back to article Google breaks heart, White Knight falls off horse

Google Fiber was supposed to be the White Knight of US broadband, riding to its rescue. But this horse has bolted. In the opposite direction. Rumours of the scale back had already been reported - we just didn’t know by how much. Now it emerges that Alphabet CEO Larry Page has not only slashed Google’s FTTH operation in half, …

  1. Anonymous Coward
    Anonymous Coward

    Google thought building infrastructure would help them capture more surveillance data, but turned out not to be worth it.

    This is the same reason why all those literal pie in the sky "laser/drone/blimp liberate the 3rd world with wimax" projects will never amount to anything.

    1. DougS Silver badge

      I think the real problem was that they assumed they wouldn't have to face any real competition, but the cable TV industry was able to keep upping their speeds without digging up all the streets.

      Running fiber to the home is silly and pointless. First of all, almost no one really needs or can make use of anything like a gigabit, no matter how much streaming or web surfing they do. Second, it's now possible to provide gigabit speeds by running fiber to neighborhoods and using existing coax or phone lines that already enter people's homes. The problem, Google doesn't own those lines, so they can't use them. That's why they are trying to rethink their strategy and skip over the "last mile" using wireless. The problem is, they won't be the only ones doing that either, AT&T is making big investments in fixed wireless broadband, though they seem to be planning on more rural areas that are currently unserved by cable/DSL so they can offer bundling deals with Directv.

      The cable company has already run the fiber in any market Google would ever consider, so when they hear Google is coming to town they move it to the front of the line for DOCSIS 3 upgrades and are offering gigabit speeds long before Google has their fiber trunks in the ground, let alone start digging up people's yards to run it to their house.

  2. Charles 9 Silver badge

    And the unspoken catch from the article is that Americans in general aren't willing to pay what market would need to support really good broadband. For many, they don't really need it yet, and furthermore the large geographic area of the United States aggravates the capex costs. It's no surprise that, when it comes to average Internet access, the strongest countries are also among the smallest and densest.

    1. Anonymous Coward
      Anonymous Coward

      "the strongest countries are also among the smallest and densest."

      Speaking as someone who lives on a 21 mile long island in the middle of the North Atlantic, we have one of the most expensive, yet slowest and unreliable network you have ever seen.

      I pay $405 a month for a 25MB download / 5MB upload. My home connection (and cable TV) has been down for four days now and they are not coming until Monday (between 9 and 1).

      When it does work, performance is best termed as "optimistic."

      1. 404 Silver badge

        I feel slightly better....

        Paying $90 a month for 15GB data over a satellite in a state established in 1792... but not by much.

      2. heyrick Silver badge

        Who gets the better value?

        I pay €70/month for a mobile/VoIP phone/internet package which gives me 2MB download (quoted as "up to 20MB") and about 0.7MB upload. This is rural France and since I'm at the end of ~4.7km of wire, this will never improve. Sure, there are political plans to bring "everybody" 10MB, 100MB, and a bunch of random values in between, but the basic bottom line fact of the matter is that "everybody" does not involve digging up a kilometre of road to lay fibre for just one person.

        The Internet connection itself is fairly reliable, my connectivity failures are mostly due to the electricity throwing in the towel. So I have never had the chance to see how Orange deal with faults. Thankfully. I pay a lot to use orange because they own the wires. My social circles are full of stories about "orange provides the line and SFR/Free/etc provides the service and they blame each other for months when something goes wrong". Given my location and the large amount of ancient overhead cabling, I'd prefer to pay more so my provider has nobody else to blame.

        Of course, it's our own damn fault for not living in the middle of a city, right? :)

    2. Dan 55 Silver badge

      The unspoken catch is that a) US telecos are regional monopolies which charge a bomb but won't invest and b) Google got bored of Google Fiber because a butterfly fluttered past the window, as with all their projects. When Google started this they must of known this was a long-term project which requires investment upfront, even if it was just going to be rolled out for a handful of cities. Now the customers who went for this are stuck... again. Obligatory xkcd.

  3. Anonymous Coward
    Anonymous Coward

    "the large geographic area of the United States aggravates the capex costs."

    The majority of the population is concentrated in urban areas. The *gaps* between those areas are immaterial, since the backbone networks connecting them already exist.

    The table at shows 497 urban areas with a population of at least 50,000; the total is 223,302,100.

    From the same 2010 census, the total US population is 308,745,538:

    So you can reach 72% of the population by connecting those 497 urban areas. (You'll get 55% of the population by connecting just the first 100 in that list)

    1. Charles 9 Silver badge

      Re: "the large geographic area of the United States aggravates the capex costs."

      "...since the backbone networks connecting them already exist."

      They exist, but I don't think they're at the terabit level just yet, which is what you'll need going forward (since it's the cities where all the broadband demand is). And that means new capex. Ever gotten the estimate on running that much fiber from New York to Los Angeles (or even further, Miami to Seattle)?

      1. Mikel

        Re: "the large geographic area of the United States aggravates the capex costs."

        > They exist, but I don't think they're at the terabit level just yet, which is what you'll need going forward (since it's the cities where all the broadband demand is). And that means new capex.

        No. It happens that this is all single mode fiber of the sort that can now carry a terabit per strand. That is why they develop the optics for that fiber type. Back in the day it was laid the electronics were barely capable of a gigabit, but that was then - and they were looking to gross $2/MB (65336[64k bps]*60/8)/$ for long distance voice calls. That was the economics that made it worthwhile to bury all those thick bundles of fiber, and lend money to someone who was doing so. But the pace of progress in optics overtook the speed of cost recovery in long distance telephone service (mostly landline!) and the competition became who could drive whom out of business fastest. Suddenly lending the money became too big a risk and it all collapsed. But not before they buried the fiber, which was sold on the auction block in the bankruptcies at millicents on the dollar of the cost to buy and bury it.

        But it -is- the terabit fiber, and the big cost isn't the electronics. It's getting the rights of way and burying the passive glass.

        1. Charles 9 Silver badge

          Re: "the large geographic area of the United States aggravates the capex costs."

          "But it -is- the terabit fiber, and the big cost isn't the electronics. It's getting the rights of way and burying the passive glass."

          That was exactly what I was talking about: laying down the actual cables so they can actually be used. If it's above ground, it's as good as dirt because the real cost of data infrastructure is the physical part of it: burying and maintaining it, and the US isn't wholly up to speed yet because of weak links in the chains: mostly in rural areas who can't be asked to do it OR give the permission for someone else to do it. And sometimes the feds get involved because the cables have to necessarily cross environmentally-sensitive ground.

    2. WolfFan Silver badge

      Re: "the large geographic area of the United States aggravates the capex costs."

      methink some of data wrong.

      Example: according to the list of urban areas noted above, Miami, Florida shows as:

      4 Miami, FL 5,502,379 3,208.0 1,238.6 1,715.2 4,442.4

      This means that Miami is the #4 urban area in the United States, New York, Los Angeles, and Chicago. Hmm. This shows a population of 5.5 million. The problem is, however, that according to, the actual United States census people, the population of the WHOLE OF MIAMI-DADE COUNTY is 2.7 million, or half that. Okay. Maybe the 'Miami urban area' includes not just Miami-Dade County, but Broward County as well, even though one would think that that's the 'Fort Lauderdale urban area'. Oh. Wait. According to, the population of the whole of Broward is... 1.9 million. Miami-Dade and Broward combined don't hit 5.5 million. Palm Beach County, north of Broward, is another 1.6 million. We'd need half of the population of Palm Beach County added to the complete populations of Broward and Miami-Dade counties to get to 5.5 million. I'm pretty sure that residents of Boca Raton would be surprised to learn that they're living in Miami. As would residents of Boynton Beach, West Palm Beach, Loxahatchee, Belle Glade, Davie, Weston, Tamarac, Coral Springs, Coral Gables, Kendal, Doral, Homestead...

      me no trust figures in Me think you should not trust them either.

    3. Badger Murphy

      Re: "the large geographic area of the United States aggravates the capex costs."

      While that's true, the federal government, as I understand it, has limited ability to work that way. When they make laws and appropriate money, it is difficult to fashion a law such that only works for "most people" strictly based on where they choose to live. If the US did as you propose, you can bet that every municipality that is not part of this connectivity rollout will gripe to high heaven and tie up the courts with litigation, incurring additional costs.

    4. MondoMan

      Re: "The majority of the population is concentrated in urban areas."

      Ahhh, AC, I feel for you -- you've been taken in by the Wiki again!

      A careful reader would discern that the so-called "urban areas" include large percentages of what civilians would call suburban or rural areas. As even the Wiki notes, the "urban area" is composed of core census tracts/blocks " *along with* adjacent territory containing non-residential urban land uses as well as territory with low population density included to link outlying densely settled territory..."

      Thus, for example, the Seattle "urban area" includes the entire circum-Puget Sound area from Port Ludlow down to Olympia and back up past Marysville, an area reaching ten to twenty miles inland all around the Sound.

      "Connecting" at least this "urban area" is much more difficult than reaching the three times as many inhabitants in Seoul.

      1. Tomato42 Silver badge

        Re: "The majority of the population is concentrated in urban areas."

        the point is that those suburban areas have higher population density than Europe, and they still have slower and vastly more expensive Internet

  4. Anonymous Coward
    Anonymous Coward

    Page is right

    Google first generation access network is just a telco GPON as-is. That is expensive as it carries lots and lots of legacy. For example, an optical access network should not need a the whole BNG + supporting infra. That is an idiocy from the days of wholesale dialup grafted onto a network that does not need it. Google should have reduced that cost already instead of verbatim copying telco architectures and using off-the-shelf kit.

    While you cannot reduce 90% this way, you can and should reduce equipment costs quite a bit which in turn should make the network competitive against the incumbents.

    You can also reduce rollout costs vs standard telco costs by approximately x3 through improved organization (I have done that modeling in a past life in a telco by the way). However, if you try that in a telco the unions will fit you some nice and comfy concrete shoes. As Google is not yet unionized, it can actually try this one.

    When you combine the technical cost reduction with rollout organization cost reduction you actually can get > 70% overall cost reduction. In order to achieve 90% you will need to do industrial scale rollouts in suburbia style areas (not touching the countryside and not touching metropolitan areas). It is probably doable (again - been there, done that math, ended up with a conviction into the telco research organization for showing that the current costs can be reduced).

  5. frank 3

    What an odd right-wing rant this 'article' is.

    US broadband is known to be dismal and most of blame for that lies in monopolistic behaviour from the telco's. In the cases of a clear market failure, it is the duty of the state to intervene to ensure that competition can flourish. And yet, this ranty article seems to completely ignore that. How odd.

    1. Charles 9 Silver badge

      Beacuse in the case of the United States, geography gets in the way, and you can't easily solve geography. If it's 100 miles to the nearest trunk line, someone's gonna have to pay for making the connection, end of. That cost (both installation and maintencnace) will have to be recovered. It's these unavoidable costs that have some capitalists asking why the Post Office doesn't charge higher postage rates to remote Alaska or Hawaii.

    2. Youngone Silver badge

      @ frank 3

      That's exactly what I thought. It's well known that the monopolistic ISP's in the US spend a fortune writing laws to prevent cities putting their own cables in, and yet somehow "regulation" is happening now.

    3. Throatwarbler Mangrove Silver badge

      Because Orlowski vs. Google.

    4. theblackhand

      Re:the US broadband monopoly

      The US broadband monopoly is caused by counties/states handing monopolies to the telcos - no one then has any interest in intervening to fix the problems.

      While Googles fibre projects were interesting, the costs were significant - the estimate of US$1b/city and low take up rates outside of richer neighbourhoods suggested it was never going to make rapid in-roads into the US telecoms market, but I thought they may give it more time.

      For the 90% funding cut, i guess that puts expansion on hold until current subscriber numbers increase to meet current costs.

    5. tom dial Silver badge

      It ain't what you don't know that gets you into trouble, but what you know that ain't so. (Attributed, probably wrongly, to Samuel Clemens, AKA Mark Twain).

      For various reasons, US internet service may be poor in places, but arguably does not rate that badly. US average connect speed (12.6 Mbit/sec) ranks about 14, roughly midway between Germany and Denmark; peak speed (,57.3 Mbit/sec) ranks about 16; and for connections better than 15 Mbits/second (24%) the rank is about 15 (full confession: based on Akamai reports for the third quarter of 2015 as reported by Wikipedia). That puts the US generally in the same group as much of Europe + Canada, and well ahead (for example) of Australia, New Zealand, and France. It could be better, and in many places is, but certainly does not qualify as "dismal" (or even abysmal) as it is.

      The state of internet service in the US also is not a result of market failure. In many areas, possibly most, it results from lack of or severe restriction of the potential market, as Google's entry into several areas has shown by inducing existing providers like Comcast and at&t to improve service and reduce prices in some combination.

      My own experience, first with Cox and lately with Comcast, has been of increasing speed over time (measured courtesy of ookla) at nearly constant price over a period of about ten years. The outlier was at&t, which failed to match Cox (as promised) albeit at a lower price. This, of course, is to be expected as a result of normal equipment replacement where the new equipment is inherently better than what is replaced due to general technology improvement.

      1. nijam

        > US average connect speed (12.6 Mbit/sec) ranks about 14

        You can achieve high average connect speeds by refusing to provide connections that wouldn't have a decent speed. Which you can do if you're a monopoly.

    6. ecarlseen

      That monopolistic behavior is legislated...

      For the majority of American customers (about 2/3 IIRC), There Can Be Only Two Providers (Telco and Cable) *by law*. Most local governments sell these "rights" to megacorporations as franchises with a guarantee of no competition, which is one of the more prominent reasons said megacorporations often go ballistic at the threat of community broadband or Google Fiber or whatever ("but you promised!!!"). Back in the day the megacorporations had to provide universal coverage no matter how stupidly expensive it was to run a pair of copper wires out to some guy living in a shack fifty kilometers from the nearest road or whatever, but in the days of cellular coverage broadband these agreements are often so out-of-date as to be meaningless.

      Anyway, no market failure because there is no market - it's a command-and-control economy mandated and run by heavily-lobbied politicians and bureaucrats. Consumers such as myself get to choose the Coke or Pepsi product, no matter how much we'd rather have a nice herbal tea.

      1. Charles 9 Silver badge

        Re: That monopolistic behavior is legislated...

        "Anyway, no market failure because there is no market - it's a command-and-control economy mandated and run by heavily-lobbied politicians and bureaucrats. Consumers such as myself get to choose the Coke or Pepsi product, no matter how much we'd rather have a nice herbal tea."

        But without Coke or Pepsi, there would be no water service to your area, making a tea option a non-starter. The main reason for the exclusivity agreements was because the cable and phone companies refused to run out to those areas at all without them. So it becomes a choice of 10% of something or 100% of nothing at a time when telephone and/or cable access was a tax matter (because having those services in town affected people's decisions to move in).

  6. Brian Miller

    Slow broadband - so what?

    Screw this garbage! I started with "broadband" when modems had acoustic couplers and uucp was the norm. Yes, I have used Teletype machines. My first modem ran at a blazing 300 baud. Some of the others around here also have a bit of perspective on this.

    High-speed data links are not available everywhere. They can't be. If someone has a need for speed, they have to move where the pipes are available. That's really all there is to it. Yes, there is unused capacity in the ground. No, it doesn't belong to you. Somebody else owns that, and if you want the use of someone else's property, you have to pay for it.

    There's really not that much information out there. Most of the slow data is stupid ads and tracking servers. Yes, it sucks that not every place has good service, but you can speed up your own connection by using JavaScript and ad blockers. If you must have the highest speeds available, you have no choice but to move. If the neighborhood where you're at doesn't have what you want, you have to move.

    Complaining about slow Internet, or the Internet divide, is just useless whining. So Google has wised up about one of its projects. Whoopee.

    1. Anonymous Coward
      Anonymous Coward

      Re: Slow broadband - so what?

      But what if you NEED that high speed (for example, remote management of videoconferencing) AND can't afford to move?

  7. NotBob

    Over here,

    The broadband you have is seldom the broadband you want...

  8. David Roberts Silver badge


    Land line spend of 30 bn a year sounds much the same as 7 to 9 bn per quarter for mobile. That seems to be about 28 to 36 bn a year?

  9. Mikel

    I don't believe this report

    Page is not this shortsighted. The report is not credible.

    It might be that he asked to see a quicker ROI. But that's it. This is a big investment long-term revenue stream, and a good performing investment.

    1. Charles 9 Silver badge

      Re: I don't believe this report

      "It might be that he asked to see a quicker ROI. But that's it. This is a big investment long-term revenue stream, and a good performing investment."

      Have you got the numbers to back that up? Last I read, the numbers really weren't backing up Google Fiber. It's like trying to sell Rolls Royces in communities where the vast majority of your customers are lucky to own Mini Metros.

  10. Mark 85 Silver badge
    Thumb Down

    You don’t get the broadband you deserve - you get the broadband you pay for.

    Bullshit. We pay for "advertised speeds" and get something way less. We pay for service 24/7 and get rolling outages and denial by the companies that anything is wrong with their system and it must be our computers. We have companies playing the game the law allows by defining "high speed" as whatever they freakin' want.

    So no, we pay for what they will give us along with paying for the huge profits and the salaries/bonuses of the C-Suite types. I'll add that for most of us, there is no choice of ISP or the choice is copper cable or DSL.

    1. Brian Miller

      Bad service? Sue them!

      If the service you receive isn't what's advertised, sue them! Moping over it won't fix a thing. Do what it takes to grab them by the profits, and either get what the terms said you'd get.

      It just doesn't make sense to let all those pitchforks and torches go to waste.

      1. Anonymous Coward
        Anonymous Coward

        Re: Bad service? Sue them!

        "Sue them!"

        Um, many have tried...and lost! What good is suing when rich media companies can basically bribe the system to their liking?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019