fiber are reporting operational cost savings averaging 20 percent annually, according to a study commissioned by the Fiber to the Home Council Americas, a non-profit group of nearly 300 companies and organizations dedicated to expanding the availability of ultra high speed, all-fiber broadband.

The survey of more than 350 telecommunications providers across North America, conducted by the market analyst firm RVA LLC, also pointed to a steady drumbeat of FTTH deployment activity, with the number of homes that can access FTTH networks increasing by 17.6 percent over a year ago to 22.7 million.

While consumers served by these smaller telcos will benefit (as will the telcos with lower OPex costs), there are many more served by large telcos like AT&T and Verizon who won't.  Neither company is upgrading its copper plant to fiber to the premise.

Israel's 1Gbps fiber will show the world what superfast broadband can really do: Cisco CEO | ZDNet

The planned deployment of gigabit fiber to the premise (FTTP) infrastructure in Israel as described in this ZDNet article Israel's 1Gbps fiber will show the world what superfast broadband can really do: Cisco CEO | ZDNet illustrates a trend in much of the industrialized world.  As the Internet grows into an all purpose communications platform, it will overtake and obsolete telephone and cable companies that built their business models on a pre-Internet world.  Some excerpts:

 The network should be completely operational in five to seven years, giving Israelis the opportunity to surf the net with downlinks of 1Gbps, ten times faster than anything the local competition — chiefly the Bezeq phone company and HOT cable service provider — can provide with their FTTN (fiber to the node) network, which delivers a top speed of 100Mbps.

Chambers predicts the network will bring in major changes: healthcare where doctors are connected instantly to providers' and hospitals' databases, with all records kept electronically and updated constantly; an education-anywhere system, where students can learn at home, in class, or elsewhere, communicating with teachers and fellow students over the internet; safer roads and streets (a major issue in road accident-prone Israel), with traffic authorities able to keep better tabs on speeders and unsafe drivers; and a proliferation of "internet of things" technology, with sensors keeping air conditioners, refrigerators, washing machines, front doors, and more connected to systems than can enable better and more efficient allocation of electricity and other resources. In a few years, all of this should be in place, according to Chambers.

Google's Project Loon no cure for the common Internet not spot

Google's experimental Project Loon unveiled this week that envisions a fleet of high altitude balloons providing Internet connectivity is likely to benefit only the most remote and undeveloped areas of the globe.  Fittingly, remote New Zealand was the site of the first experimental deployment of these 'loons, mate.

The technology won't be a panacea for more populated parts of the globe in developed nations plagued by incomplete wireline Internet infrastructure that serves only some residences and businesses while others adjacent are left off the Internet grid.

Nor can it provide sufficient throughput to support the rapid growth in bandwidth demand driven by video and the use of multiple devices common nowadays in many households.  According to Google, the high altitude digital dirigibles will provide connectivity at speeds comparable to 3G, the legacy mobile wireless service now being superseded by higher bandwidth 4G mobile wireless.

NYT op-ed complaining of low Internet subscribership undermines author's credibility

No Country for Slow Broadband - NYTimes.com: The major causes for low subscribership, as extensive survey research shows, are low interest in the Internet and minimal digital literacy. And too many American households lack the money or interest to buy a computer. As a result, more Americans subscribe to cable TV and cellphones than to Internet service. Our broadband subscription rate is 70 percent, but could easily surpass 90 percent if computer ownership and digital literacy were widespread.

So argues Richard Bennett, senior fellow at the Information Technology and Innovation Foundation.  There is a big hole in this argument.  Information and communications services are universally verging toward employing Internet Protocol (IP) to deliver them.  People are increasingly viewing video content delivered over the Internet to making voice calls using Voice Over Internet Protocol (VOIP). 

In segmenting discrete services and citing just one type of Internet-enabled service (personal computing), one has to question why someone with Bennett's level of knowledge would even make this point, undermining his credibility and suggesting a hidden agenda.

Obama administration's ConnectED program only half a solution

President Calls for High-Speed Broadband in 99% of Schools/Libraries - 2013-06-06 15:21:00 | Broadcasting & Cable: The White House Thursday announced a new initiative to get high-speed broadband to America's schools and libraries.

The so-called ConnectED program has a goal of connecting 99% of students to high-speed wired and wireless broadband (speeds of no less than 100 Mbps and preferably 1 Gbps) within five years. The president called on the FCC and National Telecommunications and Information Administration to "modernize and leverage" its E-rate program to achieve that goal. The E-rate program provides discounted broadband service to schools and libraries through the Universal Service Fund.

The President's plan while laudable is only a partial solution.  Learning now takes place both at school and in the home -- what educators describe as "blended learning."  Jeremy Meyers, superintendent of the El Dorado County (California) Office of Education, wrote about the emerging educational method in which pupils do much of their learning and class projects outside of the classroom via the Internet – arguably the world’s biggest and best stocked library.  Back in the classroom, their teachers review their projects, answer questions and lead discussions.

As Meyers notes, blended learning requires good Internet connectivity both at schools and at students' homes.  However, too many homes in Meyers' district lack even basic Internet service.  "El Dorado County faces a special challenge that is assuming greater urgency each year: How to bring all our households into broadband Internet access in a cost-effective manner," Meyers wrote. "Having large Internet 'dead zones' is not acceptable in today’s world of connectivity. It limits us academically and hurts us economically."

Are you listening, Mr. President?