How AT&T Is Planning to Rob Americans of an Open Public Telco Network
wired.com
wired.com
You can't simultaneously decry America's networking innovation and complain that AT&T is moving to IP. Everything is moving to IP. Circuit switching is dead. They're even putting TDM over IP nowadays(ITU Y.1588v2 and SyncE).
If you really want to improve the situation don't act all Luddite. Start the conversation on overhauling common carrier legislation. Bring Judge Greene back from the dead and break these carriers into thousands of competing tiny operators.
Also, I'm not sure many folks are arguing for total regulation -- instead, like smutticus, I think most people would like to see a more competitive market than the one we have now.
By comparing to the results in other countries that didn't deregulate common carriers, which have the same innovations at lower prices and with fewer restrictions on what you can do with your devices.
What you see today is a pale reflection of what might've been. You're happy you have connectivity, and that's great, but your connectivity sucks and you shouldn't be happy about it because you paid for a better network and didn't get it.
Does that make sense? AT&T and the other big telcos stole the dream of ubiquitous fast Internet by mortgaging our future for continued copper obsolescence. And we wonder why there's no money for metro fiber networks...
The first thing they did, afterward, was to have their lobbyists pressure the state legislature to let them out of their half of the deal (pass laws to that effect) -- while keeping the tax breaks and incentives, of course.
Personally, I had DSL through them, and it is the worst... "utility" service I've ever had. Three appointments before the installer showed up, who was an independent subcontractor who left bare wires twisted together on the building exterior.
Their network was a nightmare, requiring 10+ hops all over the place just to exit onto a trunk. And it would die every week or two. One service call handler who was at her wits end admitted she was subcontracted and could only file tickets. She couldn't even look at a filed ticket to see its status. Zero feedback to the customer.
And... it took on average 30 minutes to reach a service rep. By contrast, call up to order service, and your call was answered within 30 seconds. The only part of their service they appeared to spend money on: Revenue intake.
They only started getting better, and upgrading their offers, when Comcast started rolling in the same territory and eating their lunch. (Not that I'm overly lauding Comcast, these days.)
If you (grandparent comment) think the big telco's are doing you any favors... well, I disagree.
That would be one option. A more pragmatic option might be to just have the FCC classify last mile IP telecommunications service as a telecommunications service that would be subject to mild regulation instead of an information service that doesn't even have to meet common carrier requirements.
Genachowski (and whoever his replacement happens to be) will smile and do what Randall Stephenson tells him to do. McDowell will threaten to sue himself and the FCC, saying the FCC has no right to do anything, under any circumstances and any action is a socialist affront to God and Country. Clyburn will do her "what does this do for the black community" bit and if we miss the saving throw a few red necks will get pissed off, further derailing things. Unsure of Rosenworcel and Pai, hopefully one of them will say something useful, but that is only praying for Judy and Punch.
Meanwhile both representatives standing in front of the chair persons will really be lobbyists for ATT. One will say that any action suggested will never be enough and the other will say that any action will be too much. Blah, blah, blah, some things never change.
I'd genuinely love to see a conversation (on anything), but that conversation will never really happen.
The requirements for running the telephone system do not end just because the transport medium is IP. In many cases, the transition to IP services makes the existing IP infrastructure beholden to higher uptime and connectivity requirements.
Try telling federal regulators that you did not deliver an emergency call because you were in a peering dispute. Today, American backbone IP NSPs are required to deliver outage information to the FCC, and this is prior to any assumption of life-critical services.
It's costly, it's an awful use of resources, for no good benefit.
POTS is a huge ruby-goldberg device, it's almost laughable sometimes. All that so that the terminal can use 19 century tech (yep, not 20th century, 19th century)
Regulation shouldn't be about the technology details, it should be about the service provided.
Oh by the way, the US Postal Service is being driven to bankrupcy by regulation as well (I'm not saying all is bad though)
More correctly they are being driven to bankruptcy due to ballooning pension payments (and a congressional mandate to actually fund those obligations).
Most libertarians dont get economics very well, they just think taxes and government are evil. Capital is not something you can save in a silo or go negative. It just flows to where it's most needed at any given time; you see it has to be spent, saving fiat currency is just a way of moving it (if you save, someone ELSE must go into debt at the same time).
Now what does economy of our kids look like? It solely depends on what they produce and consume! But that also depends on how we spent the money in the past invest in the future. Education, infrastructure, R&D, ... Didn't invest? Well your kids will suffer. Oh, china had money burning a whole in its pocket and you decided to let them inflate rather than take the money at a negative interest rate? WTF are you an idiot?
Now what are pensions? They are simply investments for the future! But you have to pace those investments and time them just right, so you don't kill yourself now and can meet obligations in the future.
The real problem that will hit us in the ass: most of us aren't covered by pensions at all and we are finding out that 401k's aren't a great alternative. That will be a bigger burden for your kids (assuming they don't abondon you) than any kind of debt they have to pay off for us (really, they are probably just paying you anyways).
[1] http://law.justia.com/constitution/us/article-1/39-post-offi...
Which all goes to show that if the legislature wants to create a public good they should have the guts to raise taxes to pay for it, rather than trying to be clever and accomplish their aims through the back door.
If USPS were allowed to make their own decisions, they wouldn't necessarily be in this mess.
Not sure if that was intentional or a mistype but it made me laugh:)
Higher Prices: This is not a deregulation, surely it would make AT&T have to compete with other VOIP services? The issue are the necessary safety/security services and hiw these are not well governed in the TCP/IP world?
Service Disruptions: I can't see how both AT&T and smaller operators both using public IP would disrupt? I think there's a trade off between between price and quality of service, those that go with a Cadillac service vs those that know they are getting a slower service albeit at a lower rate?
Inequality and discrimination: This on eis an issue, it's the Governance of "essential service", My sense is the legislation needs to change to a) Force this be be carried over IP (I know there are technical issues) b) Subsidize or keep a lid on prices?
In the first case, he seems to be decrying IP regulation and wishing us back to a non-IP world (which we haven't been in since the dawn of DSL). In the second case, he seems to believe that having copper telephone service is essential to survivability for the elderly. The latter point is more apt and the former has had little bearing for almost 20 years.
No, what the author should've said, instead of keep copper alive, was "let's regulate IP communications". The US suffers because of a lack of modern common carriage laws, and we will continue to lag in technical development until this problem is fixed.
Forget copper, stop fighting a war from the 80's. Let's regulate IP.
This is really just an attempt by ATT to get us to help subsidize their retiring of the legacy systems that we helped subsidize in the first place.
The US comms infrastructure seems to move and evolve slower than at smaller countries for very simple reasons. When you take the lead you give late comers the advantage of looking at what you have done to improving upon. Our infrastructure is absolutely massive when compared to, say, Estonia's (not to pick on them at all, great country). This means that rolling out step improvements is very costly. There's something pro-government folk seem to insist on ignoring: ROI. Companies can't roll out new massive infrastructures and throw away investment from the prior generation. Yes, progress in the context of a significantly larger system will look and feel slower. Our next generation IP infrastructure is being incubated today through experiments and evolution. Perhaps it will take the form of 1Gb/s fiber to the home or something less radical but equally awesome. Only private for-profit enterprise can bring us this.
It's interesting then when some actual competition comes along speeds increase and prices lower. Look at what happened with Google Fiber in KC. This is an often repeated point that has been proved wrong time and time again. Sure, there is some truth to it on a basic level, upgrading services cost money. In reality the companies have the regulatory bodies under control, the politicians under control, and they don't have to innovate or invest to make insane amounts of money. Why not just improve services in cities where the population is very dense? Why not in places like New England where there are many cities close together? Why if I call to cancel my service my prices can be cut in half? They have the money, there is just zero competition so they don't care.
You have the math exactly backwards. If you have ten million customers instead of ten thousand then you have a thousand times more customers to spread your costs over and you get greater economies of scale.
>Companies can't roll out new massive infrastructures and throw away investment from the prior generation.
Sure they can. They just don't ever want to, which is why they don't and why regulators are needed to make them do it.
>Our next generation IP infrastructure is being incubated today through experiments and evolution. Perhaps it will take the form of 1Gb/s fiber to the home or something less radical but equally awesome. Only private for-profit enterprise can bring us this.
Kindly explain how some foreign countries with highly regulated or state-operated telecommunications companies already have faster internet connections at lower prices than major American cities.
> Kindly explain how some foreign countries with highly regulated or state-operated telecommunications companies already have faster internet connections than major American cities.
Because their infrastructure decisions come after our infrastructure decisions.
It takes N years to "wire" a large city like Los Angeles. After that you have to get your money back. Yes, that is not a criminal act. Meanwhile country "X", many years later, looks at what was done in Los Angeles, checks out the more current technological offerings and installs a system that yes, probably leapfrogs what the average customer in Los Angeles has.
The true test would be the next evolutionary step. When a city like Los Angeles is ready to make the next big shift in connectivity it will be really neat. Country "X" will then fall behind and it will probably be very, very difficult for them to make the next leap 'cause, well, they have to pay for the investment just as well.
The technology that is going to make a huge difference is fiber to the home. We already have it in certain places in Los Angeles. I happen to know that there's fiber in the telco box right outside my house. Why? I saw it being installed fifteen years ago. Not lit yet. It will be, eventually. Fiber is great in that you can greatly enhance transmission rates by driving it differently. There's probably a few more evolutionary steps ahead of us before fiber bandwidth is tapped out.
There's another issue a lot of people might not consider when thinking about upgrading the infrastructure of a country like the US. If tomorrow we all had 1Gb/s service at home and at work it would probably crash the Internet. The infrastructure required in order to support something like that at our scale is massive and cost a ton of money. I don't know if costs are geometric or exponential. They are huge. A small country has advantages here as well.
No, government and regulation is not the solution. It never is.
Please explain how it costs more per house to install fiber when you can buy materials in bulk and have a larger number of customers to spread fixed costs over.
>It takes N years to "wire" a large city like Los Angeles. After that you have to get your money back.
They did that decades ago. The cost of installing copper has been paid. The reason they don't install fiber (and make no mistake, they have all but stopped) is that they have no incentive to do it -- increasing average bandwidth reduces their ability to engage in price discrimination against users who are currently willing to pay a premium to reach a threshold level of bandwidth that would become the new baseline. Scarcity raises prices. Why would they ever pay money to upgrade technology that doesn't increase their revenues? The answer in a free and competitive market would be competitive pressure, but they have no effective competition.
>The true test would be the next evolutionary step.
Explain what provides the local incumbent with any incentive to be the first to take that step.
> If tomorrow we all had 1Gb/s service at home and at work it would probably crash the Internet. The infrastructure required in order to support something like that at our scale is massive and cost a ton of money. I don't know if costs are geometric or exponential.
Rubbish. The costs are sub-linear. A user who gets a connection which is 100 times faster doesn't automatically transfer 100 times more stuff. The predominant cost of installing fiber to the home has nothing to do with bandwidth or networking equipment and everything to do with paying line workers in bucket trucks to dig holes and string fiber.
>No, government and regulation is not the solution. It never is.
Do you even hear yourself? Telecommunications is a regulated industry. It's right up there with power transmission and commercial construction. No one is even suggesting changing that. If you went to AT&T and asked them whether they would be in favor of repealing all the telecommunications legislation that applies to them, thereby disallowing them from using eminent domain and providing unlicensed access to all comers to the entire radio spectrum, they would laugh you out of the building.
The only question is whether the regulations we impose are ones that benefit the public or the incumbents, e.g. by causing fiber to the home to be installed sooner vs. later (or never).