The future of Fiber | Hacker News Reader

The future of Fiber

fiber.google.com

449 pointslawdawg282 comments

I'm really, really glad they chose Kansas City as the pilot city despite not living there. If they had just gone and chosen San Francisco, Chicago or New York, it would have been no more of an experiment than offering more cable channels to the big cities in the 1990s—It just is completely uninteresting.

On a side note, If you haven't been to KC's Startup Village, the climate is electric! It feels like something really real is going on there. It is just awesome to see someone with a coding question, you can walk literally a few doors down to a different house to ask one of the startups there. When you walk inside, you see rooms that should be living rooms with hackers on laptops and would-be dining rooms with iMacs setup on tables. Meanwhile a sleepy hacker is waking up and making breakfast in the kitchen. The climate is wonderful.

I see it the opposite way: Deploying fiber is easy if the city gives you carte blanche as KC did. Doing something easy proves nothing. If Google Fiber wants to prove that their model is replicable they need to deploy it at a profit in a politically hostile environment.

You usually make things less politically hostile once you've deployed successfully in a few politically friendly zones. People's opinions aren't immutable.

> Doing something easy proves nothing.

Depends on the audience. Joe Random Voter, seeing his cousin's internet being super fast, will carry that knowledge back to his hometown. He doesn't care how easy it was for Google, only that it's possible.

The point is not to prove anything but to shame competitors into providing good quality service. Google Fibre is not about making money, it's about driving the industry to provide faster Internet, which Google believes will lead to more ecommerce and thus more profit.

"If Google Fiber wants to prove that their model is replicable they need to deploy it at a profit in a politically hostile environment."

Well actually no they don't. They're going to prove their model is replicable and profitable, while not deploying to politically hostile environments.

This.

Their goal is to change politically hostile environments by deploying in friendly cities. Then, residents of the hostile cities will start asking, "Why are we paying extortionate rates for shitty service?" Even if Google Fiber never gets to deploy there, the threat of it will hopefully be enough to get cable companies to act more fairly.

I think their strategy is good. Demand comes stronger after you show results. The pitch is much stronger when you say "look at all the benefits this city got for welcoming us in". This isn't about proving a point, it's about bringing fiber to as many people as possible.

Maybe... but it's probably better to work out the kinks first in some place that is easier so that you don't fail miserably in Philly.

> If Google Fiber wants to prove that their model is replicable they need to deploy it at a profit in a politically hostile environment.

Sun Tsu would flunk you out of strategy class. Why should Google engage with an evil, bribery-based, rent-seeking competitor, and their bought politicians, until they run out of relatively easier conquests, and have the revenue from those conquests funding further expansion, and their relationships with independent content creators at a much higher value from a larger customer base?

Particularly when those competitors are not seeking to confront Google, and are preferring to play rent-seeking games with anti-neutrality in their networks, thus making their end game even more brittle?

Sun Tzu

I'd agree completely if Google's goal was to make money from Fiber, but I suspect their goal is to demonstrate that fiber is economically viable and shame other ISPs into deploying it. By that standard, winning on the easiest setting seems unlikely to convince anyone.
See http://googlefiberblog.blogspot.com/2010/02/think-big-with-g... http://www.dslreports.com/shownews/Many-Realizing-Google-Fib... for Google's motivations.

That could be the overarching goal but it makes sense to pilot the technology first so they understand the pitfalls before engaging in a more difficult endeavor.

I suspect wmf lives in a city that he might define as politically hostile :)

Actually I'm in Austin where we're theoretically going to witness three-way fiber competition between Google, AT&T, and Grande (although I suspect they will all find ways to avoid serving my house). I do find it sad that any thread about broadband seems to devolve into whining about how crappy it is in $MY_LOCATION.

"Meanwhile a sleepy hacker is waking up and making breakfast in the kitchen. The climate is wonderful."
...really? It sounds more to me like that hacker in the kitchen is being overworked and feels compelled to not even go home at night. Going straight from the bed to the computer (or vice versa) is an unhealthy, unsustainable lifestyle.

Sometimes its fun. It's a different situation for everyone. As a young person with no obligations, I love working late into the night on big crunches, followed by some slower, relaxing weeks.

Your music is awesome :)

Thanks! :)

>feels compelled to not even go home at night
I don't think you're following. This isn't the company that has had a Series A and Series B.
The guy in that example lives there. There was another place that was hacking away with no air conditioning; It was just a bunch of passionate 18-year-olds with their baby of an idea. This isn't a W2-health-insurance-401k lifestyle, it is a Jobs-and-Woz-in-a-garage-we're-broke lifestyle—and that is one of the huge reasons it is exciting!!!
I implore you to visit. You can't walk away from a tour not feeling energized by everyone involved.

I know it would be painful, and the local politics certifiably insane, but if Google wanted draw attention to the dysfunctional monopoly of the cable industry, there would be no better place to wire than Philly. Take the competition right into Comcast's back yard. Start with a rollout of the Philly 'burbs where the Comcast execs live and work inward toward the city from there.

You live in Philadelphia, don't you?

I live in Philadelphia, where we have a subway station named after AT&T and the tallest building is the Comcast center (until they build their new Comcast building, which will be even taller), and if Verizon can't get fiber to my house (never mind that free wifi movement that sputtered out six or seven years ago) there's absolutely no hope for Google fiber happening here. Between old infrastructure and big players, decent fiber internet is just something I accept that I'm not going to get in this city. There's plenty of other reasons I like it here, but Comcast domination is not one of them.

Well Verizon was finally given approval in 2009 for FIOS in Philadelphia. but they suddenly stopped their FIOS rollout, coincidentally at the same exact time the cable companies gave Verizon a big chunk of their spectrum licenses. Of course everyone involved insists the two are completely unrelated, even though Verizon had been trying for years to get into Philadelphia.

Verizon had started selling off it's Fiber installs before 2009 (which I still scratch my head at). While they may not be directly related, I'm sure it didn't hurt the negotiation.
Edit: See https://en.wikipedia.org/wiki/FairPoint_Communications and https://en.wikipedia.org/wiki/Frontier_Communications

> Verizon had started selling off it's Fiber installs before 2009 (which I still scratch my head at).
Well, it's not like they were trying to get rid of fiber. Verizon sold off all its wireline operations in rural and non-core areas. Fiber simply went along for the ride.
What Verizon was trying to get rid of was the copper. No company would be willing to buy the copper from Verizon and then compete against FiOS. Thus, the only way that Verizon could get any buyers was to bundle the copper together with the fiber.
This would make it more difficult for Verizon to attempt a market entry at a later date, as it would have to start from a position of no infrastructure.

buddy of mine in south philly (by the stadiums) has fios at their full speed.

I live in Philly and got FIOS fiber recently. Girard station area.

Which Girard station? :) Are you in an apartment building? My house is at about 12th & Poplar, and a couple of years ago I saw a lot of Verizon activity on the numbered streets, but nothing ever coming down to house level.

I got FiOS in Germantown.

Yes, but I'm one of the fortunate ones with really fast FIOS and an actual choice between Verizon and Comcast. I wasn't necessarily making this point from the perspective of someone who wants Google Fiber. I was more saying that if Google's aim is to get incumbents to change their behavior, there is value in doing that in your primary competitor's front lawn.
I remember a Comcast exec a few years ago made the point that it was hard to do deals with Hollywood execs because the execs all lived in places in California that didn't have Comcast service, and couldn't really see the innovations Comcast was talking about.

I've seen detailed cellphone signal quality map (sensorly.com) and I can only confirm: the block around the CEO’s houses have far better signal. Having acquaintances rejoice over the ten-fold improvement (and how crisp their grand-children are on Skype now!) at the Club House would indeed hurt.

It would not hurt one bit. You're assuming that billionaire execs care about and take pride in their business. The only thing they take pride in are their fortunes and place in high society - companies are just the means to get there and nothing more. Otherwise they would not allow such mediocrity as can be seen in every mammoth tech company.

Well done sir, that may be the widest brush I've seen used to paint with today!

It would go against their strategy. By wiring up all the GF-friendly cities first, people living in GF-unfriendly cities start asking "why can't we have that, too?" Thus ratcheting up pressure on the local political systems.

This is something that works in theory, but not real life. Internet is becoming a utility, and most people only ever have experienced one utility in their life. People in the SF Bay Area choose between AT&T DSL and Comcast cable internet - two of the WORST providers, ever. Verizon FiOS eats both their lunches, yet, nobody rabble rouses to ask for fiber in that area. If people in Silicon Valley won't even pressure their politicians (there are FAR more urgent issues), why would people in other areas with far more pressing economic concerns?

We have Sonic DSL as well (with significantly better customer service and uptime than AT&T or Comcast).
That said, I'd definitely be happy to 'cut the cable' with Google Fiber in Palo Alto. There's already a bunch of dark fiber spanning the city, Google Fiber is already active in adjacent Stanford property, the residents are (in general) willing to pay for quality service, and the city is not so large as to make the project untenable.
Of course, Palo Alto has a long history of community / municipal connectivity projects with mixed success, and a notoriously-fractious decision making process, so nobody here is holding their breath...

It seems plausible that apathy would prevent people from agitating for faster internet speeds, when speeds are only slightly below the national average (http://www.speedtest.net/local/san-francisco-ca). It seems implausible to claim that such apathy will remain absolute and unchanging if/when other cities started getting major upgrades (http://www.speedtest.net/local/kansas-city-mo), and especially if the national average began pulling away. Plus, the CA regulatory environment is somewhat of an outlier in terms of its conduciveness to infrastructure progress, so it may not be the best litmus test.

Ah yes, the Wal-Mart model.

No. They should roll out to insane hamlets like Climax, NC [1].
[1]: https://en.wikipedia.org/wiki/Climax%2C_North_Carolina
no, I'm not from Climax...but it's an awesome name for a...settlement?

Kinda like Athol, Massachusetts?

Cumming, GA

Intercourse, PA

I grew up minutes from there, and there are even better ones (in the same vicinity).

Hell, Michigan.
Imagine the tagline: "Even Hell has fiber, why don't we?"

Hah "local politics certifiably insane": understatement of all time! Ahh Philadelphia politics.
Also, Verizon just rolled out some FIoS infrastructure in south philly, maybe 2-3 months ago. It was really cool walking around and seeing unfinished junction boxes with these cables running into them with huge "FIBER OPTIC CABLES" orange warnings on them.

no better place to wire than Philly...Start with a rollout of the Philly 'burbs
As someone who lives in the Philadelphia suburbs, I'm not sure if Comcast has the stranglehold in our area that you think it does. I only know one person that has Comcast and the only reason that person has it is because they don't have access to Verizon FIOS.
FIOS is faster, more stable and generally cheaper than Comcast Internet. Comcast constantly advertises how their Xfinity is faster for wireless, but they're basing that on the fact that Verizon gives you an 802.11g router while Comcast gives you an 802.11n router.
With FIOS, I pay $74.99 per month for 75 Mbps downstream, 35 Mbps upstream Internet with their Prime TV package and unlimited landline phone (need for alarm system). It's rock- of-gibraltar stable, I consistently get the same exact speed on speed tests that is faster than what I pay for, and it's cheap as hell.
With Comcast, it was ridiculously expensive, the speed was hit or miss and I generally had to reset the router and/or cable modem about once every few days.
So it would be great to have Google Fiber as an option and I hope it happens in the Philly area but there are places that are hurting for good options far more than we are.

I agree, especially where I live I actually have a choice between Verizon and Comcast. I have FIOS and love it. Best internet service I've ever had. So fast you can almost hear the bits as they blast out of the pipe.
But my point wasn't that Philadelphia isn't a competitive market. My point was that, as a Comcast executive, seeing Google fiber trucks all over your neighborhood and coffee shops is a bit of psychological warfare so to speak. Furthermore, you know Comcast has all sorts deals in place that help them maintain their monopoly. Google would bump into each of those head-on and make hay out of it in the local press and online. At a time when Philly is trying to increase its image in the national tech scene, this would be a very interesting place to shine some light.

I've been pretty happy with Cox Cable in Phoenix/Tempe... I generally get 105-120mbps up and 20-35mbps down in speed tests. I'm generally limited to however fast I can get something from the server at the other end.

s/up/down/ .. just realized I transposed my up/down, but the point stands.

Full list of cities, from the FAQ:
Arizona - Phoenix, Scottsdale, Tempe California - San Jose, Santa Clara, Sunnyvale, Mountain View, Palo Alto Georgia - Atlanta, Avondale Estates, Brookhaven, College Park, Decatur, East Point, Hapeville, Sandy Springs, Smyrna North Carolina - Charlotte, Carrboro, Cary, Chapel Hill, Durham, Garner, Morrisville, Raleigh Oregon - Portland, Beaverton, Hillsboro, Gresham, Lake Oswego, Tigard Tennessee - Nashville-Davidson Texas - San Antonio Utah - Salt Lake City

And with line breaks:
Arizona - Phoenix, Scottsdale, Tempe
California - San Jose, Santa Clara, Sunnyvale, Mountain View, Palo Alto
Georgia - Atlanta, Avondale Estates, Brookhaven, College Park, Decatur, East Point, Hapeville, Sandy Springs, Smyrna
North Carolina - Charlotte, Carrboro, Cary, Chapel Hill, Durham, Garner, Morrisville, Raleigh
Oregon - Portland, Beaverton, Hillsboro, Gresham, Lake Oswego, Tigard
Tennessee - Nashville-Davidson
Texas - San Antonio
Utah - Salt Lake City

And with frivolous awk formatting:
Arizona Phoenix Scottsdale Tempe California San Jose Santa Clara Sunnyvale Mountain View Palo Alto Georgia Atlanta Avondale Estates Brookhaven College Park Decatur East Point Hapeville Sandy Springs Smyrna North Carolina Charlotte Carrboro Cary Chapel Hill Durham Garner Morrisville Raleigh Oregon Portland Beaverton Hillsboro Gresham Lake Oswego Tigard Tennessee Nashville-Davidson Texas San Antonio Utah Salt Lake City cat jader \ | sed 's/ - /\t/g; s/, /\n\t/g' \ | awk -F'\t' '{ printf "%-18s%s\n", $1, $2 }'

Hopefully one of those cities around Atlanta will grow a line out to Winder.

There is such a high demand for this product because it fulfills a great need and requirement to be competitive, but we are being held back by our providers. Our progress, held back due to lack of innovation. Google Fiber will rule the US as soon as it can get rolled out. Cable companies better be really nice and start being competitive. Please bring it to Chandler, AZ.
This article comes to mind, I recently, through normal work, started hitting Cox's 250GB max, I work on games and can easily send 4-8GB per day in assets/code to remote repos. Cringley from 2011...
http://www.cringely.com/2011/07/28/bandwidth-caps-are-rate-h...
This isn’t about capping ISP losses, but are about increasing ISP profits. The caps are a built-in revenue bump that will kick-in 2-3 years from now, circumventing any existing regulatory structure for setting rates. The regulators just haven’t realized it yet. By the time they do it may be too late.
Most U. S. broadband customers don’t get anywhere near that 250 gigabyte cap. The few who do hit those limits are big gamers or file downloaders for the most part. Maybe they do take unfair advantage of the system, but the question is whether this is the proper way to control their consumption? I don’t think it is.
In time we will all bump into these caps and our Internet bills will suddenly double as a result, circumventing competition and ending a 15 year downward broadband price trend.
ISPs win, we lose.
Unless there is competition. Bandwidth is as needed as roads, shipping, airplanes, etc to business and economies. This is an anti-competitive hostage situation we are in in the US. This is also anti-small business as many are run from home offices and co-location etc.

I recently had to restore a 500GB cloud backup over Cox. I called them to ask if they could give me an exemption for one month, and not only did they refuse, but they could not honestly tell me what would happen if I went over the cap, except that I would get an email about it and my Internet "might slow down".
I work from home, and can't afford any degradation of service, so I've been doing the restore piecemeal and checking https://myaccount.cox.net/internettools/datausage/usage.cox every day. It's annoying but doable.
What happens to you when you hit the cap?

Nothing, so far as I can tell.
I have a 12/2 package through Cox and routinely go over the 250GB cap. Other than an automated email saying I'm over the cap, nothing happens.
A year ago, I used 750GB in a month after setting up online backup on a computer. Nary a word.

I seemed to notice the last couple months after I get the messages my service drops on download (but not on upload). I drop to their 5M connection instead. Strangely when I check known download speed sites they work with correct numbers, when I check other means I notice the difference there. So I hope they are not nerfing me wholly except when I try to measure it with known tools i.e. speedtest.net etc. They seem to be nice about it now but who knows, this is the first time they have been as vocal about it.
Sadly the next two tiers (yep we are there) only add 50GB (300) and 150GB (400 total), with the speeds of 25Mbps and up to 50Mbps but if you have higher speeds you will download more and be on HD more, get more files, backup more and always go over.

At least Comcast got something right.
Comcast charges 20 cents per gigabyte in 50GB chunks over 300 after you use up your three "courtesy" overage months. There's a little monthly report in the billing site. The price is thousands of times the actual cost to Comcast, but at least I know the limits and the cost of going over them.
My understanding is that Comcast Business has no caps.

Think about it. It's not about cable internet getting faster. It's about rent seeking monopoly. Once all their subscribers are belong to you, there is no reason to advance any further. Just erect legislative barriers to progress and charge people to your heart's content. Read Tim Wu's the Master Switch. Communications companies have been doing this since the telegraph 160 years ago.

Google Fiber will pass <~0.5% of total U.S. homes[0], even after they build out Austin and take over Provo. They have to start doing things much faster if they will make a dent this decade. Google is selectively building out 'fiber hoods' - neighborhoods that bend over backwards to get the service, pre commit and make construction super easy - not full cities, by any means.
Google Fiber was announced almost four years ago and has only a few tens of thousands of subscribers. While it's fun to get excited about what Google Fiber could be, it will be years before any material percentage of the country has the opportunity to use Google Fiber.
Google is still building out small neighborhoods in Kansas City [1]
[0]https://docs.google.com/presentation/d/1Q-sGUEiuT9VN__VPsFeo... [1]https://fiber.google.com/cities/kansascity/#zone=Kansas+City...

Google Fiber has so far only been an experiment. I see it as a very expensive, very ambitious proof of concept the likes of which only Google is capable of creating. This new announcement is evidence that their experiment of the last 4 years has succeeded.
Google can only move so fast with this given the stranglehold Comcast has on the market right now and the regulatory obstacles. There is a killing to be made by undercutting the monopolistic pricing and terrible service Comcast provides right now (not to mention the ways ownership of distribution might benefit Google's content).
TL;DR This is the next step in Google's big, slow play to become a major ISP.

I think it's likely that this exists mostly as a threat to keep the big ISPs from getting complacent. If internet speeds improve via themselves or the ISPs, Google wins either way as internet usage increases.

It always surprises me how excited people get about Google Fiber, yet how up-in-arms they get about GMail, Google+ and other Google services which tend to invade and/or expose your privacy. Do you really think that using a advertising company's fiber as your gateway to the internet is going to offer a private experience? Don't you see how much easier it will make it for them to gather your personal details and habits? Do you really think Google won't use your data to their advantage?

Google flat out tells us that they are using the content of emails to direct ads. Secret, deep-packet snooping on a broadband connection? That's a whole new level of evil.

Yes, but Google Fiber is a product I can pay for. I don't mind trading personal information / eyeballs for Google's services but I much prefer to give them money to get a superior service I'm already paying someone else for.

We're excited for better service at lower prices, and if you think Google would be (or is) the only one snooping traffic, well I think you're giving a lot of trust to people like Comcast.

You are comparing a free service where they are up front about their plans to a pay for service which doesn't detail any plans you describe. I have to ask, do you think they would?

All the other telecoms have far worse privacy records than Google does. Who were the original companies implicated in illegal NSA wiretapping? Verizon, AT&T, and Bell South. [1] What happened to the CEO of the one company that refused, Qwest? He went to jail on trumped-up insider trading charges. [2]
[1] http://usatoday30.usatoday.com/news/washington/2006-05-10-ns...
[2] http://www.washingtonpost.com/blogs/the-switch/wp/2013/09/30...

... and google is racing to catch up. There is a monopoly to be created and rent to be extracted and no matter how "not evil" an organization is, if they play in this sphere, they must adhere to this model or they won't be in business.
Everything you need to know is written in a very nice book, recently published:
http://en.wikipedia.org/wiki/Tim_Wu#The_Master_Switch
... you could have predicted well in advance that google would go this route.

When the reference point is Comcast one has a high probability of being applauded.

More competition in this space can only be a good thing. Besides, Google has done a whole lot more to protect their users from government surveillance than the telcos have.

The Google Fiber Privacy Policy seems relatively strong in the "Google Fiber Internet is a dumb pipe" sense:
Technical information collected from the use of Google Fiber Internet for network management, security or maintenance may be associated with the Google Account you use for Fiber, but such information associated with the Google Account you use for Fiber will not be used by other Google properties without your consent. Other information from the use of Google Fiber Internet (such as URLs of websites visited or content of communications) will not be associated with the Google Account you use for Fiber, except with your consent or to meet any applicable law, regulation, legal process or enforceable governmental request.
https://fiber.google.com/legal/privacy.html
(Google Fiber TV, not so much)

I guess I fundamentally do not understand what makes a city a good or bad candidate for this kind of service. Why is SF or NY not an obvious candidate? Density means that there are lots of customers for a given physical length of infrastructure rollout, they're both wealthy cities with a large proportion of techies who could use faster internet...
I'm sure they're both nightmares to deal with permitting processes for. Is that it? Or is there something else that makes mid-tier, more spread-out cities more attractive?

I suspect they mostly want a compliant political infrastructure and relatively easy construction. NYC and SF aren't really known for either. Also, NYC would probably be biting off more than Google Fiber has any interest in chewing at the moment. They're still scaling up, after all.

YUP.
The short answer is: it's much easier to build in a greenfield environment where there is a "helpful" local government, ready access to rights-of-way, and underserved population, easy construction costs, access to labor, favorable tax climate, and a host of other reasons that come before "lots of people who want gigabit speeds".
The economics of building (or overbuilding) a gigabit fiber network drive you towards areas where you can get a high degree of penetration for your investment. That's not going to happen in a place like NYC or SF. The construction costs (directional boring) and tax regime (network cables are "property" for property tax purposes) mean that SF and NYC are probably going to be the LAST places to get Google Fiber.

> (network cables are "property" for property tax purposes)
WUT.

Look it up. If you install $15M of fiber in/around town, you can bet your bottom dollar that the local taxing authority considers that real property and will tax you on it at appropriate rates.
In TX, colo providers have had to send business property tax bills to people who locate their servers in those centers. Sometimes those bills are quite shocking.
In CT, the local phone companies started putting DSLAMs on poles because it got them off the ground and saved them millions on property taxes because there was a rate differential between property that was pole attached and property that was on the ground.
Never underestimate the effect that taxes can have on the deployment, operation, and maintenance of business property. You can explain great deals of financial shenanigans to the real savings associated with avoiding taxes.
And who gets screwed? Anyone who doesn't have the scale or size to make avoiding that tax worthwhile. e.g. the small-business owner.

Good answer. I always post this: http://www.fiercetelecom.com/story/att-ordered-stop-u-verse-... when people ask why SF isn't at the top of the list. Existing telcos already can't easily roll out faster service!

ATT is now rolling out U-Verse in San Francisco:
http://www.noevalleyvoice.com/2013/May/Att.htm
The challenge to ATT's plans was manifold:
1) ATT refused to put their equipment underground. SF's sidewalks don't need more cabinets.
2) U-Verse is DSL service. Sonic.net already offers 40mbps DSL service in SF, so U-Verse is not that much faster. Why does SF need more cabinets to get pretty much the same data service from the incumbent telco?
3) ATT is putting cabinets on the sidewalks, and running fiber to those cabinets to serve ~45mbps connections to folks ~800 feet away. If ATT had planned to run fiber from those new cabinets into residences ~800 feet away and provide 100mbps or 1gbps service, folks would have welcomed their project.
4) Monkeybrains [0] and Sonic.net [1][2] have had (for several years) workable plans to eventually deliver 1gbps service to the whole city with (they say) far fewer aboveground boxes that ATT "requires" for U-Verse, and don't even need to dig up the street to lay the fiber. I'm not sure why they've been ignored by the Mayor's Office and the DPW.
[0] https://www.monkeybrains.net/DPW-MonkeyBrains-March-2011.pdf
[1] http://corp.sonic.net/ceo/2010/07/14/micro-trenching-at-soni...
[2] http://corp.sonic.net/ceo/2011/12/15/sonic-net-plans-gigabit...

Putting cabinets on sidewalks still seems like a very minor point and still part of the "People United Against Everything" syndrome affecting so much of America. Maybe U-Verse is not that much faster, but faster is still important!
If Sonic / Monkeybrains can deliver 1gbps, all the better.

Sonic.net leases the lines from AT&T. 40mbps as far as I remember is a bonded ADSL+ service requiring two phone lines. Even then 20mbps is a theoretical maximum which falls off the further you are from Sonic's DSLAM. The U-Verse boxes brings the DSLAM much closer. In my apartment's case, I can only get 5mbps from Sonic.net while AT&T is able to deliver over 20mbps.

The beautiful thing about this in the long-term is that if Google Fiber succeeds and becomes more widespread, people will start to say, "Why don't we have this?" If this succeeds — and I have little doubt it will — local governments across the country will be under increasing pressure from their constituents to move away from the monopolistic deals most cities have now with Comcast.
I see this as a huge step in the right direction and cannot wait to see how it turns out.

To be fair, San Jose has a pretty awful political climate as well. I mean look at VTA.

I'm in San Jose, what's wrong with VTA?

This is answered in their FAQ. The summary is: a wide range of variables.

Probably because they are larger so it would take more time than a lot of the other cities here.

> Why is SF or NY not an obvious candidate?
At least in NYC most of the buildings are older buildings that have already been wired for copper and are very tough to rewire. It's a much bigger task to rewire a building than to run a drop to a house. So then you end up having no idea whether fiber is available in any specific building. Plus many buildings have signed exclusivity agreements. FiOS has the exact same problems and so is generally just available in newer buildings.

That sounds strange to me. I live in Sweden and went apartment-hunting last year. Not one of the apartments I looked at was without fiber (technically, ethernet in the apartments, fiber to the building). Including buildings that pre-date WW1. They all had pre-existing cable and DSL as well. If there's a will, there's a way. I guess in the US there just isn't much of a will yet.

Vacancies in NYC are incredibly low, so a landlord doesn't gain much by being able to advertise fiber internet. This, added to the expense of wiring the old buildings, means that its not a worthwhile investment.

Actually, I and many others I know would pay a significant rent premium for a building that had fiber access, so I think a case could be made for it financially.

In sweden this sort of thing is done by the city/state not 'the market' which makes for a different set of arguments: https://en.wikipedia.org/wiki/Fiber_to_the_premises_by_count...

What about the actual building wiring? It doesn't matter whether there's fiber to the building if cable has enough bandwidth to fill the copper between the cable box and the apartments.

It might work, assuming the building can support multiple cable providers (e.g. it's not all just 1 trunk from CableCo), but it would cost more in equipment. You would need to convert the fiber to coax (maybe with an intermediate step of ethernet), and then the customer would need a modem to convert from coax to ethernet. It'd be a pricey installation and continual maintenance of hardware.

> Vacancies in NYC are incredibly low
Same in Sweden. Often you need to be on a waiting list for 2+ years to get an rental contract in a large city in Sweden (longer in Stockholm[0]). The impatient among us flit around on 6-month second hand rentals. Or you buy an apartment with an untenable loan and hope the bubble keeps inflating...
[0]The wait is currently on average 7 years in Stockholm municipality. 15 years if you want a centrally-located apartment. http://www.svd.se/nyheter/stockholm/lagenheten-ar-din-om-att...

Is there some sort of regulation going on there? In a normal market I would expect rent to rise until the wait time was down to something reasonable. You can always find an apartment within a week or so in NYC - its just going to be really expensive. COL here is about double the national average, largely driven by housing costs.

Yeah, there's some quite heavy rent control http://www.thelocal.se/20131113/rent-control-blamed-for-hous... http://www.thelocal.se/20130723/49206

Same for me. My apartment building in Stockholm is from 1943. It has copper for phone (DSL) and cable, plus fibre right into my hallway.

I wish Google Fiber could be deployed in NY.
However if you consider the cost of deployment, it's much cheaper to do it in a rural area than in the middle of the dense city. I.e. cost per meter of deployed cable can differ a lot.

None of the cities listed on that page are "rural" by any stretch of the imagination.

Then it's not an issue and Google is not scared of digging the earth. On the other hand they write they want to reuse existing conduits, rather than perform new construction.

I don't understand your point. Just because Atlanta isn't as densely developed as New York City doesn't make it "rural".

As I said, if Google is not scared to dig - good for them. But they seem to want to avoid it. See their FAQ.

In every metro area they've chosen, the vast majority of dwellings are single-family, owner occupied. So they can run cables above ground next to the existing telco equipment, and approach homeowners directly when it comes to installing the drop.
In Chicago, about 60 percent of the city is NOT owner-occupied. I'm sure New York is significantly higher. MDU runs are much more expensive, even with cooperative landlords.

Just ask any NYC folks. How many of us are still waiting for that FIOS? I am waiting and waiting. And they promised to complete by 2014. We are almost three months into 2014 and I still have no FiOS access here. A bit above my area has already gotten FiOS two-three years ago and my area still hasn't gotten any!
I am angry. We need to break this franchises.
> The company also appears to be blaming landlords for any hold ups in deployments
Yeah. Excuse. I haven't heard any landlord in my neighborhood ever complain about this. A lot of my neighbor landlords (they don't actually live there, just for rent), when they come by we would have conversation about these things (because they know I do some computer stuff and they would come and ask for advice when TW goes down), they say they'd make more rent if FiOS was here.
http://www.dslreports.com/shownews/Verizon-Accused-of-Laggin...

VZW stopped rolling out FIOS across the board I believe.

Yeah I've heard/seen this as well, they've done it in favor of expanding LTE coverage [0]
I'm astounded at how anyone at VZ thinks this could be a good idea.
[0] http://www.dslreports.com/shownews/Verizon-Again-Confirms-Fi...

I believe the FIOS demand wasn't high enough to justify the costs of expanding it any further.

fios penetration rate is 35% which is not bad. http://www.lightreading.com/fios-slows-but-still-grows/d/d-i...

There are a lot of folks on forums like this who think networks like FiOS should be regulated into being essentially dumb pipes. Verizon and their investors see some chance of this happening, and even if it didn't, it turns out there aren't a lot of "value added" services you can sell to consumers over a proprietary network; they just want the internet.
You can't make the margins as a "dumb pipe" that Verizon needs to justify the capital investment in FiOS. That's the long and the short of it. They had dreams of selling pay-per-view movies, video chat, and a dozen other applications over their network for big $$$. Instead they're fighting with Comcast over the same $120 in cable and internet bills, while Netflix, Google, and Apple build their high-margin businesses on VZ's fiber.

It's amazing to me how many people here on HN complain on one hand about companies not wanting to make massive capital investments to build out faster internet service while simultaneously espousing regulating such services into low-margin dumb pipes.

I think there's a simple solution - municipal fiber everywhere with companies allowed to lease access. Internet access looks like a utility and quacks like a utility, so it should be treated like one. The timeline for breaking even on the investment has been pretty reasonably short in the examples I've seen.

The municipalities have no money to build, and more importantly, to maintain and continually upgrade, such networks. Heck, they can barely maintain utility systems that evolve far slower than internet connectivity: power and water. See: http://www.nytimes.com/2009/04/18/us/18water.html?_r=0. People outside of HN won't be anywhere near sufficiently eager to earmark sufficient public dollars to properly maintain such networks, not when pensions systems are underfunded and schools are putting teachers on furlough.

Yep, it'll never happen.
Except where it did before the State stepped in to ensure Comcast/TWC didn't have competition.
But other than that it'll never happen.

I don't know enough about fiber to have a very intelligent discussion about it, but from what I've heard, one of the nice things about fiber is that you can run increasing amounts of bandwidth through the same fiber once it's laid as the endpoints improve.
I would expect that the buildout could be funded by a municipal bond issue, with the repayment being handled via the revenue from the customer-facing companies that lease access to the infrastructure. I've heard of municipal networks like this being run profitably and being repaid in a reasonable timeframe (on the order of a decade or two), but unfortunately I don't know more than that.

I think the point is: local governments can barely pay for essential services (water, power) that most everyone in their district needs. High speed fibre that a small subset of their constituents would use is way down on the list.

I think you're right that that was his point. It wouldn't necessarily come out of their budgets, though, it could be done with debt financing on the project itself, repaid purely from the revenues from leasing access to the network to the customer facing ISPs.
I also don't think it would necessarily be used by a small subset of the constituents - it seems as though it could potentially be much cheaper than alternative high-speed arrangements, since the customer facing companies would actually have some competition. Once the initial construction debt was paid down, the govt. could likely reduce the access lease cost, which would be passed on by the competitive ISPs, further enabling access by poorer residents.
I think it would be great for the long-term economic prospects of those areas as it broadens access and makes heavy uses like remote work more viable.
Plus, a good internet connection is a much cheaper form of entertainment than many of the alternatives, including cable, so it could actually save many people money.

Large capital expenditures undertaken by municipalities are usually funded by bonds.
A municipality would issue bonds to fund the replacement of their water mains, which are breaking and pouring water into the streets every winter. It would issue bonds to renovate their elementary school, which has a leaking roof, asbestos insulation everywhere, and drinking fountains that dispense leaded water.
Municipalities have limited borrowing power, so muni fiber does have to compete against all those other priorities. And it's been a tough market for muni bonds lately.
What's more, muni fiber would be considered a non-essential revenue bond. These types of bonds carry higher interest rates than essential services bonds, or general obligation bonds.

Ah great, thanks for the concrete info!
I imagine that a fiber rollout would create entirely new revenue, though, where upgrading water mains and other existing infrastructure wouldn't, so it might be viewed by the govt as offsetting any difference in interest rates?

Well, it's pretty common for municipal power companies to return a profit to the city, which is then used to offset property taxes. In fact, a number of successful muni fiber projects have been managed by the municipal power company, or the municipal cable provider.
The problem is that a lot of municipalities aren't run all that well. Local politics can sometimes be very ugly. Muni fiber only really works when the municipal government works. Sadly, that's not as common as it ought to be.

Hm, that is a shame. I wonder if there's a way for technology to help change that. Maybe by offering a graphical budget breakdown of spending that the public can browse? I remember seeing something like that for the federal government, and it was pretty eye opening. I guess there's no incentive for a terribly run government to roll that out, though.

Many other countries have built high speed consumer fiber networks, some of them heavily regulated. Something isn't working in the United States.

The U.S. is ranked #8 in average broadband connection speed in Akamai's most recent State of the Internet. That puts it ahead of most of Europe: http://www.akamai.com/dl/documents/akamai_soti_q213.pdf?WT.m... (page 13). If you look at more densely populated coastal states, the U.S. looks even better. Yeah, South Korea has widespread fiber internet. More than half the country lives in the Seoul metro area. You can't compare.

Nation-wide averages tell a story that no one is interested in. Do you have data on densely populated coastal states? I want to compare speed and speed/Mbps to areas like Warsaw and Stockholm with high density cities in the US.

The problem with drilling deeper is that America is organized quite differently than European countries. It is typical for a European city like Stockholm or London for 40-60% of the people in a metro area to live in the denser, core city. In the U.S., that number is more typically 10-20%. Moreover, because of historical reasons, the dense inner cities in the U.S. that are most geographically amenable to building out fiber have huge proportions of poor people who couldn't afford it. Middle class people who might buy fiber are disproportionately more likely to live out in suburban areas less suitable for building such infrastructure.
Very concrete example: Delaware is one of the fastest states in the U.S., with average connection speeds in Akamai's study comparable to the top 5 countries in the global list. I live in the largest city in the state, a major rail hub, in the heart of downtown, and can't get fast internet because the area is so poor. But the suburbs right around us have fiber.
Per-city breakouts aren't in Akamai's most recent report, but here's one for Q4 '11: http://tumblr.thefjp.org/post/22270191498/the-worlds-fastest.... The fastest city in Sweden at 11.3 mbps wasn't that dramatically faster than the fastest in the U.S. at 8.4 mbps. Asia of course dominates, but that's because Japan and Korea have tons of very densely-built cities.

Hey thanks for the link.
I would love to test all your hypotheses, I'm skeptical. My impression is that, even if you compare dense high-income areas (Manhattan, SFBay), many other countries are still coming out on top. For instance, Warsaw (!) metro area: 631.4/km2, Manhattan: 27,227.1/km2. I don't have the numbers but income is much higher in Manhattan.
Wikipedia tells me I can get 60mbit/s down for under 25 USD in Warsaw and I think that is uncapped. I don't believe any comparable deal exists for Manhattan. SFBay is similar.

Its a shame, regulated pipes is starting to look like the most reasonable solution when providers are doing exactly what you said, eyeing up higher margin models and drooling.
Regardless of opinions on free market, political leanings, etc, the fact that this is standing in the way of progress is bad news for everyone.

I wonder how many people honestly care about what internet provider they have? Do most people want FIOS? Or would more people prefer to have a much better connection on the phone/iPad?
On top of that, as was mentioned above, it's probably cheaper to update to LTE than it is to run fiber everywhere, and on top of that, it's probably worth more money to them on the $x/mo for LTE service since to add another customer in the area is trivial compared to FIOS.

It isn't about the provider, it's the speed we care.
Are they going to let my router able to get stable internet from LTE technology? I have desktop and laptops and I don't want to add wireless card to my desktop in particular. I rather serve things over my own router. How are they going to market to home users? FIOS has been said to be high speed, reliable, fiber-optic. Great. I am willing to add $20 to my monthly bill to get one.
And how can they drop the agreement like this?

Yes. This. Having had FiOS before is the worst part, knowing first-hand just how subpar my TWC service is (there's no going back after getting 35mbps upload speeds). It seems completely ridiculous to me that in 2014 Brooklyn we're still choosing block-by-block where to live based on the internet service.

What's funny is that the rent premium for fiber is higher than the prices people complain about for Cable.

Utah: large NSA datacencer -> Good. See how this works?

Why couldn't they lay fiber to core areas, then setup some sort of wireless redistribution - there'd be some lack of speed due to more users using same connection, but it would mean less construction required. As they could just setup more and more access points throughout the city, and some sort of wireless login for specific users... --Might be a possible solution.

That sounds like LTE, which you can already get.

Doing any kind of utility work in NYC is a nightmare. There's lots of stuff underground (power, telecom, water, steam, sewer) that has been built up over the years and is in various states of obsolescence and disrepair, and in many cases isn't really marked on any utility maps. It's not a trivial matter to just dig a new trench for fiber.
Here's a interesting story of what happened when an electrician found a 250VDC feed that was supposed to be abandoned decades ago: http://www.electriciantalk.com/f5/fixture-tails-44009/#post8...

And here is a cool cross section of what lies underneath New York City:
http://www.nationalgeographic.com/features/97/nyunderground/

... dangit, I wish I knew more about electrical work. Why was that so dangerous? I mean, besides 250 volts and rotted-out insulation. The writer makes it sound like he just discovered an unexploded bomb.

The steam pipes that blow up every couple years are the unexploded bombs.
http://en.wikipedia.org/wiki/New_York_City_steam_system
"At least 12 steam pipe explosions have occurred in New York City since 1987."
http://en.wikipedia.org/wiki/2007_New_York_City_steam_explos...

Good question! AC voltage is relatively safer because it crosses the zero point 120 times a second. If it passes through your body, your muscles will spasm but you will have an opportunity to pull away. Also, AC voltage will probably not sustain an arc at such a low voltage, again, because it is constantly switching polarity.
On the other hand DC voltage, once a connection has been made, will continue to flow through that path to complete the circuit. You can think of it as the electrons on the hot wire want very badly to get back to the place where they were generated (the source transformer under that covered up manhole). If they find a path through your body, it will continue to flow through your blood vessels until either you or the wire burns up, or the upstream transform faults. You basically become a fuse. If you are part of the circuit, you'd better hope there is somebody nearby to hit you you with a pole, because otherwise you are pretty much stuck that way. Also, DC voltage is much better at sustaining an arc, even below 100 volts, so it can set off an explosion that can ignite whatever dust is lying around the equipment. Google "arc flash" to see what this looks like.

> Also, DC voltage is much better at sustaining an arc, even below 100 volts, so it can set off an explosion that can ignite whatever dust is lying around the equipment
Ahh, that's what I was missing. I assume the "painted black" part was to insulate the box and mitigate this risk when the box was closed?

AC is lethal at lower RMS voltages. Two reasons: (1) peak-to-peak voltage is higher (2) It is not synced to heart, so can cause more damage

Yes absolutely. I never said AC wasn't dangerous.

When major Telco players decide how much to invest in opposing Google, the first place they look for a measuring stick is the potential lost revenue numbers.
The huge customers in cities on the coasts would give them every reason to fight tooth and nail. Going for small potatoes first represents a larger financial risk for Google, but it gives them a better chance to work out the kinks in their product in relative peace.

I looked at the map and was curious - are these cities each in somewhat close proximity to existing google data centers? It would make sense if fiber, much like app engine is just productizing existing google infrastructure

I think their strategy is to get it out to as many people as possible in the most efficient way possible. I imagine doing construction in SF and NY is a nightmare so it would be a fight to get it done. If you do it in the easy places first, show all the benefits, and get the big cities jealous, you can drum up more demand than if you go through a painstaking multi year process just to get it in one city.
Their rollout to Kansas City was impressively fast. Had it taken years it would have done terrible things for their campaign.

When Google came to Provo, UT, it wasn't because of the friendly government. It was simply that it already had plenty of fiber and a bankrupt ISP that Google could buy out. There's plenty of fiber that's been buried in the ground since the 80's. Some is in use, some is not. Some has more potential. Here in Utah, we have another fiber company called Utopia... I wouldn't be surprised if Google isn't just going around buying these companies and using the existing fiber. I'm probably way off base.

I had the same thought when I heard this. With Utopia already laying down the fiber (and keeping track of where it is compared to Provo) it would be smart move for Google to buy up Utopia.

I wonder if the existance of "dark fiber" is a factor. In the late 1990s a half dozen fiber companies tore up our industrial park street laying new fiber. I doubt if much of it is used. One new fiber was laid last year since then.

Woo! Moving to Mountain View in June, so I'm excited to see this.
On another note, I wonder if this will have any impact on the Comcast/Time Warner merger. Comcast has specifically called out Google Fiber as a source of legitimate competition in defending the merger[1]. With the announcement that Google could increase the number of Fiber cities 10-fold, that claim might have a little more weight.
1. http://gigaom.com/2014/02/13/comcast-cites-competition-from-...

I've been hearing that there will be no real threat to that merger. If that's true, then I see this as a real threat to the new Comcast; basically Google saying "you will have real competition even if you are massive".
Edit: Otherwise, why would they announce this at this stage? They've only announced other cities when they were certain.

> "Otherwise, why would they announce this at this stage?"
Drum up local interest and support to help with any local political roadblocks.

Why is there not a single city in the northeast or the Rust Belt?
Chicago, Philadelphia, and Pittsburgh are all major cities that are at least "on the map" as far as tech is concerned. I know for a fact that the local authorities in Detroit, Cincinnati and Cleveland would bend over backwards for any project that has even the slightest whiff of economic development.
I can understand NYC being a special case that they don't want to deal with, but there are plenty of other cities in the region.

I would have liked to have seen one in Ohio.

The Denver-Boulder corridor is going to start losing it's place as one of the top tech startup areas in the country if it doesn't get itself on this list of potential fiber installs.

As a resident of greater Denver-Boulder, I, too, was disappointed to see the big empty space on the map around Colorado. At least one ISP here is experimenting with residential gigabit, though -- I'm hoping that maybe someday I can get it in my neighborhood: http://fiber.forethought.net/

interesting. Their website is a bit lacking in info. Do they offer residential service at all right now? Looks like they are just servicing larger multi-tenant buildings, I'm guessing where everyone needs to sign up to get them to come and install.

It's in the extremely nascent stages right now. I think they'll be turning on the first residential gigabit customers in downtown shortly. We've occasionally discussed the idea of getting my condo complex in Arvada lit up -- it's technically feasible, although I suppose the numbers would have to work out.

I want Colorado on the list too.
All I can think of is: Google's doing this to put pressure on other cities. Is it working? (Not living in Denver anymore, or I'd just ask around :) )

Yeah, also disappointed. I thought for sure it would at least be planned.

The Triangle region of NC would be a great place for Google to start with this.
Seriously, with all of the technology workers working in RTP, and all the university students in the area, as well as the emerging startup hub(s) in Durham and Raleigh, the area could really put Google Fiber to great use.

Cool to see they're expanding, but I'm disappointed to see that there's no love for the upper Midwest. Wouldn't expect it in Chicago but Milwaukee (where I am), Madison, Minneapolis, Indianapolis, Cleveland, Columbus, Cincinnati, Detroit, and possibly a few smaller cities would all seem like reasonable candidates (although I acknowledge there's issues with all of them).

Minneapolis kinda sorta not-quite has at least some parts with fibre.
http://fiber.usinternet.com/coverage-areas/
Barely though to be honest. >.<

Love to see challenges to the pathetic cable monopolies. I'd be willing to pay more than I am now if it meant sticking it to TimeWarner.

I wonder when this will be a reality in New York. The place where we all want to stick it to Time Warner.

Don't worry comcast will make you pay more after the merger, so you'll get your wish granted by an evil genie.

Why not San Francisco, Washington D.C., Boston and NYC?
San Francisco, NYC and Boston because they are big tech hubs where a lot of Googlers live and Washington D.C. because it is the seat of politics in this country and the best way to show what good can happen when we have broadband connectivity.
In fact, rolling out Google Fiber in the capital cities of each state makes the most sense in general. You lobby the political class by giving their home base (state capitals) excellent broadband.

1. Sacramento would be an odd choice for California. 2. I believe this would have the opposite benefit you intend - if you want lawmakers to be on your side, they need to feel your pain.

NYC at least because each building is like its own little fiefdom that's already wired for at least one competing service. FiOS has the same problems rolling out, a lot of the older buildings would require substantial building-wide upgrades to support fiber, something Google is likely not interested in at this point.

Fingers crossed that DC is in the next group of cities :)

I think that NYC and DC already have Verizon FIOS, as do the suburbs of Boston, but not Boston proper. Why compete with an existing and decent fiber service when you can stick it to Time Warner- sorry, Comcast and AT&T instead?

I can't speak for the other three cities, but San Francisco has a number of problems that would make it difficult for a Google Fiber rollout.
First, AT&T is already in the process of rolling out fiber to the home, or in some cases, fiber to the node with last mile copper. They have faced nothing but problems trying to do this from people in the city: http://www.sfgate.com/bayarea/article/Groups-sue-S-F-trying-...
Density can be a nice benefit, but lack of space makes it hard to place the equipment necessary to serve fiber.
Second, a local provider Sonic.net is also rolling out fiber by taking the approach of starting in the closest thing San Francisco has to a less dense suburb (the Sunset). They've faced exactly the same issue (battles over placing Utility boxes).

People have been pushing for municipal-owned fiber for a long time in SF: https://www.change.org/petitions/make-fiber-broadband-a-prio...
But the Mayor thinks that WiFi is good enough. :-(
ATT's last-mile copper solution will barely be better than their current "Uverse".
The City (SF) has tons of fiber running all over the City. With some infrastructure investment, it would not be hard to use that as a backbone for municipal fiber. Residents can then sign up with any of the 3rd parties to jump on to the Internet.

Suburbs of DC (at least NOVA) has a decent number of places that have FiOS. Also the cable situation isn't that bad if you have Cox rather than Time Warner.
But the dominating factor of what determines where people live in the suburbs seems to be schools more so than internet. Google fiber here would be more of a novelty.

Yes... there's also RCN in DC as well yet in my neighborhood I can only get Comcast.

I agree, this is completely lame until Google can serve the western tech capitol, San Francisco.
SF High Speed options:
Comcast. Max out at 100/10.
Sonic offers fiber in certain blocks of the Sunset, but are deploying painfully slow.
I have an ATT U-Verse box permitted to install on my block but it hasn't happened yet, and I'm not thrilled with ATT.
Monkeybrains requires LoS to Sutro tower, and permission from the building landlord.
Webpass requires at least 20 units in a building.
Google remains the only hope; build it, TAKE MY MONEY.

MonkeyBrains requires LoS to any one of their radios, which aren't all located on Sutro. Permission shouldn't be a problem if the landlord is OK with tenants' satellite dishes on the roof. Even then, you can install any communications antenna (within reason) on areas you control in your lease (e.g. a balcony) thanks to OTARD [1]
[1] http://www.fcc.gov/guides/over-air-reception-devices-rule

I live in DC and I'd love to have Google fiber. We supposedly have a municipal fiber network here, but as far as I can tell it's doing diddly squat and I'm stuck with Comcast.

At least for SLC there is already _Some_ fiber around the city: http://www.utopianet.org/
Much like Provo, I think google saw that utopia already had some infrastructure in place and wanted to swoop it up.
So while people are asking, why not this x,y,z city perhaps it's due to infrastructure not already being in place.

Atlanta! Looks like they're targeting the nicer suburbs too. My mom's going to get fiber before I am.

>We’re asking cities to ensure that we, and other providers, can access and lease existing infrastructure.
This is probably my favorite part. They're not just trying to get special privileges for themselves, they are trying to level the barriers to entry so that there is actual competition in this space.

Wish they had Google Fiber in Toronto :(

It would be quite fun watching the incumbents playing the "but we're Canadian" card in order to prevent such a thing.
Hard as it might be for Americans to believe the ISP scene in Canada is in danger of falling behind even them.

Canada related question. Do all the major ISPs in Canada have bandwidth caps on consumer broadband?

Yes, every major ISP has bandwidth caps, some of which are prohibitively small (20GB). Even the smaller upstart ISPs, like TekSavvy are forced to have a cap (though at 300GB) due to wholesale laws. At least they offer unlimited for a reasonable amount of money.
Canada is definitely behind the US for internet access, it's even behind many (if not most) third world countries.

I've seen a lot of complaints about the caps (wasn't sure if they were widespread or not). What keeps the caps in place if they're universally disliked? I've always thought of Canada as a fairly responsive democracy.

Canada is a land where a lot of the big businesses maintain their position by wheeling and dealing with the government under the guise of economic protectionism. i.e. they'll protest that without these limits Canadian jobs will be lost. The most productive stuff here is the resource extraction going on in the west, and a shocking proportion of the rest of the country is subsidised by this, so the gov are incredibly sensitive to any noise about jobs. The excessive regulatory environment of the telecoms sector is a side effect of this larger picture.

>I've always thought of Canada as a fairly responsive democracy.
Nope. We elect one group of corrupt assholes to run things, and then complain about them being corrupt assholes. We keep re-electing them till they do something so hugely scandalous that we elect the other group of corrupt assholes instead. And then keep re-electing them until they fuck up so badly that we switch back to electing the first group of corrupt assholes again. Our government is currently pushing for more restrictive drug laws, including for weed. They are pushing for mandatory minimum sentences for non violent crimes, and for privatized prisons. These things are almost universally opposed by Canadians. It simply doesn't matter, we re-elect them anyways.

> Our government is currently pushing for more restrictive drug laws, including for weed. They are pushing for mandatory minimum sentences for non violent crimes, and for privatized prisons.
WTF!?! Is it that US is trying out National Healthcare, so now Canada is going to try out some of the worst American ideas? I don't know much about Canadian politics, but the things you listed are some of the worst policies United States ever implemented. Why on earth is Canada trying those things, I can assure you, private prisons and mandatory minimum sentences are NOT working here in US.

No they don't. Only out east. I'm on Shaw. No caps.

Not sure what package you have, but Shaw has caps. http://www.shaw.ca/internet/plans/

Don't know why they put that on their website, maybe so people get used to it?? They suspended all caps.

I have BellAliant 80Mb/30Mb in NS. No cap.

You don't have an explicit one, but they have an "excessive use of bandwidth" clause in the contract, which is similar to the early days of Rogers Cable.

I've had Cogeco (180down/20up) in the suburbs and currently have Rogers (250down/20up) in downtown Toronto without caps. Rogers charges a bit extra for it, but no-cap plans are definitely available.
> Canada is definitely behind the US for internet access, it's even behind many (if not most) third world countries.
I completely disagree with this. Cogeco kicks the pants off of the options available in downtown SF in my experience.

I believe so - I have to use a relatively small player that resells Bell infrastructure.
The way it works is pick any two of speed, capped usage or reasonable price, but then forget the last one because it doesn't exist anywhere.
I consider my, still expensive, 4G mobile way better value than the land line ISP, but I need the capacity for work. Frankly I think 4G is going to wipe out land lines in far more cases than people realise.

>Frankly I think 4G is going to wipe out land lines in far more cases than people realise.
Even if it were consistently fast and reliable (which it is neither), at $40/month for half a gig of data? Not likely.

I'm on telus. It's 60 bucks a month for 6gb of data + phone/text ect.

Yes, you can talk them into a slightly less terrible deal than their advertised rates (which is $50 for 5gb, plus your talk/text/network fee costs). That is still absolutely horrible, and essentially useless as an alternative to actual internet access. Normal people hit their 100GB monthly caps on a regular basis just from stuff like game patches, buying new games, watching TV/movies, etc.

You don't have to wait for Google (or Bell, etc.)... http://www.cbc.ca/news/technology/small-alberta-town-gets-ma...

How is the residential Internet in the San Jose area cities listed? I'm in SF, and I can't believe how bad the Internet service is. It's worse and more expensive than what I had 4 years ago in my Midwest town of 100k people.

I live in Sunnyvale where the options are Comcast (who will happily sell a 105Mbit connection for $$$), or DSL from ATT or Sonic. I personally use Sonic since they are a great ISP.

Got my hopes up there for a little bit... if only they came to Santa Barbara -- we have plenty of tech people here willing to pay for awesome internet! (and get away from Cox...)
I'll keep dreaming :)

I wish they'll come to Santa Barbara so badly (As I'm sure everyone also does for their cities). We do have lots of tech here and even good connections to Google, but I feel we're probably too small. Even Verizon pulled out from installing FiOS.

At least Cox isn't too bad in comparison with the horror stories I've heard about dealing with Comcast or TWC... still, it is not fun to deal with them. Plus, 1 gbps for $70 sounds a lot more appetizing than 50 mbps for $74.

It is hard to wait for signups to begin in Austin.

The FAQ does not mention details about bandwidth other than "Internet that’s up to 100 times faster than basic broadband", but Wikipedia says the free service is 5 Mbps down (1 Mbps up) and the pay service is 1 Gbps down and up.
https://en.wikipedia.org/wiki/Google_Fiber#Technical_specifi...

Excited for Salt Lake City. We are growing our reddit office there a lot. It's a really exciting city to be in and this helps us a lot!

Really excited if they offer service between SLC and Provo

Reddit has an office in Salt Lake City?? Need any DevOps? :-P

It's redditgifts.com. Doesn't look like they're hiring for DevOps at the moment, but they are looking for devs (among other things): http://redditgifts.com/jobs/

Very excited to see Portland,OR on the list! I'm tired of Comcast and the other options are quite lacking in one way or another.

East Portland is wired up with Fibre backbone DSL from Century Link. Its $30 a month for a connection that's roughly 40 mbps down / 15 mbps up. I generally get 4 megabytes a sec downloads, and my friend who lives south around Holgate and 65th sees about the same. No data caps that I'm aware of either.
Its faster than Comcast's two lower tier internet packages. I've heard North Portland isn't as lucky but its certainly worth checking out.
Also various complexes around town are already offering FiOS.

I'd love to see Google buy up Frontier's FIOS lines. Their service isn't bad, but I get the feeling they don't really want to be in the fiber business and just got stuck with it in the Verizon purchase.

How far east? I'm east of 82nd, and the fastest CenturyLink offers here is 1.5 mbps.

So excited to to see Nashville on this list. Our technology scene has been thriving recently and this only adds fuel to the fire.

Nashville is a great city...I was totally unaware of the tech scene. Startups, big companies? I have no idea if I'll ever move away from the west coast, but Nashville is one of the few places in the eastern US that I could imagine living.

I'm interested to see where it would start to spread from (if it comes here at all), any thoughts?

Seriously? No Seattle?

Up until the last month or two, Gigabit Squared was supposed to be taking over the city's fiber, which may have been keeping Google away.

Joke's on us :(

No kidding. God damn.

Would love to see Google, or anyone, come shake up the Australian market. We are still beholden to low bandwidth, almost non existent upload and data usage caps.
We also have a national broadband network that most people will never see due to the politics played by the current government.

This +infinity.
The National Broadband Network is the one policy that had reasonably bi-partisan voter support, but the new conservative government has seen fit to abandon the admittedly more expensive, but more future-proof 'Fiber to the Premises' in exchange for what was going to be 'Fiber to the Node' but is now 'Multi-Technology Mix' or more appropriately 'Massive Telecommunications Mistake'...

Still no San Diego, but still happy to see this as the more Google Fiber expands within different regions the more the dinosaur ISPs in those regions will be forced to compete, hopefully rising boats even in areas not yet covered by Google Fiber.

Google needs to bring Google Fiber to Australia. The coalition have plans for a fiber/copper hybrid in which the fibre runs to a box and then you connect via outdated copper from the street to your house. It's like driving a sports car 90% of the way and a horse and cart the rest of the 10% — I think what Google are doing is great, they need to expand though. I know New Zealand could use something like Google Fiber as well.

Which is precisely why a re-nationalized telecommunications provider was such a terrible, terrible idea. Google cannot install fiber in Australia even if it wanted to, because competition to the NBN is illegal. It's now 7 years since it was announced, virtually nobody has it, and private companies like Google are precluded from filling in the gap, and the local telcos have just let their existing network go to ruin because they are being paid to rip it up. A shocking outcome.

While I won't argue that Fibre-to-the-cabinet (FTTC) is better than Fibre-to-the-premises (FTTP) it is, at the very least, much cheaper.
The cost per consumer for "last mile" connections can be an order of magnitude more than those used in the core network because of the economies of scale at play. [1]
Most of the UK is currently being set up with FTTC, representing a dramatic increase in the country's average broadband speed over the past few years. [2] [1] http://arstechnica.com/tech-policy/2010/03/fiber-its-not-all... [2] http://media.ofcom.org.uk/2013/03/14/average-uk-broadband-sp...

Is it cheaper? It depends on how long you expect your investment to be relevant for. If you accept that FTTH is going to be an ultimate 'end game', at least for the foreseeable future, then going FTTC is not cheaper. It may require less capital investment today than FTTH, but in 5 years, you're having to invest a similar amount all over again, in new equipment or by going FTTH.
FTTC is just delaying FTTH.

It seems one of the biggest costs of fiber is installation; i.e. trenching. Why is it not hung from poles? It must be much cheaper and quicker, surely?

Yes, whereas Europe puts everything underground, US operators will go cheap and string stuff across poles. My supplier of infrastructure management software was astonished when they saw at our data an realized we have almost no aerial spans in France.

Where poles exist, fiber is run on them. Where utilities are underground, fiber is underground.

When I saw the headline, I cringed. At this point, whenever I see a Google article titled, "The future of [google product]," I assume it's an announcement that the product is being phased out. Needless to say, I'm glad that's not what's happening here, but I'm still not holding my breath that Google Fiber is going to take over the world.

I wish Google would publish more technical information about their fiber project. I am quite interested in what equipment they have chosen, what types of cabling they are using, etc. (and why) Also, any other information that could be used by community internet providers looking to roll out fiber!

Always wondered why they didn't buy out Surewest or Consolidated Communications. Instant fiber subscriber base, right-of-ways, and complementary service areas. And the money found in Google's couch cushions would probably more than cover the costs. Easy way to add 100K subscribers...

Is mountain view count as San Jose?

Yep, according to the FAQ: "San Jose, Santa Clara, Sunnyvale, Mountain View, Palo Alto"

No Campbell. Guess I'm too sticks for Google Fiber. :\

LOL, I'm right on the border of SJ and Cupertino around De Anza and 85. I am so hoping I can get it just to dump Comcast.

San Jose is the nearest major city to Mt. View, Sunnyvale, Palo Alto & Santa Clara... they're using it as a Metro area grouping.
EDIT: Yes, these are part of the "The San Jose-Sunnyvale-Santa Clara, CA Metropolitan Statistical Area comprising Santa Clara and San Benito counties".

Awesome. They added Leawood to the list of cities. Now they need to complete some of the neighboring areas they have already started on and make some progress. Although I wouldn't complain if they just abandoned those and started on Leawood.

Google should install their fiber in the cities of corporate headquarters for at&t, time warner, quest, verizon, and cox.
Should have an interesting effect on the top-down mentality of price setting.
FCC also needs to rule ISPs as common carriers.

"announce the next round of cities who’ll be getting Google Fiber by the end of 2014."
- announce which cities are chosen by the end of 2014? or those cities that will have fiber in them by the end of 2014?

If you want to save money with Fiber, it's okay. But please use VPN with encryption or other traffic-encryption methods to no let Big Brother intercept all of your your data.

I live in Scottsdale and this is excellent news! Cox is the longstanding primary internet/cable provider here and their speeds are good but service is atrocious.

I live near DC... shit! I wanted Google Fiber just for the awesome TV. DirecTV is incredibly awful. Hopefully they can expand to even more cities in the future.

I don't understand, obviously there is a demand for this, so why aren't more companies rolling this stuff out? Give the people what they want!

"15/1 should be enough for anybody!" - something TWC would say.
Seriously though. Why are most of us still in the bandwidth dark ages?

Does 'do no evil' include canonical net-neutrality, or will it end up like Animal Farm: but already it was impossible to say which was which.

how do people feel about mobile networks? In a lot of locations the speeds are getting as good as or better than broadband. They're cost ineffective right now for anything but short term tethering but that problem could be solved in 2 - 4 years. I just wonder if laying down fiber house to house is going to seem outdated in 5 years or so for anyone except the hard core work at home dev.

How is the free plan going to work? Are they going to monetize it somehow by collecting data or serving ads?

It is free for "only" the first seven years.

I have a question does the cities Google picks have anything to do with their CDN locations?

So I guess I have a 1 in 34 chance of getting Google Fiber. Better than nothing...

Pfft. I'm waiting on Google Drone/Blimp; it solves the last-mile problem.

anything to get away from Comcast or Time Warner helps the greater population!

Not a single city in the upper-Midwest, Great Lakes region... :(

Great. Google should just push for it everywhere gradually.

Why San Antonio over Dallas/Houston?

As an Englishman, I feel left out :(

I currently live in NYC, and I've lived in Raleigh. It seems pretty clear that tearing up New York City to lay fiber is a much more expensive and tedious process than it would be in Raleigh. The state and local governments also seem much more friendly to that kind of thing.

Doesn't raleigh already have municipal fiber? I know there is fiber in in the RTP area.

Nope. There is municipal fiber in Wilson, NC which is about an hour away (greenlightnc.com). I'm not sure about RTP.
A crappy anti-muni broadband law was passed at the state level a couple of years ago (http://arstechnica.com/tech-policy/2011/03/cable-backed-anti...), so apparently the network has to be run by a private company?
There is joint initiative by local municipalities and universities to bring in a private company to run fiber (http://ncngn.net). It's not clear to me how this will be affected by the Google announcement. (They released a rather vague statement about it.)

Maybe I was thinking that they were going to get muni fiber but then they passed a law to not allow it.

I currently live in NYC, and I've lived in Raleigh
Howdy neighbor: I live in NYC too. A lot of NYC already does have fiber in the form of FiOS. I'll also observe that NYC's overall Internet situation appears to be much better than many places; right now I'm paying $50 a month for 50 Mbs down / 5 Mbs up through RCN. In Tucson, where I used to live, I was paying $70 for 12 down / 1 or 2 up.
Google is presumably targeting the places with the crappiest service and the most amenable local governments; NYC's service isn't terrible and its government isn't exactly known for being responsive.

RCN is a solid provider from what I hear. I'm JUST out of their coverage area and am consigned to Comcast. Tis a terrible fate.

I'm JUST out of their coverage area and am consigned to Comcast. Tis a terrible fate.
That hurts. We're looking for a new place to live and I have been asking landlords if their buildings get FiOS service. It doesn't seem to be an uncommon question, and the ones without it seem somewhat unhappy about the situation.

> right now I'm paying $50 a month for 50 Mbs down / 5 Mbs up through RCN
And I'm paying $54 for 20 Mbs/down on Time Warner, which is already a "discounted rate". (I'm in Manhattan).
The building next door to me has FiOS, but my landlord has no incentive to let them wire my building too.
I agree that New York City isn't the best target for Google Fiber, but the Internet situation here varies very heavily based on where in the city you live (not just the neighborhood, but which physical building and management company).

Hear hear. I've never been more frustrated about the ISP situation than when I lived in NYC. I have some friends that had such a bad experience with TWC (daily outages) that they voluntarily dropped down to ridiculously slow DSL, their only other option.

"The building next door to me has FiOS, but my landlord has no incentive to let them wire my building too."
Is your building rent controlled?
If not, I suspect your landlord is short sighted. I am curious to see whether property values will start to diverge based on quality of internet access. With more people wanting to work from home, I think availability, speed, cost, and quality of Internet access will have an impact on property values.
(Of course, you're in Manhattan, which means landlords can do almost anything and still charge insane rents, so this may not apply to you. Sorry.)

In New York City, buildings aren't rent-controlled (or rent-stabilized); individual apartments are. I highly doubt that any units in my building are rent-controlled (that's very rare in NYC these days). There may be one or two that's rent-stabilized, but given the neighborhood and the tenancy of the building, I doubt it.
> I think availability, speed, cost, and quality of Internet access will have an impact on property values.
In New York, that's vastly dwarfed by the many other factors that affect property values. As much as you and I think Internet access is important, most NYC tenants care more about location, bedbug history, sunlight, etc.

I'm on TWC and would love to get off it. How happy are you with RCN?

Fairly happy. We had the option of TWC or RCN and chose RCN based on reputation: TWC's is horrible, and anything I can do to avoid paying them money is probably a net Good Thing.

I live in Tucson now and I'm paying $70/month for 50 down / 10 up, but that's for residential. Commercial lines are horrible; $99/month for 10 down / 2 up.

So far Google isn't "tearing up" anything; they've been targeting cities with overhead power lines they can piggyback.

I guess that cancels out cities and suburbs that have underground power lines.....

NYC is already quite full of fiber. It's just not accessible to residential customers. Some of the world's major internet exchanges are in NYC, as are any number of datacenters. Most large office buildings in NYC should have a fiber connection from one of the commercial providers.

i just checked online, and it said it was available in my zip code. got super excited, and called Verizon. after 20 minutes on hold I asked them if "Fios is available in my apartment" after which I was asked 20 personal questions and told that there is some "good news" and i'd be transferred to a "Verizon specialist", who asked me 10 more personal questions. Finally i was given a great introductory price of $59.99.. great!! and then he informed that Fios is not available in my apartment, but they can still give me reliable fast internet which will cover all those things i told them i was going to use it for. I told him that it was really misleading how I got conned into answering all these questions all for a marketing pitch of a service i didn't ask for, and the guy said that he was "upfront with me about Fios not being available in my area" he just needed to spend some time to perform this check.. WTF?!
i feel so violated.. at some point i even asked how those questions are relevant to checking Fios availability, to which i was told "these are just part of standard questions i ask everyone"
if you call Verizon, refuse to answer any questions until they tell you if Fios is available at your address.

Nice try Google, I was really happy about Atlanta until I saw the cities listed.
Most of the cities they listed are in very sketchy areas that probably dont even know the difference between their regular connection and a faster fiber connection.
Avondale Estates - not so nice Brookhaven - will probably benefit College Park - crime area, could care less about fiber Decatur - some areas will benefit East Point - crime area Hapeville - crime area Sandy Springs - will benefit, lots of Apartment Complexes Smyrna - will benefit in some areas

Yeah, God forbid anyone should bring high-speed networking to poor areas. Fiber should be reserved for rich white people.

When they laid fiber down in KC they had each neighborhood sign up and based on how much interest each community had they would have precedence over the others.
If the cities they choose dont have an interest then what is the point?
Who said anything about white people?

You sound like one of those people in the suburbs stunting Atlanta's public transit because it would bring in more black people. The areas you are describing are not high crime areas, just poor.

No I want Marta just like I want my fiber. Accessible anywhere in and outside the perimeter.

Reply on news.ycombinator.com