Hacker News new | past | comments | ask | show | jobs | submit login
AT&T Pretends It, Too, Will Build A 1 Gigabit Network (techdirt.com)
224 points by mtgx on Apr 10, 2013 | hide | past | web | favorite | 144 comments

This reminds me a bit of the situation we had in Zurich. The (city owned) electric utility had a few hundred million reserves and decided to invest it into a fiber network.

Actually - since the utility was owned by the city - the decision was based on a popular referendum, which sailed through. Needless to say that the established telcos and cable providers fought tooth and nail to kill the referendum.

The interesting thing was that Swisscom (the largest telco and ex-federal monopolist) argued that fiber is not necessary and too costly.

After EWZ (the utility) started construction the private providers went into hyperventilation mode and started their own construction. Later the networks merged into one fiber network.

The network is managed, and I think wisely so, in a way that EWZ grants equal access to private ISPs which then compete for customers. They don't act as an ISP.

While in the first step construction was in the most dense areas of the city the network is now extended to the entire city. Extension was also accepted by popular referendum.

Would construction not have been started by a government owned utility (well, that's us the citizens - we don't necessarily see government as the enemy) we could be sure to wait an additional 10 years until private entities would have deemed it profitable enough to step in.

> The network is managed, and I think wisely so, in a way that EWZ grants equal access to private ISPs which then compete for customers. They don't act as an ISP.

This is the model I see as ideal and it always baffles me why it's so rarely mentioned as an option. Make the fiber network a public grid like roads, water, sewer, power. Then sell access to private companies to deal with the connectivity, services (TV/VoIP) and support. I'd say the technology is mature enough that we don't need competition to pick what kind of network works best anymore.

edit: Rephrased: I don't understand why the cities would give so much free stuff to Google for them to set up their fiber network instead of just... setting up their own fiber network that's then actually owned by the public.

That is simple. The city doesn't know what it's doing. They know Google does. They don't have to give any money to do it, just not charge Google to put their stuff in. Essentially they have to do nothing, get an ass-load of credit for doing it, and if it all goes wrong... well how can you blame them for trusting the company that some people literally think IS the internet?

> That is simple. The city doesn't know what it's doing. They know Google does

In either case they're "picking winners". They also "know" that Comcast, Time Warner, etc "know what they're doing", and we've all seen how that turned out. At least if they're the one commissioning it, they have any kind of power to fix things when they go wrong.

Even AT&T knows how to run a service that requires customer contact/billing as telecoms does. Google has zero experience in both building/running/maintaining physical telco plant.

And Google has terrible /nonexistant customer service Ernestine the operator would be a major improvement in customer service for google.

  Google has zero experience in both building/running/maintaining physical telco plant.
The network Google is deploying is nothing more than a bunch of glorified Ethernet network switches. I'm sure they have quite enough experience building/running/maintaining networks.. especially since it's well known that they get their own kit built specifically for them.

The Customer Service is an issue, but I think their innovative approach to CS could be one of the biggest benefits, throwing brains at problems as opposed to more clueless people answering phones seems like a good thing. Infact, on this note, analysing vast amounts of data for trends and problems is exactly the sort of thing Google is amazing at.. and most faults with internet service can be identified, and remotely resolved, in this very manner.

Customer service, from the telecom standpoint, isn't so much about troubleshooting as it is with customer relations and dealings.

- Setting up billing and payments

- Altering billing payments

- Adjusting for outages

- Scheduling repair vans

- Scheduling installation

- Answering clueless customer questions

Now I completely understand how Google may just defer much of those types of issues to a central website but I know plenty of people that refuse to use a service because of lacking telephone support. Google might be happy going that way but I wouldn't pretend it is good CS.

This may be AT&T's core business, but I can tell you that they do not do it expertly.

Thers a world of difference in doing a small network of say 3 buildings and 3k ports ie what you bog standard CCNA can do. And building infrastructure including the local loop is quite different to a few custom DC's which is where Googles experience lies.

Theres the whole "oles and poles" aspect and providing power to the cabs and the plant to name but two diferences.

and in Google datacentres they don't care of a blade or a rack drops off but in this case thats x subs that have just lost service.

Well that wasn't really what great-grand parent was saying. They were asking why not build your own.

Sure AT&T knows what they are doing. However they have a very clear track record of over charging and under delivering.

>I don't understand why the cities would give so much free stuff to Google for them to set up their fiber network instead of just... setting up their own fiber network that's then actually owned by the public.

Isn't Google providing the actual capital to pay workers in trucks to dig holes and string fiber? It looks to me like the "free stuff" falls almost entirely into the category of the expedited cutting of red tape.

You have to remember that there is a level of politics with all of this too. A municipal network may benefit the public, but who is going to pay lobbyists and lawyers to beat back AT&T's efforts to stop it? Whereas in the case of Google Fiber the 'who' is clearly Google.

Maybe I'm too cynical, but public utilities are often seen as the bad guys. Cries of "socialism" often deter anyone from championing a public good, even if it actually goes further toward creating a robust competitive economy.

Not a problem most of Europe has though. We love a good bit of socialism over here! ... Or most of us do anyway...

It's what's happening currently with the National Broadband Network (NBN) in Australia too. NBN Co (the government owned company building/running the network) is not an ISP but provides about 120 datacentres around the country where the retail service providers (the privately owned ISPs) bring their backhaul and hook into the network.

I agree that this is 100% the best way to do it - currently here, internet is delivered over a decaying copper network owned by privatised monopoly that is openly hostile to open access, who also competes in the retail space. So the new model should be so much better and encourage so much more competition.

We're just hoping that the opposition government (which are likely to come into power in September) will have enough pressure from voters and industry to not put in their awful policy which stops the fibre to the premises rollout and installs fibre to the node instead (costing almost as much, and offering a maximum of 50Mb/s instead of the 1Gb/s that the current plan will provide).

Fibre in the UK is rolled out by two providers: Virgin, which owns "all" of the cable network in the UK (there are one or two tiny little pockets left, I believe, but no other country wide providers), and BT OpenReach which is in a similar position: They're a subsidiary of the ex-monopolist that is required to offer access to anyone that wants to, and with caps on what they can charge. Currently they're mainly deploying fibre to the cabinet.

But the thing is, as much as I'm absolutely for a similar model to this, the technology is not mature enough that we know how to make it work best. BT OpenReach is struggling to provide the same speeds as Virgin. They claim to be able to offer up to 100Mbps, or 150Mbps in some regions, but my provider which packages BT OpenReach fibre access does not advertise more than 80Mbps and often advises to expect less (I was told to expect no more than 66Mbps based on distance from the cabinet because despite BT OpenReach's claims, they very rarely manage to meet it. Meanwhile Virgin reportedly have no problems reaching 100Mbps.

Without competition, there'd be little incentive for BT OpenReach to try to resolve the issues limiting their bandwidth. As it is, it matters so much that they keep misleadingly claiming they can deliver more (and many ISPs echo their claims).

I'd argue that while it is important to have a "baseline" provider that is required to offer 100% equal access, it is still important to provide wider access. There's really little reason not to require that every time a cable duct is put down to the local cabinet, that extra strands of fibre are laid down, for example, so other operators can rent dark fibre and install their own equipment. Or that cabinets be made big enough to accommodate some extra equipment for a small number of additional providers (and put similar access requirements on these providers, given that space in the cabinets will still be at a premium).

Today in the UK, other providers can put equipment in BT's exchanges, but apart from the densest city cores there's no practical way for them to provide connections to the end-user premises other than for Virgin (which is only in that situation because they merged UK's last two large cable providers who each had spent many years gobbling up smaller operators, often at bargain prices as competition got fierce, and had already written off most of the cost of building out the physical infrastructure).

It isn't necessary to have 10 providers all operating their own equipment and their own fibre, but there's little practical reason not to make space for at least 2-3 providers all the way to the local telco cabinets (even if not immediately), and allow the subscribers connection to be reconnected to the equipment of any of these based on the subscribers wishes.

Because in many states, like Texas, the incumbent carriers have lobbied for, and received, legislation that prevents municipalities from doing just that: http://www.baller.com/pdfs/BallerHerbstStateBarriers%287-1-1...

I don't understand why the cities would give so much free stuff to Google for them to set up their fiber network instead of just... setting up their own fiber network that's then actually owned by the public.

Because cities don't have the first clue about running a fiber network. As is the case with most of the other stuff they get involved in.

This is quite usual in (Eastern?) Europe for DSL networks (on landlines). And surprisingly very similary for railways. What used to be state-owned companies split into two - infrastructure owner and service provider. Anti-monopoly restrictions grand access to infrastructure to other companies as well and they compete for customers.

The city of Munich in Germany does something similar (though without the battles) with FTTB service provided by M-Net (regional provider, closely affiliated with local utility Stadtwerke München)

Yeah just got M-Net in January.

But we are blessed compared to the rest of the country.

Our genius federal government built the infrastructure first and then created a private monopoly who owns all of it. The worst of both worlds. The funny thing is they bought with all the money an US provider and are now the underdog over there.

We can blame Thatcher for introducing wide-spread privatization for this.

20+ years ago BT offered the then thatcher government FTC in return for a 20 year payback period - they where turned down as competition would magically build a competing network.

And of course vast amounts of capital was expended in building a cable network all the operators of which got into financial trouble and effectively got brought out of bankruptcy which is how virgin got its current network on the cheap.

And thank her for what the competition has brought us. One of the fastest and cheapest broadband Internet that I know of.

Just some facts of pre-privatisation: it typically took 6-12 months for bit to install a landline when it was state owned. Now: 0-4 weeks. Not to mention how much cheaper it is £4.99 for a broadband line?! It's insane!

Well, the price difference is largely because the investion cost of building the infrastructure is long since paid back, so all that has to be paid for now is a little maintenance and customer service.

I recall visiting my grandparents in Wales many years ago. Something that was astounding was that they paid every time you lifted the handset (a dial-tone access fee?). And then paid some more, depending on where you called. International calls required booking in advance.

We have basically the same model in Reykjavik, Iceland. High speed fiber managed by one company (owned by the energy company, which again is city owned), several ISPs compete for the customers.

Here in the UK, we have a peculiar paradox regarding the rollout of fibre. We have what amounts to a telecoms duopoly, split between the old telephone monopoly (BT) and the sole national cable company (Virgin Media). BT cover every home, but Virgin Media serve only the most populous areas. BT are progressively upgrading their POTS network, while Virgin's network was FTTC from the start.

BT only bother upgrading areas to fibre where they've got competition from Virgin Media, so people tend to either have the choice between two 100mbps fibre providers or none at all. The disparity in broadband speeds is growing rapidly, because the areas not served by Virgin Media tend to be those furthest from BT's exchanges and hence having the slowest ADSL speeds.

I'm not normally one to advocate government intervention, but the situation seems absolutely absurd.

What's absurd about it? It doesn't cost dramatically more to wire up a mile of fiber in a dense urban area (and in fact can be cheaper if you can re-use existing utility conduits), but can hit far more customers. So why shouldn't you expect fiber providers to only hit the densely populated areas?

And you can have government intervention, but much of the US's woes are the result of that intervention. US municipalities created all these monopolies in exchange for the guarantee that they would serve everyone and not just the most profitable people, and unsurprisingly those monopolies don't have much incentive to keep up with the latest in technology. But that's precisely the bargain all those municipalities made! There is no free lunch here.

It is absurd because if BT or AT&T wanted to upgrade certain areas to the new network to improve service to their consumers and gain additional profits out of it, they wouldn't choose cities where there already is a competitor with an equivalent fast network.

For example, if AT&T is convinced they are capable of providing a speedy 1 Gbps network, when Google announced they were going to Kansas City last year, AT&T could have decided to go to Austin and improve their service in the area.

However, this is not how it's playing out at all. Instead of choosing another densely populated area (only 2 cities in the US have this type of network, so there's plenty to choose from), they select (or signal) they will go only to where the competition is, and that is pretty absurd IMHO (not from a gaming theory perspective though).

Either way, Google's strategy in this case is flawless, and regardless of what other network providers do, it signals to the consumers that the technology is available and at reasonable costs. Therefore, it will start creating the push for network upgrades in cities where Google never intended to go in the first place at all, which is what this is all about. Having a fast and reliable network available to as many users as possible serves Google's interests (incidentally the consumers needs too), and that's what they are pushing for.

Regarding AT&T, they are signaling being willing to bring to the ground any business models relying on such upgrades, which as relevant as it could be to Comcast or Verizon, it's meaningless to Google as these investments are literally sunk money to defend their traditional business model.

In the same way they already defended in the past for other entrenched markets which could undermine their advertising business model, such as: Android (mobile market, avoiding iOS powerful position) or Chrome (internet browsers and default search provider settings).

The absurdity is the duplication of infrastructure. Half of the money being spent on providing fibre connections is effectively wasted, because the market incentives do not align with the priorities of society.

You present a false dichotomy between libertarianism and plutocracy. The US is exceptionally bad at market regulation by developed-world standards, largely because of how politics in the US is funded. The obvious counter-example is South Korea, where the government are providing billions of usd in funding to ensure universal service of high-speed broadband. That's no free lunch either, but the costs of a digital divide can't be ignored.

Broadband is an increasingly vital infrastructure, arguably as important as roads or mains water. The national economy is increasingly dependent on the ubiquitous availability of fast internet connections. We should take seriously the social responsibility to provide and maintain that infrastructure.

Yes, not to mention some rural areas which have no broadband at all and are still on 56k or satellite. In some cases locals have taken it into their own hands and established coops to build their own broadband systems.

Even if you have no interest in the internet, stuff like that has to hurt your house price.

And it should affect house prices. It's inefficient to have people living in the middle of goddamn nowhere, and it makes no sense for the government to constantly subsidize doing so.

Um, middle of nowhere is where food is grown and livestock are raised. It'd be disastrous to not have people living there doing these things. As more services move online, it is a public good to ensure all constituencies have access.

'Middle of nowhere' is sometimes 2 miles from a town (with cable) or a mile from a regional airport.

We finally got 3MB DSL last year.

This is why Google Fiber is (probably) not going to have that big of an impact as people hope it will.

Internet connectivity is geographically bounded, and to be disruptive, you need to cover most areas. This is quite unlike other disruptive technology that e.g. Apple have produced, which isn't constrained by geography.

If Google Fiber never comes to your state/county/city, you'll have just as crappy and expensive service as you had before - the existing providers will try to squeeze out as much as possible - which they can continue to do until there's competition in that geographic area

The upside of living in a large city is the plethora of available competition for almost everything. That has always be the case before the internet was even born.

The government targets for rural areas is a joke, and damaging to the economy. Not helped by 4G rollout 'issues'.

We have a microwave link on the church tower to cover the village here. It really shouldn't be left up to the locals to manage their own network.

Why not? You get a better service and its likely going to be cheaper. I wouldn't see any advantage of having companies like AT&T neglect such areas. Get the real competition going.

A lot of these services can be pretty laggy and at times unreliable. Good broadband ultimately needs fiber which requires capital and people who can dig up streets etc.

Most telcos will work on the basis of recouping this after X number of years, but can only do so because they have deep pockets. Putting fiber into west london is a no brainer, putting fiber into a village with ~300 residents , half of whom are over 65; not so much.

>putting fiber into a village with ~300 residents , half of whom are over 65

Actually, that's probably a no brainer as well. A village of 300 can't cost _that_ much, and improving network connectivity to a small town means that town becomes more attractive to the people who will be inheriting and buying the homes once the 65-year olds shuffle off this mortal coil.

Also, don't knock seniors. My 77-year old father blogs daily and my 71-year old mother spends a lot of her time playing with her new iPad mini. They live in a small town with new fiber.

It can cost a metric fuckton of money. When I worked up in scotland for BT the press officer for Scotland told me a story about how BT had miscalculated the cost of providing service to a very remote location and ended up having to pay around £200,000 to provide service to one property.

Actually, that's not a bad price, considering that they get to use the outlier every time they need to justify not providing service to remote customers.

Well it was over the max allowed for providing service - normally they hit the government up for subsidy to provide service to the highlands and islands.

I don't know about the rest of the world, but often in the US the choice is between laggy, unreliable local services or no service at all. Much too often the company that is supposed to service rural areas can't be bothered because it isn't economically viable. Although, it doesn't stop them from suing the locals because it's an unfair economic advantage if the locals decide to build their own.

My Grandmother (75) is the member of my family with the highest speed connection.

If national governments weren't thoroughly corrupted by the vested telecom interests, they would fund local-owned efforts to set their own baselines, allowing profit-driven enterprises to address sub-areas that are profitable within that area to deliver even better access.

The problem is that it's the opposite (at least in the USA) - state and national governments are lobbied by the telecom/cable industry to pass laws prohibiting local initiatives, so local governments have to give massive subsidies or concessions in order to get anything.

In the UK it's normally a few local individuals bonding ADSL lines together at point x where the service is an acceptable speed. A few microwave links later and it's ready to be consumed by 10-100 people. It's not real competition, it's not cheap, and it's not well support.

It's a poor replacement for fibre to the cabinet.

I'm currently trying to buy a house in UK, I check each potential purchase for distance to cabinet and FTCC on the cabinet. In most cases not having a nearby cabinet (<2KM) has been the deciding factor in not putting an offer in on that house (typically in this area that means a speed of ~1MBps).

That is exactly what is happening where my parents live (in upstate New York.) The local telephone company now has fiber throughout the village and a WIFI network for customers that's accessible from most places in the village.

Their base-line internet service is roughly five times faster than my base-line internet service (from AT&T) in Chicago.

In the UK in particular both broadband AND cellular/mobile service should be nationalised.

Both businesses and consumers NEED good quality infrastructure at an affordable price. Essentially the country's economic growth directly depends on it.

To accomplish that the government should control all fibre, all phone lines, and all cell towers. It should then re-sell access to these things to ISPs/telephone companies.

With this in place you might still buy your cable from Virgin and your mobile phone service from Orange, but Virgin and Orange would be renting space on the government's infrastructure to accomplish it.

Similar to the Network Rail model but less terrible.

That's a terrible idea. You probably can't remember how bad the telephone services were when they were nationalised. Having to wait 6-12 months for a line was the norm. Prices were high.

People seems to forget just how badly state run services typically are, even compared to AT&T and the like..

My father was a surveyor in the city of London who worked on new office construction. He used to tell me how back in the days of nationalised phone service the first thing you did when you were building a new office building, before you'd even laid the foundations, was to order the phone lines, otherwise they wouldn't be ready in time for opening the building

Doesn't that makes sense anyway? It isn't like have to get your phone lines after anything else.

No, it does not. In most cases it's not as if they had to lay down new copper from a central office to the new building or would have to run copper throughout the building. If the thing was built on top of existing infrastructure then it "should" not be a big deal to hook the building into said infrastructure. Of course, I say this without knowing the quality of the infrastructure in question.

infrastructure provided by the government which sells it to ISPs that hunt for customers. That models work pretty well already in many cities and small countries.

The Network Rail is what we've got already - BT Openreach provides the infrastructure, which is resold by myriad companies who have access on equal terms. The problem is that Openreach has no universal service commitment. Their customers are the ISPs, not end-users.

We've had nationalisation and it was a complete shambles. BT had no interest whatsoever in providing Broadband until Telewest came along. You couldn't even buy your own phone or answering machine until 1982 - before then, only a BT engineer was allowed to connect equipment to their network.

The straightforward solution is simply to offer a grant or tax credit to any company that provides the first high-speed connection to a house. Progressively increase the subsidy over a number of months or years to create a Dutch auction between BT, Virgin and whoever else throws their hat in the ring. This would progressively push up the LTV of customers relative to the cost of acquisition, creating the maximum number of connections at the minimum cost to the treasury.

Another model is to separate the entities that own and sell the infrastructure from the entities who own and sell the service. It's all still privatized but there is no monopoly.

It's a good idea in theory.

In practice a conservative government will end up selling off and privatizing the company. Subsequent governments will still act like they own it though, because that one company now controls 99% of the country's infrastructure. The share price will then stagnate forever.

Kinda damned if you do and damned if you don't.

Can a conservative government also privatize public roads and pathways? I mean, nowadays it can be argued that broadband/communication is as important as the transportation network. It's at the core of our day to day communications. Although if public roads can be sold to be managed by private companies, then we're screwed in both situations. But I'd rather have my tax money going to pay for infrastructure and innovation than to bailout this or that bank.

There are privatized "public" roads in the US. They are called "tollways"

Even the SWP and Keith Have given up on that one :-) and network rail has worked so well hasn't it NOT!

To be fair, the current telephone network, including fibre is owned by BT Openreach. Under agreement with ofcom they have to provide equal and fair access to their network to any company who wants it.

That is totally not true. I know of several people, myself included, who now have access to BT infinity and never have and still don't have access to virgin media.

Well, in theory you can make just as good a case for the government providing internet access as the government providing roads. In practice I see two major problems, though. First, norms against the government denying the use of the roads to people are very strong, but there aren't really the same norms against the government refusing to provide the internet to people considered bad. The upgrade cycle for roads is also much, much slower than for internet connections.

It's a lot easier to injure or kill someone with a car on a road than on the internet, though. I hope nothing like an internet license happens.

As a counterpoint based purely in practical terms, there are nonetheless almost a dozen companies competing to actually take my money in exchange for a service, many of which are far faster and far cheaper (and without any data limit) than typical offerings in the US.

True, I have no need for fast speeds; 20 Mbs is more than enough for me.

BT only bother upgrading areas to fibre where they've got competition from Virgin Media

I live in a small town in a very rural part of the country and there's no hope of cable ever reaching us.. but we got Infinity 2 FTTC last year FWIW. I think the council threw grant money at BT though ;-)

The best kind of government intervention in this case is to eliminiate the duopoly and open the market to more competition. The more competition the better the service and prices will be.

It's not a legal duopoly, innit. It's just a business with extremely high barriers to entry given all the infrastructure.

The only tools at your disposal is some kind of subsidy or legislate them into eating the costs of laying down infrastructure in less profitable areas.

This is the really important takeaway of this piece:

"AT&T just admitted that they'll offer better service if there's real competition, so how do we make sure there's real competition?"

AT&T and the others have been claiming for years that they are doing the best they can and raking in huge profits; this kind of abrupt turnaround just shows what a lie that was.

As I recall, the telcos already got their chance to hook everyone up with fiber with super favorable terms from governments and huge tax breaks in the 90's. They made the promise, took the money and ran.


Google's turn.

that ain't right. let's ask for our money back..

TFA talks about favorable deals for telcos. Telcos have received one-time incentives to do things that they may or may not have done, depending on who you read. The recurring incentive is usually to keep phone service working in rural areas.

What Google got (at least in KC) that AT&T wants is no franchies fee (typically cable companies and fiber providers pay the town a percentage of their revenue for the right run lines), no pole-attach fees, and cheap/easy permitting.

The impressive thing that Google has accomplished is not delivering Gig to the home. It is convincing cities, in a budget crisis economy, that the mere presence of gig is important enough to not only give up a traditional revenue stream (franchise fee) but also to give up cost recovery charges like pole-attach and permit fees. Apparently they still have a 5% fee in Olathe, but I'd bet they have other concessions that make up for it.

This: http://techliberation.com/2012/08/07/what-google-fiber-says-... and this:

I don't know, I find it impressive that Google is doing this without begging for money, then canceling the project after doing no work on it because of hand wavy "expenses", like AT&T and their crowd have traditionally done with putting fiber everywhere.

All marketing lies.

Make any company seeking such guarantees put up a $1 Billion bond that will be forfeit if they don't complete in a few years and watch AT&T back down.

And for pete's sake stop giving tax deferments on the FRONT END. If a company claims they are going to make jobs or supply a service do not give them the tax break until that actually happens.

All great ideas, except for lobbyists. In reality, it's probably not reasonable to force someone to put up a bond to build a network. And a billion dollars is probably an order of magnitude (or two) higher than what it'll cost AT&T to completely build out Austin.

I believe google recently said they could build out most of the major population centers in the US for 11 Billion, so that gives you some context for a single city out of the top 100 metros.

Billion is just an example of something meaningful, and they'd get it back upon completion. If not, the amount should be used to complete the network for them and then the government would own the network or auction it off.

So the amount should be what it costs to build the network for the location.

Is it just European bias or is the US largely near a 3rd world country in terms of internet broadband access?

Bias, sort of.

The US is actually two separate but interlinked networks.

On the one hand, residential service is 3rd world in most areas. On the other hand, I would argue that the US has the only first world broadband economy, which is our commercial network. The real networks in the US are very fast, 100gigabit fast at the core, but they slow down when you get to residences. Between Datacenters, the Internet is really amazing.

On the resi side, there was a Bush memo that reclassified resi service as "information services" instead of "Telecommunications services"[1] so they gave up the ability to enforce common carriage which is the term for sharing the last mile access.

In short, that is European bias when talking about our business networks, but quite right when referring to our residential networks.

It's a lot more complicated than this, but that's the gist of US broadband. You can't nationalize it because that's socialism and you can't force companies to build networks because that's fascist, so the telcos only build where there's competition: the datacenter.


Edited to add link.

When I read elsewhere "We saw a lot of 100 GbE deployments primarily in core networks over the past couple of years, and now 100 GbE peering is taking off too. Several IXPs around the world, most of whom are Brocade customers, have announced the availability of 100 GbE peering ports or the intent offer them this year: AMS-IX (Amsterdam), DE-CIX (Frankfurt), JPIX (Tokyo), JPNAP (Tokyo), LINX (London), Netnod (Stockholm) and NIX.CZ (Prague)."

I can't help but think that European data centers have core internet of 100Gb as well. Following up on that Stockholm company, http://www.netnod.se/bahnhof-becomes-first-netnod-100-gbps-c... says "Netnod is one out of only four Internet Exchanges in the world that offers 100 GE services." Elsewhere, euNetworks Group Ltd says "The new 100G capability is available between the key cities of Frankfurt, London, Amsterdam and Paris, and also between Dusseldorf, Hamburg, Dublin, and Manchester. " (These are from last year.)

It's therefore hard for me to believe "that the US has the only first world broadband economy" given that many other non-US cities have 100 Gbit service.

I also think it's a bit disingenuous to split a country into its first-world and third-worlds parts. A third-world country can have parts which are very first-world. I think an aspect of being first-world is that private people also have access to good services. In Stockholm, for example, 100 Mbit residential fiber is about $60/month.

In the US, it is virtually impossible to find a circuit of 10 mbits up and 10 mbits down for a residence. This was the original goal stated by the Clinton initiatives but this was never reached.

Contrast that with Datacenters where 10gigabit links are common. Sure Europe has 100gigabit core networks, but the US, in terms of aggregate capacity, dwarfs Europe.

But the point is about the difference between our residential and our commercial networks, which are completely separate from a regulatory perspective. We treat residential bandwidth like its cannon fodder for telcos, which is decidedly not first-world.

You make great points but I still think mine are valid.

I don't think its bias. I cannot get DSL, CABLE, Fiber-anything... My choices are: Dialup, Satellite, or wireless internet.

I can get 3G cell coverage -- sometimes.

My street does have cable television, they just refuse to service the area with internet access. My country has an exclusivity agreement with this cable provider, so no other providers can come into the area.

The sickening factor for me is the fact that I live practically on the county line. I can SEE houses with time warner service from the end of my driveway. They just can't go that extra half mile.

Crazy. So, basically, I end up paying $50/mo for 1down/.3 up. That's with lots of packet loss, too.

We lived with that situation for years. Finally we banded together to start a WISP, signed a few people, and put up some antennas (with the blessing of local government) We then started passing out flyers.

The flyers did it. The a cable company had everyone hooked up in less than 6 weeks. They had previously told the county that geography made this impossible while reupping their exclusivity contract.

That's not a bad idea. I'm going to look into this. Thanks!

Great point about the flyers!

> My county has an exclusivity agreement with this cable provider, so no other providers can come into the area

I see this over and over, and I don't understand WHY a local government would ever do this. What do they get out of these exclusivity deals?

Meanwhile in my old town in Sweden, the city-owned power company built a supplier-neutral fiber network to promote competition (and a price war).

Grant County, in Washington state, did the same thing. The Public Utility District (the power company) built a county-wide fiber network; local ISPs have sprung up to sell access. My parents use a microwave link to the top of a nearby mountain range, where a tower ties them into the fiber net. Speeds aren't great, but when the other choices are satellite or dialup, it's pretty good.

It probably helped that Grant County has several hydro dams in it and sells power all over the place... I assume the PUD was pretty well-funded in this effort.

> I see this over and over, and I don't understand WHY a local government would ever do this. What do they get out of these exclusivity deals?

Short-term available money, e.g. to fund urgent renovation, pay back overdue loans or fund federally mandated services. The latter is popular in Germany, the federal government forces availability of services like child daycare paid by regional authorities, but the federal government doesn't have to foot the bill.

This is how you do internet. Public wire, competitive gateways into the backbone run by private competitors.

Hell, the majority of US phone / cable wire was already paid by taxpayers over the course of a hundred years, local governments just did short term gains for long term screwing by selling them off to big telecos.

> I can SEE houses with time warner service from the end of my driveway. They just can't go that extra half mile.

Is there a way to create some local network with people who have access in that half mile? Then you can share the bandwidth accessible from their location.

My father works with the guy (or did before he retired) and he's cool with us putting up a tower if we want. Unfortunately, line of site is iffy due to the tree coverage on that end of the property.

There is a company here that does fiber to place of business for either corporations or schools. Unfortunately, it will cost me about $5000 to get everything set up (having the cable run, etc.) and about $120/mo for 10/10.

There's another fiber company in the next county over that says they plan on expanding into the area later this year. I don't know how serious they are, but they offer 50/50, 100/100, and are working on building out gigabit service.

That would be nice.

Have you considered moving?

I have. I popped back into into the middle of nowhere to help take care of some family things. Fortunately and unfortunately, I landed a programming gig while here with a great company. The pay is not market rate but I love the company and what they do (think non-profit free netflix for deaf/blind students with educational materials.)

I'll end up heading back to California eventually.

No, that would be us here in Canada :) Especially on pricing and caps it feels that way sometimes.

Internet prices in Canada are awful... Videotron is charging me 50$/mo for 10 down / 1 up and a supplement of 30$ for unlimited download... and the other provider in Montreal (Bell) has exactly the same pricing scheme... we have a lot of choices.

Start an ISP. You'll make a lot of money.

You're kidding right? The issue is not a lack of entrepreneurs.

Forgive me for my ignorance, but if rural telcos are building their own networks, what is to stop you from also building your own Network?

Am I missing something?

what rural telco is building their own network?

in canada, the taxpayers built the network, then a corrupt government signed it over to a private cartel so they could charge rent.

it sounds like you're making some kind of Ayn Rand free market allusion. free markets dont apply when Dagne Taggart owns the government and closes all barriers to entry so that her railroads will never have meaningful competition.

You are talking about terrestrial networks; that's not what WISPs are building. In a number of developing countries, the ground networks are owned and regulated, but wireless operators are not regulated in the same way.

In San Francisco we have a TON of small wireless ISPs fulfilling the needs of the companies that do business here, without using the last mile telco access.

I'm not living in a dystopian Ayn Rand reality; I'm suggesting an alternative method of delivery, one which is not subject to the same regulatory hurdles.

Of course you can't go and jack into BT's network and undercut them, but you can probably build a WISP without running afoul of the law.

Please correct me if I'm mistaken :).

You may be right about WISPs.

I was referring to cellular networks and wired networks.

I'm guessing the hurdle is still legal/political. I'm gonna look into this, thanks for the insight

No worries. Check out the meshpotato. If you're really gonna do it, this is the key to service delivery.


Basically it's an Analog Telephone Adapter and a Fixed Wireless receiver built into one, so for a very small price you can deliver bandwidth+telephony to your users.

Cellular and Wired are completely fubarr'd, that's why Africa bypassed them.

I have Verizon DSL with 3mbps up / 1mbps down a hundred miles north of Philly. Local cable co would be an option, but my mother watches a lot of netflix and I download youtube videos by night (because I can't concurrently watch anything across 2 screens) and the cable company due to their monopoly puts up 50 gig caps. And that is the $60 a month 15/1 plan.

I used 195 GB in January, for example.

So I'm stuck on Verizon. They have jacked the rates up from $25 in 2004 to $40 today for the exact same (nonexistent) service. No other options in the area. At least they don't put bandwidth caps on their DSL service. Yet.

If that happens, I'm moving to Kansas City or Austin. I'll take a plane and go door to door at startups or tech companies to find work from a hotel room. My college had crap internet too, but back home it is so bad it is like the dark ages.

I'm getting 81 Mbps download speeds right now on my laptop and have 4G on my phone wherever I go, not really near a 3rd world country as far as I know.

How? Are you in Google Kansas?

Comcast has had 100mbps service available in some areas for a while now.

Among many other ISPs. If you live in any of the major metropolitan areas it's the norm.

What major metropolitan areas? Would San Francisco or Los Angeles count?

I noticed that there have been a lot of comments lately saying "Interesting, the US is nearly a 3rd world country in terms of measure X", where X is income distribution, incarceration rates, civil rights, certain diseases/causes of death, etc.

US is very much behind in terms of quality and adequate pricing for Internet access. It's mostly the result of sick monopolies developed around it.

No, yes.

I am so glad Google is putting AT&T (and the other broadband providers) in the hot seat. Sadly I don't think San Francisco will get Google Fiber anytime soon. Why is it we don't even have Verizon FIOS?!

Because this is what happens when you try to do stuff like that in the Bay Area:


Sonic.net proposed gigabit in SF with 170 above ground repeaters compared to AT&Ts 800+. That also was voted down.

It's not a sick joke, it's San Franciscan :/. I sometimes feel like we work in a city that's all things at once: tech bastion, hipster haven, and a home to a generation of hippies. There's a lot of culture and there's a part of it that hates ugly things and a part of it that hates technology and every time someone wants to build some cool tech like city wide wifi, this is the inevitable result.

The problem is that we grant each and every individual de facto filibuster power over any construction project. CEQA is so broad and vague that all you need to do is come up with some crazy, tenuous claim that the environment will be impacted in some way, and even if your claim has no merit and you can get no fellow citizens to support you, you still get to obstruct the development process for years.

Um you know that city wide wifi woudl require an insane amount of AP's and backhaul - hence why wimax doesnt seem to be going any where.

Insane is relative, backhaul is also relative.

I don't think WiMax was meant to solve the problems to which people have applied it. If you look at WiMax actual range, it's quite good, but the advertised range is at least double (hence all of the big ClearWire lawsuits).

I don't think setting up city wide wifi is unreasonable with present technology. I think it's an entirely attainable goal, the things that hold it back are bureaucracy and the same individuals who hold back things like project lightspeed.

But my broad point in response to your comment is that the benefit of saturating san francisco with AP's are manifold, whereas the costs and risks are minimal.

Sf has a population of over 6 and a half million and covers 600 square Km. I woudl love to see your estimate for how much that woudl cost to build and maintain.

Less than 100M to implement and under 10M to maintain per year.

I definitely don't think it's expensive compared to the benefits. Personally, if I'm pipe dreaming, I'd rather have ubiquitous 1gigabit fiber, which google pegged at 11Billion for almost all of the US to be covered.

You have to ask why they required those freestanding boxes, instead of negotiate access to basement utility areas in buildings across the network. Boxes like that may be necessary in less built up areas, but in densely built areas they're a cost-cutting measure.

Maybe it wouldn't be profitable if they had to deal with extortionate landlords, but on the other hand it'd seem likely you'd find a sufficient number of building owners that'd be happy to allow space for cheap or nothing in order to ensure great access.

They need 24-hour immediate access to any box. That means it has to be somewhere accessible. And they can't bury them, because they need ventilation, and the ventilation unit would have to go above ground, and would be even bigger than the utility box was.

The NIMBYs are pretty terrible. The best part is how you won't see a single solitary wind turbine in the "bay area" proper, but as soon as you start heading up Altamont on I-580, BAM everywhere. It's ok though, it's just the central valley out there, hicks deserve it.

This is just a sick joke, I hope?

These fools should be banned off into an Amish-like community.

Because the FCC ruled that wired networks are subject to network neutrality rules, and that wireless networks aren't.

Verizon has every incentive to move efforts to broadband wireless.

Some apartment buildings (Including mine) have webpass which is 100mbps up and down (and in some locations 200mbps) for $50 per month. Other than that internet in SF is surprisingly bad

Or, in my case, 10Mb down and 55Mb up. It's like reverse Comcast.

When I saw the original press release earlier today, I thought the company name was a typo.

It's clear now that AT&T is making a political statement instead of announcing a premeditated expansion plan. Regardless, this seems like one of the worst possible ways to draw attention to purported "special incentives" that the city of Austin is granting Google. There is a large chance that AT&T won't come through on the plan at all--and when all this has blown over, AT&T is much more likely to be remembered for reneging on their plan than for being a proponent of fair regulatory laws.

> AT&T's expanded fiber plans in Austin anticipate it will be granted the same terms and conditions as Google on issues such as geographic scope of offerings, rights of way, permitting, state licenses and any investment incentives.

Sounds more like AT&T is expecting to make a deal with Google to resell bandwidth off Google's own fiber after Google puts up all the initial expense to lay it. That would indeed give them "exactly the same terms and conditions as Google."

Isn't Google laying minimal fiber infrastructure because they bought a bunch of existing dark fiber years back?

Hey, if they want to do that and Google will co-operate, sounds fine & dandy to me.

If Google slyly becomes America's fibre wholesaler like that, I will tip my hat to them.

> Municipalities have been giving AT&T and other incumbents incredibly favorable deals for years, and AT&T has tended to return the favor by providing the bare minimum in quality of service to its broadband customers, while focusing most of its efforts on trying to block any hint of competition from showing up.

I"ve heard this a lot - I'd love to see some objective background information on this.

There are many news stories about such companies suing local municipalities to prevent them from building their own network because the companies refuse to service the area properly. Just pick whichever news source is the most objective to you.

> "this press release confirms exactly the message that AT&T has been trying to deny for years: that when there's real competition, then AT&T will invest in making a better service."

In this cyber week in Congress, I wish our loyal representatives would pay more attention to this instead of focusing on taking away our rights or ramming another CISPA/SOPA down our throats.

This whole thing is really sad. Fiber is such a long-term investment in a city's infrastructure that it should be a public utility. Let all the ISPs compete to sell service over the public fiber infrastructure, don't let the infrastucture rollout languish in the hands of people who are only looking for short-term profits.

I love the disclaimer at the bottom:

"Information set forth in this press release contains financial estimates and other forward-looking statements that are subject to risks and uncertainties, and actual results might differ materially. "

It might worth it to AT&T to actually build a 1 gigabit network to dissuade Google from encroaching on their territory in other cities. As the only gigabit player in a location, Google can charge users much higher rates than if they're one of two providers in the market. If AT&T sends the message that they're going to be in a two-player market in any location where AT&T is the incumbent broadband provider, Google might go after other markets, first.

Of course, that assumes that Google is rolling out fiber to the most profitable cities first, which might not be a valid assumption.

For those fortunate Austinites who will face the difficult decision of choosing between two gigabit fiber ISPs: AT&T is a founding member of the Copyright Alert System scheme [1][2].

[1] https://en.wikipedia.org/wiki/Copyright_Alert_System#History

[2] http://www.copyrightinformation.org/about-cci/

Even if they do end up following through on this, knowing AT&T it will be subject to the same 150 GB/month cap they have on U-Verse.

Let me just download this stuff at 1Gbps... annnnnnd we're out of internet.

What's incredibly unfair is that Austin will now have 2 Gigabits of internet while the rest of the world is bottlenecked at megabits.

That's a poor poker face, AT&T.

Once again it approves competition is good. ATT has the chance to do it years ago, sigh.

Makes sense-- pretending to provide service is AT&T's core competency.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact