7 ms·

Economics of Google Fiber and what it means for U.S. broadband

163 points by iProject 14y ago

gamble 14y ago
Isn't the more obvious conclusion that Google Fiber is just a bargaining chip for Google to use against traditional ISPs and cell phone networks? It won't even serve all of Kansas City for the foreseeable future, much less the rest of the country. What it does do is demonstrate that Google has the will to play the carriers' game, without actually committing the capital necessary to build out a competitive network. Remember how gung-ho Google was about network neutrality, before it became necessary to choose between net neutrality and Verizon's support with Android. I find it hard to believe that Google's shareholders would appreciate the company devoting the massive amounts of capital necessary to build out a fiber network across the US, unless it was an existential question. And even then it will only be built out to the degree necessary to punish the carriers.
- throwaway64 14y ago
  Google already has a massive nationwide network to serve as a CDN for stuff like youtube. They are peered with every ISP of any significant size, and have data centers and points of presence in every major city. Not to mention they have bought up almost all the "dark fiber" they could get their hands on for the last 10 years. Their network already exists, and its everywhere, the only thing remaining is rolling it out to residential areas. There is a large upfront free that customers have to pay for the installation. If the economics have been worked out correctly by Google, the customers themselves will pay almost all of the roll out costs. Edit: I should clarify, a large upfront fee, or a reasonably large monthly fee.
  - gamble 14y ago
    What fee? The only requirements are a 2-year contract, or a $300 upfront fee for the free option. And they give you a free Nexus 7 tablet with the contract. At $70-120/month, the money they're collecting on a 2-year contract isn't remotely enough to cover construction of the network and the end-user electronics.
    - throwaway64 14y ago
      $120 a month for 2 years is $2880 guaranteed, not to mention it is unlikely customers will leave considering the competition comes nowhere even close to that kind of offer. Some of that will be eaten up by fixed costs of providing the service (for stuff like bandwidth, Google can probably get better prices than almost anyone else in the world, due to extensive peering), and for the provided electronics (which Google can order in huge quantities and pay comparatively little for, or even build themselves, like they do with their own internal servers and networking equipment). Whats left over is still a very large chunk of change, especially when you consider the fact that Google wont roll out fiber till a decent percentage of a neighborhood is willing to sign on (their calculated numbers are almost certainly designed to cover roll out costs) I think this will easily pay for their roll out costs. Remember, they don't need to build out a national network to serve tons of places, they already have one. YouTube itself already uses something like 40% of the internet traffic to residential connections. The important thing to consider here, is as far as Google is concerned, they only need to break even, they arn't in the same game the other ISPs are where it is their only profit center.
    - dedward 14y ago
      One would imagine they were thinking long term and big - they didn't start laying the foundations for this 10 years ago for no reason.... and while they may have toyed with some "lab" projects that never took off... this one represents a hell of an investment - they're probably several moves ahead of everyone else already. Now imagine what they have up their sleeve that they aren't telling anyone about.
- wmf 14y ago
  I think it's been demonstrated that the "bargaining chip" theory does not work very well; incumbent ISPs only increase service/reduce prices in cities where a competitor appears. But by the time Google Fiber is in a city, Google has already spent a lot of money. Maybe Google could perform some kind of franchise judo where they have the smallest possible buildout in each city, forcing incumbents to respond disproportionately...
nodrama 14y ago
it's funny that they speak about "making a profit". We have the same price per Mbps in my country as Google's $70/month offer. And we have this since at least 4 years ago. The only valid point is about building the network, because the cost of that differs from place to place. This is offset with the overselling: who uses 1Gps at home? I have 100Mps and most of the time I don't need it, or, when I need it, the server it's too slow anyway.
jmillikin 14y ago
One aspect I've not seen discussed is the impact of a commercial fiber service on city- and county-owned networks. In the past, when local governments have tried to roll out high-speed municipal internet access, they've faced bitter opposition from entrenched network providers (Comcast, Time Warner, AT&T, etc). These companies refuse to roll out fiber, but also lobby for state laws against government-funded fiber[1]. The end result is that residents are left on decrepit old copper lines. Now there's a large[2] company rolling out fiber, and they have shown explicit desire to work with local governments. Depending on how the laws are written, it might be possible to achieve something close to a municipal network by contracting out to Google. And the Google fiber seems to be cheaper -- I can't find any municipal fiber services offering gigabit for less than $120. (bias disclaimer: I work for Google, but am not involved at all in fiber stuff) [1] http://arstechnica.com/tech-policy/2011/03/cable-backed-anti-muni-broadband-bill-advances-in-north-carolina/ http://arstechnica.com/tech-policy/2011/03/cable-backed-anti... [2] Sonic.net deploys fiber, also priced at $70 a month, but they are a very small regional ISP. http://arstechnica.com/tech-policy/2011/06/1gbps-fiber-for-70in-america-yup/ http://arstechnica.com/tech-policy/2011/06/1gbps-fiber-for-7...
- pasbesoin 14y ago
  As I commented elsewhere, perhaps Google is the party with the deep pockets and legal expertise -- and lobbyists -- to take on and beat these entrenched monopolies/oligopolies. When a community of some thousands goes up against a corporation of 10's of thousands and a largely captive customer base of millions, in e.g. lobbying the state legislature, well, the deck seems kind of stacked. In Illinois, I recall specifically, SBC -- now AT&T -- received hundreds of millions of dollars in tax breaks and incentives, specifically with the agreement to promote and complete universal high speed access. They then promptly set their lawyers and lobbyists to finding ways to break their side of that agreement -- while keeping the incentives, of course. If Google wants unfettered access to unfettered customers (or eyeballs, if we're going to call the advertisers the customers... not to go there, at the moment), such partnerships might be a good way to identify and speed deployment. And... plenty of publicity about what people in many other countries are paying. Little seems to motivate the American consumer -- and voter -- more than a bargain, or feeling they're getting ripped off. And, since I'm typing away here, I hope Google will do it while treating their linepersons (technicians, etc.) decently. As a recent story here, with its high rating, pointed out, people are tired of seeing the line workers being abused. I'm not familiar with sonic.net in detail, but its CEO, Dane, seems a decent sort, and I bet they're not the worst employer -- whether direct or contract. (And since Google's already talking to him, maybe you all can pick his brains in this specific area.) Google still does some things that piss me off. [1] But I support this initiative -- I'll sign up as soon as I can, if and when -- unless Dane beats you to it. [1] Mentioning just one item: Give it up, already, with the "true names" initiative. Anonymity has its value and place within the online domain. Not just one of convenience, but one of necessity, for many people and contexts. P.S. I'm pretty tired, and I'm probably being to optimistic/idealistic, here. OTOH, having directly observed the crap dished out by AT&T, Comcast, and the like, I can readily believe there is plenty of room for improvement.
obituary_latte 14y ago
As a layman, newb, wannabe hacker etc. this does nothing but excite me. I long ago conceded the privacy vs convenience point regarding google. I've been a long time customer. I felt (read: convinced myself) then as I do now that the engineering backbone would stifle less than acceptable behavior by the company. I live in the largest city of a northeastern us state. My options for acceptable Internet are nil. +$120/month for (possibly) 70 Mbps. Which, from experience, is heavily weighted on concurrent usage. Everything else aside, I want so bad the option to have [insert company (google is winning the race)] to come in an offer me something better. If nothing else, as has been stated in other threads, this will be an enormous boon for technological state of ISP's in the so far backward US.
- mhurron 14y ago
  > My options for acceptable Internet are nil And it will remain that way for a long time. Google doesn't serve the whole metropolitan areas they are in and there is no known plan to expand beyond the two areas they are in. Google has stated they do not want to be an ISP so no plans to expand may not just be them being quiet. So what does this mean for existing large ISP's? Nothing. The same nothing that every little regional ISP means to them.
jlarocco 14y ago
I'm not a big fan of my current provider, but I think I prefer them to the privacy concerns I'd have using Google. Almost every existing Google service is based on snooping people's browsing and serving up ads. They've made billions and billions of dollars doing it, and as the ISP they'd be able to snoop on everything. I'm sure they promise not to, but I just don't trust them, and I think there's too much money to be made for them not to do it.
- ineedtosleep 14y ago
  I don't buy this. Google would be offering you an internet service here. If you choose not to use their web services, great; you are free to not use them. If you're afraid Google are going to mine your traffic, use ssl, ssh, etc.. I don't see how this is any different with most of the ISPs monopolizing the market now.
- avree 14y ago
  You trust your current ISP? I've never understood these arguments. Google doesn't 'snoop' on your browsing—they anonymously track your habits to give you better ads as their business model. You used to get terrible ads on the internet—flashy popups, tricky spam ads, etc. Google has changed this, since they realized that they could avoid needing to trick people into click on ads by delivering ads that are actually relevant. Your current ISP can 'snoop' on everything you do too. Why don't you have this odd mistrust of them?
  - thaumaturgy 14y ago
    Your own comment explains why: Google has built an empire upon the ability to efficiently compile information about people. I've worked for a small ISP. They are not nearly as competent in that regard. Most of the time, they're just hoping they can finish the day with fewer tech support tickets than they started the day with. I trust Comcast / SBC / others less -- especially since they voluntarily opted to cooperate with the RIAA/MPAA against their users' interests (http://www.dslreports.com/shownews/ISP-Six-Strikes-Plan-Arrives-in-July-118834 http://www.dslreports.com/shownews/ISP-Six-Strikes-Plan-Arri...) ... but I trust Google least of all, speaking just from a privacy standpoint. I realize that it's a little irrational, especially since, as you pointed out, the information is anonymous. (We think.) However, it's not so much that I'm concerned with what Google is doing with that information now; I'm concerned with what they might do with it later. Once the data has been collected, it's impossible to ever know for certain that it's been removed. Data collected about customers will survive middle management and CEOs and entire company boards. It may be sold at any time to anyone for any reason -- or given away freely. It may be used for purposes which I am not imaginative enough to conjure up. That probably sounds a little paranoid, but then again, when I was dialing in to eWorld in 1995, I could not have imagined that 17 years later the cable company would be my ISP and they'd be snitching on me if I downloaded unauthorized content. 17 years is not beyond the life time of current storage methods; who knows what will come next? Finally, Google can't survive forever on advertising. I frankly expected the online advertising market to be having greater problems by now than it is. Still, AdBlock Plus is popular and I've seen a few articles recently that are tentatively suggesting that buying advertising on Facebook probably isn't worth it. Google's building some actual consumer hardware now, which is super neat, but their core competency is in data mining. I think it would be foolish to believe that if the online advertising market started to really struggle, Google would just drop the thing that they're best at and move on to something else. So I don't think it's such a bad idea to treat Google with a little mistrust. (But their fiber project is really damn cool.)
shawnee_ 14y ago
Kansas City is pretty close to the geographic center of the US. ... to reduce the cost of the actual last mile to users’ homes it’s telling people in Kansas City that if they want to be the first to get fiber, they’ll have to convince their neighbors to sign up. The goal is to get a critical mass of between 5 percent and 25 percent of the homes in a given neighborhood (Google calls it a fiberhood) committed to signing up for Google Fiber before ever sending out technicians. Speaking of cost-per-mile as it relates to "the last mile", population-density surely has to factor into how cost-effectively this service can be unrolled. In a densely-populated area, less fiber can serve more people, and can probably greatly reduce the cost-per-household to install. Seems like the Midwest would require more cost for fewer households able to benefit, at least in the beginning of this endeavor. Wouldn't it make more sense to start on both / either coast, and literally reduce the "cost" of unrolling it marginally to the less-dense areas of the country, proportional to physical distance?
- wmeredith 14y ago
  Kansas City's metro area is as population dense as any other developed urban core. It's not like they're starting out in a wheat field in Kansas.
  - humbledrone 14y ago
    This table seems to disagree: http://www.demographia.com/db-uscity98.htm http://www.demographia.com/db-uscity98.htm . From the totals at the bottom, I found that the average population density for the 600 largest cities in the U.S. in 2000 was (total population)/(total land area) = 99,000,000 / 34,000 = 2911 people per square mile. Kansas City's density is 1100-1400 people per square mile, meaning that not only is it false that it's as "dense as any other developed urban core", but in fact it is not even half as dense as the average city (with population 50,000 or higher).
    - tmrggns 14y ago
      I think part of that is the way the city is laid out. I think Kansas City is more consistent than most cities in having major roads travelling about every five or six blocks. This probably helps when physically putting the fiber down.
fsckin 14y ago
Imagine for a moment, if they fudged the numbers a bit on the storage capacity of every box going into your home. They advertise the usable capacity of 1 or 2TB. In reality, it has 30-200% more storage that is not usable by the customer. They just made a very fast, low latency, distributed mesh CDN that the customer pays Google for the connectivity AND electrical bill.
- throwaway64 14y ago
  they don't really save anything on the connectivity, as everything would still be routed through Google, thus it would use exactly the same amount of bandwidth as if the storage was sitting inside a data center. As for electricity, the cost of packaging up and sending drive units to the customer + upkeep would dwarf the costs of electricity for the total lifetime of the drives. I don't think this really makes sense.
  - dullcrisp 14y ago
    Also regarding the electricity, would that not be considered theft, if nothing else.
    - derleth 14y ago
      I really don't see how.
      - dullcrisp 14y ago
        If they don't disclose that the box they gave you purportedly to record TV shows is actually also drawing power from your line and using it to run part of their datacenter? I would think that that could be considered theft by deception. It's certainly very shady. I'll admit that the line is a bit blurry.
        SquareWheel 14y ago
        If they were to actually do this, they would of course disclose that fact. It would be written in small print, but still disclosed.
whatusername 14y ago
Does anyone in Australia want to take a stab at what this is likely to mean for the politics of the NBN?
- mattmiller 14y ago
  I thought NBN already planned to do fiber to the door for most of the country.
  - whatusername 14y ago
    So one side of politics will claim this proves that FTTP is the correct solution. The other side will claim that Private Investment is the way to build Fiber rollouts.
zmmmmm 14y ago
So a key part of the "cost saving" is that AT&T gave them cheap access to poles. Doesn't that mean they are relying on one of the very competitors they will displace to collaborate with them in this? Seems to me that either AT&T was forced to do this by competition laws OR they indulged this little experiment as a one-off but the chance of them or any other telco letting Google scale this up to the point where it matters is zero. Does anybody know which one it is?
- jws 14y ago
  I believe most of the poles are not AT&T's, but rather owned by the county and it's Board of public utilities. The combined power and communication poles have a standard fee for attachment. Google got a special deal to use the power area of the pole for free as part of their enticement. Labor cost is higher for the power area of the pole. The AT&T deal is probably for just the odd pole where the communications approaches a building from a different path.
amalag 14y ago
Curious why they don't have another mid range offering, 5Mb/s free then 1000Mbs at $70 are the only options? I would love something in the $15-$25 range.
- dsr_ 14y ago
  By making the low tier free, they avoid many costs. Billing. Tech support. Compared to their gigabit-chewing neighbors, the free tier isn't even using bandwidth of any significance. Supporting a middle tier would cost as much as the high tier in everything except bandwidth. The one-time payment of the free tier helps pay for the infrastructure in the neighborhood, but doesn't incur any real costs after that. (I expect customer support for free will consist of a web form to report an outage, and a wiki.)
  - Avenger42 14y ago
    > customer support for free will consist of a web form to report an outage And how do you get to the web form if there's an outage? :)
    - Retric 14y ago
      Cellphone? However, I suspect they could automatically detect outages based the in home battery backed up router not responding or losing contact with all other devices on the network.
- splinterofchaos 14y ago
  I suspect offerings like that will be rolled out within the next year or two, if all goes well.
Jun8 14y ago
"From the infrastructure on the back end to the TV and Wi-Fi routers in the home, Google has built its own stuff. Most carriers rely on outside vendors to sell them networking gear and even set-top boxes." If I were Comcast I would really be afraid at this point, because Google now has access to top-notch set-top-box, as well as other CATV equipment know-how through Motorola Mobility's Home division. If then can incorporate that knowledge into their system they would have an end-to-end entertainment delivery system.
- vyrotek 14y ago
  I really hope they are. Comcast did some lobbying some time ago in Utah which prevented local fiber ISPs from being able to install new lines in many cities. It's very frustrating!

163 points by iProject 14y ago

Isn't the more obvious conclusion that Google Fiber is just a bargaining chip for Google to use against traditional ISPs and cell phone networks? It won't even serve all of Kansas City for the foreseeable future, much less the rest of the country. What it does do is demonstrate that Google has the will to play the carriers' game, without actually committing the capital necessary to build out a competitive network. Remember how gung-ho Google was about network neutrality, before it became necessary to choose between net neutrality and Verizon's support with Android. I find it hard to believe that Google's shareholders would appreciate the company devoting the massive amounts of capital necessary to build out a fiber network across the US, unless it was an existential question. And even then it will only be built out to the degree necessary to punish the carriers.

throwaway64 14y ago
Google already has a massive nationwide network to serve as a CDN for stuff like youtube. They are peered with every ISP of any significant size, and have data centers and points of presence in every major city. Not to mention they have bought up almost all the "dark fiber" they could get their hands on for the last 10 years. Their network already exists, and its everywhere, the only thing remaining is rolling it out to residential areas. There is a large upfront free that customers have to pay for the installation. If the economics have been worked out correctly by Google, the customers themselves will pay almost all of the roll out costs. Edit: I should clarify, a large upfront fee, or a reasonably large monthly fee.
- gamble 14y ago
  What fee? The only requirements are a 2-year contract, or a $300 upfront fee for the free option. And they give you a free Nexus 7 tablet with the contract. At $70-120/month, the money they're collecting on a 2-year contract isn't remotely enough to cover construction of the network and the end-user electronics.
  - throwaway64 14y ago
    $120 a month for 2 years is $2880 guaranteed, not to mention it is unlikely customers will leave considering the competition comes nowhere even close to that kind of offer. Some of that will be eaten up by fixed costs of providing the service (for stuff like bandwidth, Google can probably get better prices than almost anyone else in the world, due to extensive peering), and for the provided electronics (which Google can order in huge quantities and pay comparatively little for, or even build themselves, like they do with their own internal servers and networking equipment). Whats left over is still a very large chunk of change, especially when you consider the fact that Google wont roll out fiber till a decent percentage of a neighborhood is willing to sign on (their calculated numbers are almost certainly designed to cover roll out costs) I think this will easily pay for their roll out costs. Remember, they don't need to build out a national network to serve tons of places, they already have one. YouTube itself already uses something like 40% of the internet traffic to residential connections. The important thing to consider here, is as far as Google is concerned, they only need to break even, they arn't in the same game the other ISPs are where it is their only profit center.
  - dedward 14y ago
    One would imagine they were thinking long term and big - they didn't start laying the foundations for this 10 years ago for no reason.... and while they may have toyed with some "lab" projects that never took off... this one represents a hell of an investment - they're probably several moves ahead of everyone else already. Now imagine what they have up their sleeve that they aren't telling anyone about.
wmf 14y ago
I think it's been demonstrated that the "bargaining chip" theory does not work very well; incumbent ISPs only increase service/reduce prices in cities where a competitor appears. But by the time Google Fiber is in a city, Google has already spent a lot of money. Maybe Google could perform some kind of franchise judo where they have the smallest possible buildout in each city, forcing incumbents to respond disproportionately...

nodrama 14y ago

it's funny that they speak about "making a profit". We have the same price per Mbps in my country as Google's $70/month offer. And we have this since at least 4 years ago. The only valid point is about building the network, because the cost of that differs from place to place. This is offset with the overselling: who uses 1Gps at home? I have 100Mps and most of the time I don't need it, or, when I need it, the server it's too slow anyway.

jmillikin 14y ago

One aspect I've not seen discussed is the impact of a commercial fiber service on city- and county-owned networks. In the past, when local governments have tried to roll out high-speed municipal internet access, they've faced bitter opposition from entrenched network providers (Comcast, Time Warner, AT&T, etc). These companies refuse to roll out fiber, but also lobby for state laws against government-funded fiber[1]. The end result is that residents are left on decrepit old copper lines. Now there's a large[2] company rolling out fiber, and they have shown explicit desire to work with local governments. Depending on how the laws are written, it might be possible to achieve something close to a municipal network by contracting out to Google. And the Google fiber seems to be cheaper -- I can't find any municipal fiber services offering gigabit for less than $120. (bias disclaimer: I work for Google, but am not involved at all in fiber stuff) [1] http://arstechnica.com/tech-policy/2011/03/cable-backed-anti-muni-broadband-bill-advances-in-north-carolina/ http://arstechnica.com/tech-policy/2011/03/cable-backed-anti... [2] Sonic.net deploys fiber, also priced at $70 a month, but they are a very small regional ISP. http://arstechnica.com/tech-policy/2011/06/1gbps-fiber-for-70in-america-yup/ http://arstechnica.com/tech-policy/2011/06/1gbps-fiber-for-7...

pasbesoin 14y ago
As I commented elsewhere, perhaps Google is the party with the deep pockets and legal expertise -- and lobbyists -- to take on and beat these entrenched monopolies/oligopolies. When a community of some thousands goes up against a corporation of 10's of thousands and a largely captive customer base of millions, in e.g. lobbying the state legislature, well, the deck seems kind of stacked. In Illinois, I recall specifically, SBC -- now AT&T -- received hundreds of millions of dollars in tax breaks and incentives, specifically with the agreement to promote and complete universal high speed access. They then promptly set their lawyers and lobbyists to finding ways to break their side of that agreement -- while keeping the incentives, of course. If Google wants unfettered access to unfettered customers (or eyeballs, if we're going to call the advertisers the customers... not to go there, at the moment), such partnerships might be a good way to identify and speed deployment. And... plenty of publicity about what people in many other countries are paying. Little seems to motivate the American consumer -- and voter -- more than a bargain, or feeling they're getting ripped off. And, since I'm typing away here, I hope Google will do it while treating their linepersons (technicians, etc.) decently. As a recent story here, with its high rating, pointed out, people are tired of seeing the line workers being abused. I'm not familiar with sonic.net in detail, but its CEO, Dane, seems a decent sort, and I bet they're not the worst employer -- whether direct or contract. (And since Google's already talking to him, maybe you all can pick his brains in this specific area.) Google still does some things that piss me off. [1] But I support this initiative -- I'll sign up as soon as I can, if and when -- unless Dane beats you to it. [1] Mentioning just one item: Give it up, already, with the "true names" initiative. Anonymity has its value and place within the online domain. Not just one of convenience, but one of necessity, for many people and contexts. P.S. I'm pretty tired, and I'm probably being to optimistic/idealistic, here. OTOH, having directly observed the crap dished out by AT&T, Comcast, and the like, I can readily believe there is plenty of room for improvement.

obituary_latte 14y ago

As a layman, newb, wannabe hacker etc. this does nothing but excite me. I long ago conceded the privacy vs convenience point regarding google. I've been a long time customer. I felt (read: convinced myself) then as I do now that the engineering backbone would stifle less than acceptable behavior by the company. I live in the largest city of a northeastern us state. My options for acceptable Internet are nil. +$120/month for (possibly) 70 Mbps. Which, from experience, is heavily weighted on concurrent usage. Everything else aside, I want so bad the option to have [insert company (google is winning the race)] to come in an offer me something better. If nothing else, as has been stated in other threads, this will be an enormous boon for technological state of ISP's in the so far backward US.

mhurron 14y ago
> My options for acceptable Internet are nil And it will remain that way for a long time. Google doesn't serve the whole metropolitan areas they are in and there is no known plan to expand beyond the two areas they are in. Google has stated they do not want to be an ISP so no plans to expand may not just be them being quiet. So what does this mean for existing large ISP's? Nothing. The same nothing that every little regional ISP means to them.

jlarocco 14y ago

I'm not a big fan of my current provider, but I think I prefer them to the privacy concerns I'd have using Google. Almost every existing Google service is based on snooping people's browsing and serving up ads. They've made billions and billions of dollars doing it, and as the ISP they'd be able to snoop on everything. I'm sure they promise not to, but I just don't trust them, and I think there's too much money to be made for them not to do it.

ineedtosleep 14y ago
I don't buy this. Google would be offering you an internet service here. If you choose not to use their web services, great; you are free to not use them. If you're afraid Google are going to mine your traffic, use ssl, ssh, etc.. I don't see how this is any different with most of the ISPs monopolizing the market now.
avree 14y ago
You trust your current ISP? I've never understood these arguments. Google doesn't 'snoop' on your browsing—they anonymously track your habits to give you better ads as their business model. You used to get terrible ads on the internet—flashy popups, tricky spam ads, etc. Google has changed this, since they realized that they could avoid needing to trick people into click on ads by delivering ads that are actually relevant. Your current ISP can 'snoop' on everything you do too. Why don't you have this odd mistrust of them?
- thaumaturgy 14y ago
  Your own comment explains why: Google has built an empire upon the ability to efficiently compile information about people. I've worked for a small ISP. They are not nearly as competent in that regard. Most of the time, they're just hoping they can finish the day with fewer tech support tickets than they started the day with. I trust Comcast / SBC / others less -- especially since they voluntarily opted to cooperate with the RIAA/MPAA against their users' interests (http://www.dslreports.com/shownews/ISP-Six-Strikes-Plan-Arrives-in-July-118834 http://www.dslreports.com/shownews/ISP-Six-Strikes-Plan-Arri...) ... but I trust Google least of all, speaking just from a privacy standpoint. I realize that it's a little irrational, especially since, as you pointed out, the information is anonymous. (We think.) However, it's not so much that I'm concerned with what Google is doing with that information now; I'm concerned with what they might do with it later. Once the data has been collected, it's impossible to ever know for certain that it's been removed. Data collected about customers will survive middle management and CEOs and entire company boards. It may be sold at any time to anyone for any reason -- or given away freely. It may be used for purposes which I am not imaginative enough to conjure up. That probably sounds a little paranoid, but then again, when I was dialing in to eWorld in 1995, I could not have imagined that 17 years later the cable company would be my ISP and they'd be snitching on me if I downloaded unauthorized content. 17 years is not beyond the life time of current storage methods; who knows what will come next? Finally, Google can't survive forever on advertising. I frankly expected the online advertising market to be having greater problems by now than it is. Still, AdBlock Plus is popular and I've seen a few articles recently that are tentatively suggesting that buying advertising on Facebook probably isn't worth it. Google's building some actual consumer hardware now, which is super neat, but their core competency is in data mining. I think it would be foolish to believe that if the online advertising market started to really struggle, Google would just drop the thing that they're best at and move on to something else. So I don't think it's such a bad idea to treat Google with a little mistrust. (But their fiber project is really damn cool.)

shawnee_ 14y ago

Kansas City is pretty close to the geographic center of the US. ... to reduce the cost of the actual last mile to users’ homes it’s telling people in Kansas City that if they want to be the first to get fiber, they’ll have to convince their neighbors to sign up. The goal is to get a critical mass of between 5 percent and 25 percent of the homes in a given neighborhood (Google calls it a fiberhood) committed to signing up for Google Fiber before ever sending out technicians. Speaking of cost-per-mile as it relates to "the last mile", population-density surely has to factor into how cost-effectively this service can be unrolled. In a densely-populated area, less fiber can serve more people, and can probably greatly reduce the cost-per-household to install. Seems like the Midwest would require more cost for fewer households able to benefit, at least in the beginning of this endeavor. Wouldn't it make more sense to start on both / either coast, and literally reduce the "cost" of unrolling it marginally to the less-dense areas of the country, proportional to physical distance?

wmeredith 14y ago
Kansas City's metro area is as population dense as any other developed urban core. It's not like they're starting out in a wheat field in Kansas.
- humbledrone 14y ago
  This table seems to disagree: http://www.demographia.com/db-uscity98.htm http://www.demographia.com/db-uscity98.htm . From the totals at the bottom, I found that the average population density for the 600 largest cities in the U.S. in 2000 was (total population)/(total land area) = 99,000,000 / 34,000 = 2911 people per square mile. Kansas City's density is 1100-1400 people per square mile, meaning that not only is it false that it's as "dense as any other developed urban core", but in fact it is not even half as dense as the average city (with population 50,000 or higher).
  - tmrggns 14y ago
    I think part of that is the way the city is laid out. I think Kansas City is more consistent than most cities in having major roads travelling about every five or six blocks. This probably helps when physically putting the fiber down.

fsckin 14y ago

Imagine for a moment, if they fudged the numbers a bit on the storage capacity of every box going into your home. They advertise the usable capacity of 1 or 2TB. In reality, it has 30-200% more storage that is not usable by the customer. They just made a very fast, low latency, distributed mesh CDN that the customer pays Google for the connectivity AND electrical bill.

throwaway64 14y ago
they don't really save anything on the connectivity, as everything would still be routed through Google, thus it would use exactly the same amount of bandwidth as if the storage was sitting inside a data center. As for electricity, the cost of packaging up and sending drive units to the customer + upkeep would dwarf the costs of electricity for the total lifetime of the drives. I don't think this really makes sense.
- dullcrisp 14y ago
  Also regarding the electricity, would that not be considered theft, if nothing else.
  - derleth 14y ago
    I really don't see how.
    - dullcrisp 14y ago
      If they don't disclose that the box they gave you purportedly to record TV shows is actually also drawing power from your line and using it to run part of their datacenter? I would think that that could be considered theft by deception. It's certainly very shady. I'll admit that the line is a bit blurry.
      - SquareWheel 14y ago
        If they were to actually do this, they would of course disclose that fact. It would be written in small print, but still disclosed.

whatusername 14y ago

Does anyone in Australia want to take a stab at what this is likely to mean for the politics of the NBN?

mattmiller 14y ago
I thought NBN already planned to do fiber to the door for most of the country.
- whatusername 14y ago
  So one side of politics will claim this proves that FTTP is the correct solution. The other side will claim that Private Investment is the way to build Fiber rollouts.

zmmmmm 14y ago

So a key part of the "cost saving" is that AT&T gave them cheap access to poles. Doesn't that mean they are relying on one of the very competitors they will displace to collaborate with them in this? Seems to me that either AT&T was forced to do this by competition laws OR they indulged this little experiment as a one-off but the chance of them or any other telco letting Google scale this up to the point where it matters is zero. Does anybody know which one it is?

jws 14y ago
I believe most of the poles are not AT&T's, but rather owned by the county and it's Board of public utilities. The combined power and communication poles have a standard fee for attachment. Google got a special deal to use the power area of the pole for free as part of their enticement. Labor cost is higher for the power area of the pole. The AT&T deal is probably for just the odd pole where the communications approaches a building from a different path.

amalag 14y ago

Curious why they don't have another mid range offering, 5Mb/s free then 1000Mbs at $70 are the only options? I would love something in the $15-$25 range.

dsr_ 14y ago
By making the low tier free, they avoid many costs. Billing. Tech support. Compared to their gigabit-chewing neighbors, the free tier isn't even using bandwidth of any significance. Supporting a middle tier would cost as much as the high tier in everything except bandwidth. The one-time payment of the free tier helps pay for the infrastructure in the neighborhood, but doesn't incur any real costs after that. (I expect customer support for free will consist of a web form to report an outage, and a wiki.)
- Avenger42 14y ago
  > customer support for free will consist of a web form to report an outage And how do you get to the web form if there's an outage? :)
  - Retric 14y ago
    Cellphone? However, I suspect they could automatically detect outages based the in home battery backed up router not responding or losing contact with all other devices on the network.
splinterofchaos 14y ago
I suspect offerings like that will be rolled out within the next year or two, if all goes well.

Jun8 14y ago

"From the infrastructure on the back end to the TV and Wi-Fi routers in the home, Google has built its own stuff. Most carriers rely on outside vendors to sell them networking gear and even set-top boxes." If I were Comcast I would really be afraid at this point, because Google now has access to top-notch set-top-box, as well as other CATV equipment know-how through Motorola Mobility's Home division. If then can incorporate that knowledge into their system they would have an end-to-end entertainment delivery system.

vyrotek 14y ago
I really hope they are. Comcast did some lobbying some time ago in Utah which prevented local fiber ISPs from being able to install new lines in many cities. It's very frustrating!