Everyone that's wondering why this doesn't happen, the answer is simple: private enterprise.<p>The ideal network is publicly owned and privately operated/maintained, but that will never happen because the state would be impinging on the liberty of corporations to profit. In a reasonable state, the gov't would connect every home and operators would compete on service. Instead, we take the tacit monopoly granted to early operators and extend that into the abysmal dystopia we live in today, one where carriers control all.<p>You could have ubiquitous 10Gb links to every home, if only some one would let you, and then folks start to say "well why don't we do this for every industry. Why if everyone paid their fair share we could all have cheap healthcare!". And therein lies the problem.<p>Profit and public interest are diametrically opposed in the case of utilities. Nevermind that most carriers operate over publicly licensed spectrum, or send cables along publicly owned land; a monopoly is the epitome of capitalism and Telecom is printing money.<p>In short, ubiquitous fast internet access has never been a technical problem or a distribution problem, it is a social problem; just like healthcare and, to a lesser extent, power and water.<p>::EDIT:: I am not saying that profit or capitalism is wrong, only that in the case of utilities, on which we all rely, it is a bit strange.
Hopefully I haven't messed any of the numbers up:<p>$11 billion per year for 20 million homes or roughly $550/home/year or $45/home/month<p>And, for perspective:
From 1936 to 1938, the Rural Electrification Administration spent $210 million ($3.37 billion in 2012 dollars) to connect 100,000 rural farms to the electrical grid at a cost $950 each (over $15,000 in 2012 dollars).<p>Edit: Per the title update from "annually" to "over 5 years" the above, non-historical numbers are off by a factor of 5. The costs would be $110/home/year or $9/home/month over 5 years!
UPDATE 3:45pm CT: Kirjner wrote us to say that he had "picked up a typo in our write-up. The $11 billion is not annual but total. I corrected it in our files. Please do so too if you can." In other words, rather than Google spending $11 billion annually, the analysts estimate this is what Google would spend over five years
Only $11 billion? That's nothing (we probably spend more in month or two on Iraq). Why can't the government roll out high speed Internet to everyone and lease the lines to ISPs? Somehow we have gone backwards, there was more choice and competition in the dial-up networking era than now.
I've had a couple of experiences in my career where the enormity of something just sort of changed my world view. The first time was when I was working as an intern at IBM during the summer and came across the complete list of IBM's property holdings. Back in the 70's they would have made the largest REITs look puny. The second time was about a month into my tenure at Google when I discovered a tool called 'network weather' (which, appropriately, gave an overview of what was going on across the global network). Without being specific, I can say that Google fully appreciated the importance of network bandwidth on their ability to operate.
their analysis of why Google is doing this in the first place:<p>"we believe Google Fiber has two related objectives: first, Google is seeking to figure out whether or not, or under what conditions, it can make money as a facilities-based provider of broadband and pay TV services; second, it is an opportunity for Google to test new applications, new ad formats and delivery models (e.g., targeted TV ads) and to get further insight into consumer behavior."<p>This completely misses what I believe to be the primary reason, which is raising the bar for internet access, regardless of who's providing it. Chrome started out similarly, though my impression is that Google does care about dominance for that now.<p>But the point is: if google fiber in 3 cities means comcast gets better in 1000, that's a win for google.
A collective action startup could help solve the problem for neighborhoods wanting FTTH. In a Groupon/Kickstarter model, each house commits ~$3,000 toward a goal. When 250 households in a neighborhood commit and collectively commit ~$750,000 (i.e. $3,000 x ~250 homes) then sub contractors lay the fiber in the neighborhood.<p>As an aside, the Australian government spent ~$4,000 per household to connect each house to FTTH.
"Probably will never happen" is a phrase that has been applied to human flight, walking on the moon, the internet, just about everything we take for granted now. I tend to dismiss someone's opinion when they utter such words.
How about this math 11,000 million dollars per year divided by 115 million households, divided by $70/ month, divided by 12 months per year. That's 11% percent of households to break even?<p>65% of adults already have broadband at home in the US. So will about 1 in 6 switch to 100x better for marginally more money?
$11B year for 20M homes = $550/year per home = $45 monthly<p>Currently Google charges $70/mo for the service. The numbers doesn't look that bad to me, am i missing something?
Ironically their analysis actually convinces me of the opposite proposition. Frankly, the most neglected part of their analysis is the a can Google actually make money from customers who only pay the initial fee and never pay again for 7 years. Could they conceivably make money on a per customer bases on a customer who paid nothing? I believe they can. Does Google Fiber have the reputation of being the best ISP in the United States. I believe it does. Does that branding offer a regulatory advantage? Will municipalities forgo revenue for high quality utilities? That's probably the question that keeps ISP execs up at nights.
What's missing is the return on investment data.If it's good , the $11B number would be an advantage: a good investment at a huge scale, something many companies are looking for.
There is also a great by-product to going national... Google will now not only know everything that passes through their search engine, they will also know every customers web habbits and, unlike traditional ISPs, can capitalize on them. This in itself has to worth a couple Billion.<p>Also, what if, by chance, google allowed you to reduce your monthly bill by say $10 a month by making your default browser google.com?
$11b to destroy a media and internet distribution monopoly?<p>Owning the distribution can allow them to enter markets that don't seem so crazy when you consider that google could do offer a YouTube user an upgrade to publish their channel to their local market for a fee.<p>There are probably a dozen things I can't think of that Google can do with a media monopoly so I'm not ready to say that $11b is too much.
A month ago this would have seemed perfectly sensible, but if this is just a platform for experimentation and data-gathering, why do another city? Austin is apparently similar in demographics to Kansas City, so I'm not sure you'd get considerably different results, and I can't imagine KC is too small to provide a reasonable sample... seems like there has to be more going on here than that.
If done by private enterprise, the spectre of blackspots in non-revenue generating areas that would entrench rich/poor district disadvantages is what I'm concerned about.
Indian Billionaires doing the same.<p><a href="http://www.hindustantimes.com/business-news/CorporateNews/Ambani-brothers-bury-hatchet-sign-Rs-1-200-crore-deal/Article1-1036119.aspx" rel="nofollow">http://www.hindustantimes.com/business-news/CorporateNews/Am...</a>