Blog Post

Google's Fiber Network Could Foil ISPs and Fuel Innovation

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Google (s goog) said today it plans to build an experimental fiber-to-the-home network in select areas of the country that would offer speeds of around 1 Gigabit per second. It says it plans to serve between 50,000 and 500,000 people and will offer the service for a “competitive cost.” The company is currently seeking a response to its request for information (RFI) by March 26 from municipalities that may want such a service for their citizens, but let me be the first to put my hometown of Austin, Texas, in the running.

When asked about the characteristics of those communities, a Google spokeswoman emailed the following response:

Above all, we’re interested in deploying our network efficiently and quickly, and are hoping to identify interested community partners that will work with us to achieve this goal. To that end, we’ll use our RFI to identify interested communities and to assess local factors that will impact the efficiency and speed of our deployment, such as the level of community support, local resources, weather conditions, approved construction methods and local regulatory issues. We will also take into account broadband availability and speeds that are already offered to users within a community.

Google’s announcement, which has been five years in the making, could positively shift the telecommunications landscape if it leads to new services that galvanize the FCC, communities and consumers to start demanding faster broadband. It also creates a potential testbed for innovative services that rely on broadband as a platform to work — benefiting entrepreneurs and those who invest in them.

With its web DNA and commitment to openness, Google will likely attract entrepreneurs to its network that are willing to try something new on the services front. Presumably it will also offer a faster path to the end consumer than what an existing ISP might. I’ve always felt that much of the innovation around broadband has occurred despite the ISP or even by bypassing the ISP, so imagine what projects we might see if the pipe owner were an active contributor to that innovation.

We at GigaOM have said for years that broadband is the platform for innovation, and Google no doubt agrees. The pace of technological innovation in terms of video conferencing, telemedicine and remote education are rapidly surpassing the average American’s connection speed, which ranges from 3 Mbps to 7 Mbps depending on the study. And without the demand for such services, or a cost-effective way to get there, ISPs and entrepreneurs that want to deliver products that require fat pipes are reluctant to invest. Think of it as a chicken-and-egg issue. Google can help change this.

The network can help spur innovation, but could also become a foil to the lobbying efforts from existing ISPs, many of whom are less than up-front about how their network costs are reflected in their prices, and tend to react with hysteria when faced with regulations that will limit their ability to control the bits running over their pipes. We’ve all read stories about how the web will break under the weight of network neutrality, or how the ISPs needs to raise prices or implement tiered pricing plans because some consumers are using too many resources. What we don’t have is the data showing that there’s an economic reason for this other than profiteering in an uncompetitive market.

Google makes almost all of its money from selling ads over the Internet, not from selling the pipe itself. Therefore, if it is truly transparent about its costs and traffic demands, it could provide valuable data to the FCC and the industry that telecommunications companies do not. It wouldn’t exactly make the broadband market competitive, but it could help make the economics of operating such a network more transparent. And that could help regulators determine how competitively priced broadband is.

Om wrote about Google’s interest in controlling its own bandwidth back in 2005 for Business 2.0, laying out an economic rationale for the creation of what he called the GoogleNet:

An even more compelling reason for Google to build its own network is that it could save the company millions of dollars a month. Here’s why: Every time a user performs a search on Google, the data is transmitted over a network owned by an ISP–say, Comcast–which links up with Google’s servers via a wholesaler like AboveNet. When AboveNet bridges that gap between Google and Comcast, Google has to pay as much as $60 per megabit in IP transit fees. As Google adds bandwidth-intensive services, those costs will increase. Big networks owned by the likes of AT&T get around transit fees by striking “peering” arrangements, in which the networks swap traffic and no money is exchanged. By cutting out middlemen like AboveNet, Google could share traffic directly with ISPs to avoid fees.

It didn’t happen five years ago and my hunch is that Google was waiting for a last-mile technology that would last. DSL and cable don’t have the capacity to reach 1 Gbps (although cable can offer up to 200 Mbps), and in 2005, fiber to the home was still an expensive pipe dream. But now fiber to the premise, which has the capacity to meet bandwidth demand for decades, is a reality.

It took Verizon’s (s vz) $23 billion investment in its FiOS network to drive innovations for delivering fiber to the last mile and lowering the costs. Technologies such as bendable fiber and smaller optical network terminals that fit on desks rather than inside entire closets were pioneered for the FiOS effort. Google will surely take advantage of that with its deployment.

The fiber GoogleNet could become an indirect threat to Verizon and other ISPs because it could open the kimono on actual network costs and lead to services that would suck even more bandwidth on ISPs’ existing networks. So it’s ironic that the creation of such a network might have a large debt to Verizon at it makes its own fiber push. Personally, I can’t wait.

Related content from GigaOM Pro:

When It Comes to Pain at the Pipe, Upstream Is the New Downstream

79 Responses to “Google's Fiber Network Could Foil ISPs and Fuel Innovation”

  1. Google’s fiber network delivery 1 gig broadband Internet to homes in the select community or communities is a major deal. Longview, TX was one of the 600 plus cities trying to woo Google for fiber optics. Best of luck to all participating communities!

  2. Sorry Stacie but the Incumbent phone companies, CLEC’s and cable companies are not invited to this party nor will they be. Google will need a strong local private Internet Service Provider (ISP) to roll this out to cover all sectors of health, education, city, business and residential to roll this out appropriately. Don’t know of any participating cities that have this necessary infrastructure but Rancho Googlemonga (Rancho Cucamonga, CA) that will be the first host city.

  3. Longview, TX isn’t interested in Cisco fiber. Instead the East Texas city is vying for Google fiber. Longview is hoping to form a long, prosperous relationship with the giant and welcome them to the lone star state with open arms.

  4. This doesn’t pass the sniff test.

    It’s one thing to manage thousands of servers. But networks are a totally different animal. They’re not contained in nice, comfy air-conditioned buildings. Networks are out amongst the rain, sleet and earthquakes. And if that doesn’t cause a problem or 2, there’s always a backhoe waiting to dig and bust your cables.

    And just how does Google plan to do this? They can’t afford to dig up streets and you can bet the incumbent players won’t let them into their conduit. This means they’ll have to lease dark fiber from someone else. But that only gets them to within maybe a mile or so of the customer? How do they actually traverse the last 100 yards? Are they really prepared for the cost of trucks and installers?

    This might work – maybe – in a dense metro area where you can serve a lot of people in a high-rise environment. But if you live on Maple Street, I wouldn’t be holding my breath.

    I will say this – if their puffery manages to get the current crop of ISPs and Telecoms providers to up their game then I’m all for it. Because one thing about the Google announcement is true – bandwidth is a national resource and a competitive edge. So to that end, I tip my hat to them sparking the debate.

    • There are dozens of outside plant contractors that will bill Google for the task of trench and cover, other CLECs and regional incumbents that will hand over pole and duct space if Goog can get the PUC franchises.

      The regional net is trivial, the local POP is trivial, the residential pull is the beast, but not beyond doing. Only takes $$$.

  5. This is an interesting move by Google. I personally thought the Wifi network in Mountain View was a trial as an ISP.

    This appears to be a another trial with a larger sample and different infrastructure.

    You have know that eventually, depending upon what they learn from these trials, they will own an ISP network on some level.

  6. More ISP competition is exactly what the industry needs, especially if it’s transparent competition that offers crazy-fast connectivity. ISPs need a real kick in the pants. I’m tired of only having one real option that’s limited to 10mbit peak down.

    • Brett Glass

      If you think that more ISP competition is needed, you’d better oppose Google’s scheme with all your might. Google’s goal is to create municipal monopoly carriers, eliminating all choice.

      • Electricity is a monopoly with exactly zero competition. I can’t remember the last time my lights went out, or the last time I had trouble pulling enough juice to run my air conditioner. When you ISPs and last-mile providers can match that level and quality of service only THEN can you start harping about change being destructive. The way I see it – as a consumer – is Google might finally be able to provide what all of you ISPs and last-mile providers have been unable or unwilling to provide. You’ve had long enough, now it’s time to move on with a real plan for innovation and progress. Hail Google, our broadband white knight, here to save us from decades of fierce oppression and abuse from our broadband overlords.

  7. For years the carriers have been taking people for a ride – just look at what text messages cost users and the 6,500% markup. Free market or totally controlled ripoff market. Google go and help generate a new era of innovation and low costs.


  8. “DSL and cable don’t have the capacity to reach 1 Gbps…”

    Not really true; cable can go to 2.4 Gbs with current technology, and DSL speed is limited mainly by distance.

    Anyhow, gigagbit networking is already available in Japan, Korea, and Sweden, so the enterprising reporter may want to take a look at what goes on in those testbeds. Japan had to impose a usage cap on the gigabit accounts, for example.

  9. Reading the article on Google’s intent to build their own broadband infrastructure had some interesting scenarios run through me. I first relate this to real estate, build your own warehouse, factory, office and own it instead of leasing it from someone else which then becomes part of a valauble asset. Secondly, Google wants to jump start the innovation in infrastructure development which will provide it with the express way it needs to deliver more high end boradband serives so they will no longer be held hostage to what it can and cannot deliver due to infrastructure constraints. This should and will shake up the fibre optics industry. Also reminds me of when the rail road was build, it connected the East to the West and made the cost of goods come down, a revolution which united a country, perhaps this will be another revolution which will connect small communities to the wired world.
    I love it.

  10. More competition is good. More competition for consumers against regional monopolies like VZ, Comcast, AT&T is even better.
    Even if nothing comes out of this experiment, it would be wonderful if the broadband ISP market is shaken by the entry of Google.

    Typical of Google, their entry is at a large scale, with a very large / differentiating product. I hope their prices will be enticing enough to shake the regional monopolies / duopolies.

      • judging from your other comments, you seem quite frightened by this. why so worried, isp guy? afraid of the big bad google that’s going to help innovate yet another industry full of stagnation?

      • Brett Glass

        Google is already doing great harm to broadband users and the hard working providers who work slavishly day and night to serve them. Its lobbying in Washington, for example, threatens to impose innovation-killing, stifling innovation that will lower the quality of broadband service, raise its cost, and hinder deployment.

        ISPs like myself are the innovators, not Google — which has bought virtually all of the technology it uses. We are not “afraid” of Google. However, we’re well aware of the damage it has done and is trying to do — aided and abetted by credulous writers like Stacey.

        We’ll stop it, mark my words.

      • Your just mad because this will turn you into “dialup” meaning cable dsl will become the AOL of the world. We should all still have AOL service and not Cable, Cable/DSL service ran poor dialup out of town.

  11. Brett Glass

    Stacey, any product for which the provider charges less than its costs is anticompetitive. Google proposes to use taxpayer dollars to undermine hard working commercial ISPs. And you’re advocating that it do so. We’ll fight that to the death, because it IS a life and death matter for our livelihoods and our companies.

    And Google does have a monopoly: at least 80% market share in Internet advertising. And it has an equal lock on users. You might have some idea of how to change the default search engine in your browser, but 99.9% of Internet users do not — and almost as large a percentage is saddled with Google’s spyware toolbar as well. And more than 90% of the Net’s video is streamed by YouTube.

    As for changing ISPs: People in our area can change ISPs with one telephone call or e-mail to us. We do the rest.

    • Your not serious are you Brett? The Windows operating system is on the majority of the worlds computers, which has a default browser known as IE , with a default search engine, which aint google. For google to have such a large share of the search engine market share, people must be figuring out how to choose search engines just fine.

      Why fight to the death? Instead, innovate and be creative, you will make a living just fine that way. Seems to be working for Google. Resting on past accomplishments is prehistoric.

    • NotSureWhatYouAreSaying

      the so-called 80% share is not because it has some sort of a hold on advertisers that they cant switch (as in MS having a monopoly). they have a high share (high 60’s in the search market, what 80% are you referring to?) becuase they are most effective in connecting the advertiser to the consumer of the ads. if they are not, ad owners will shift – there is no monopolistic behaviour tying them back.

    • If the overlords of the last mile hadn’t been sitting on their cans for the last 2 decades they wouldn’t find themselves on the verge of being outclassed by Google. The telecoms took $2 billion of our taxpayer funds in 1996 with the promise of Gigibit FTTH installations, and what did we get for that investment? – DSL. DS-freaking-L, and that’s only if you live close to a CO with a functional DSLAM. I for one can’t wait for all of the current last-mile operators, their ancient infrastructure and even more ancient business practices to get swiftly and solidly kicked to the curb. Good riddance to you and your delusional views – you’ve been about as fun to deal with as an abusive spouse.

    • Xbox and PS3 both charge less then it costs to produce units of gaming systems because of the free market of choice.

      “99.9 percent of users do not now how to change the default search engine in the browser?” Thats got to be a fact, lol. I luckily fall in the .01 percent category, I use the address bar and type or because bing sometimes has cool pictures. This computer happens to have live search with one click I can change the search provider.

  12. Brett, what anti-competitive practices are you speaking of exactly?

    Are you sure you aren’t confused between “Google” and “Regional ISPs?” This is how you can remember: Google doesn’t have any monopolies–Yes, even in search. I could change my search engine in literally 2 clicks of my mouse. Try that with an ISP.

  13. Google world domination is piecing together one piece at a time. There’s Google Public DNS, IPv6 just got supported by Google owned YouTube, Android mobile platform spreads like crazy, now here comes 1Gbps fiber. What’s still missing though is 4G. Be it WiMax or LTE I bet it’ll mimic the FON business model. Bye bye ISPs, we don’t need ya any more.

    • IPv6 is not a choice, its a reality not enough IPs. Google has no effect on whether IPv6 is implemented or not it was accepted long ago IETF around 1996.

  14. i better idea would be for google to invest the same amount of money to build out a large as possible of a free WiFi network across the nation. this could reach millions instead of thousands of people.

    this would also force a look at the pricing of what will dominate future internet access: wireless in various forms.

    they could build QOS controls into the WiFi network that support voice calling from android handsets without needing cellular service. this together with free google voice service would create massive handset sales in the areas covered.

  15. Chris Roczinski

    Sounds promising, but not sure if Google cherry picking one location – weather? – will necessarily reveal any relevant generalized insights into the economics of broadband. Will have about as little relevance as South Korean broadband rates/speeds…course doesn’t stop people from using them. sigh

    • Indu S Das

      don’t’ know what competition u talking about ? In our area (ie a suburb Raleigh, NC) the max upload we get is 512kbps, while the max download is 18mbps. The TWC & AT&T still live in the 1990s restricting end users better upload speed thinking folks will start streaming or hosting.

      This is all because they don’t’ have competition. If my online backup takes days, so be it. Where I am going to go.