Slashdot Mirror
Gigabit Speeds At Home In the US
An anonymous reader writes "The Electric Power Board of Chattanooga is preparing to offer 1 Gigabit speeds at home by the end of the year. 'The city-owned utility announced today it will boost its broadband service to 1 Gigabit throughout its service territory by the end of 2010. Such a connection will be 200 times faster than the average broadband speed in America and the fastest of any US city.' The NY Times reports that the service will cost $350 per month. 'Mr. DePriest of EPB does not expect brisk demand for the one-gigabit service anytime soon. So why offer it? "The simple answer is because we can," he said.'"
$3,5 per mbps is pretty close to the wholesale prices - and it would be pretty hard to get that for just 1 gbps. Where's the catch ?:)
Additional verbage. http://www.chattanoogagig.com/
... the Chattanooga Choo-Choo!
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
Get 199 friends and split the bill to get 5Mbps for 1.75$US per month!
I'd certainly pay $35 for 100 meg though.
If you could split it 7 ways, that would be a 18 MB line each at $50, which is a good deal compared to the semi-monopoly prices you usually get. Of course, this could vary depending on how close to a gigabit the line will actually get you (although it shouldn't be worse than the big ISPs, and may be significantly better).
My webcomic
Where I live 400Mbit is about $1000/month (that has to be adjusted for the fact that the price and salary levels here are generally a fair bit above the US, but still). The 1000 Mbit option is "call us for price". I think you'd better be sitting down if making that call for a quote.
The only consolation is that we don't oversubscribe over here. You get what you pay for. But boy, do you ever pay.
Some (few) things are best provided in a monopolistic environment. Utilities (like power) and infrastructure (like this) are typically in that category. However, that's best in a public monopoly, not a for-profit, private monopoly.
>> 'Mr. DePriest of EPB does not expect brisk demand for the one-gigabit service anytime soon. So why offer it?
Because there is a huge opportunity for resale or inclusion in basic services of multi-tenant (residential or business).
Give 10 businesses 100MB/s for $50 / month and you're making money or for offer it free and it's a cheap inducement lease space
Give 100 tenants 10MB/s for $10 / month and you're making more money or for offer free and it's a cheap inducement to renters
Then the Midwest might bring down the average speed. But there's absolutely no reason why San Francisco, LA, Chicago, NYC, shouldn't have the same high speeds as entire countries like Japan, Korea, etc.
However, the argument you're using isn't even a good one for the Midwest. Sparsely populated places are easy to reach with long fibers, and so cheap to bring high bandwidth to. It doesn't take a huge operation or investment to bring fiber to nearly everyone in Montana or Wyoming.
The real answer is that US the telecom network cartel has never been aggressive in bringing Internet to homes. Quite the opposite: every time there's a push to increase the reach or speed of the network, the telcos have been there to push back, claiming the new traffic load will kill the existing network, or some other malarkey. What they're afraid of is that more bandwidth creates more opportunities to compete with them, and gives them less time to milk ancient services for a dragged out period of pure profitability before investing in a new generation. And that's exactly what they've got, and what we're stuck with. Except when an org not in their cartel provides some actual competition, like this municipal network operator.
--
make install -not war
There's no incentive to upgrade if the consumers are forced to pick from amongst X number of similarly-priced oligopoloies. I didn't spell that right, but I don't think most people know what that word means anyway so fuck it.
Anyway, in Canada there's not even that. I can get cable from Shaw or ADSL from Telus. Those are exactly all of my choices unless I want to go to dial-up.
---
ECHELON is a government program to find words like bomb, jihad, plutonium, assassinate, and anarchy.
This just shows what can happen when private companies are allowed to compete without regulation to provide services much more cheaply and efficiently than the gov... oops, hang on, I'll try again.
...government bureaucracies try to implement services that could be done more cheaply and efficiently than the private secto...wait a second...
...
Shit.
I can see this subscribed to by small businesses with data heavy uploads (film production companies, ad agencies etc). Spread across an office of 20 employees, $350 is peanuts when each worker is getting 50mps, assuming it's symmetrical.
However I think the price for the gigabit service will drop to something hotly competitive like $99 within 36 months as the electric utility begins poaching customers from the established players when it hits home that selling access to information is more profitable than burning coal.
It wouldn't surprise me if shareholders and even regulators eventually order a spinoff of this tail-wagging-the-dog broadband division, and it winds up with a cable co, where it all gets dialed back to the current offerings.
- js.
(stops laughing)
Realistic situation:
College students living in a house with ~5 people, $30/month five ways for 1-mbit service comes to $6 per month per person, which two of the people don't pay until you padlock their rooms with a sign saying "see me".
Unless the house is near an open wi-fi, then nobody even brings up the issue of getting internet for the house.
Well,
Simply recruit 10 neighbors and hook them to a 10 port router and wallah! At 35 bucks plus taxes, it's cheaper than many solutions and the speed is almost guaranteed to be superb 100% of the time. How about that?
It is easy to offer gigabit speeds. You provide a line that signals at a gigabit, probably just Ethernet. The hard part is having the infrastructure above that which can maintain it. This is particularly the case if you have multiple lines.
My bet is that at that price, they have insufficient upstream. So you sell your gig line out and you discover that really you are lucky to get 100mbit at the best of times. Thus your customers are getting less than they paid for and so on.
how much bandwidth per node / headend backend?
Geeks get too obsessed with the big pipe numbers and don't stop to think the costs of backing all that up whit the infrastructure upstream you need to maintain that speed. That is something I've observed is common in many of the countries with the really fast Internet. I remember a guy from Japan posting on Slashdot how great his 100mbit Internet was, he could download a CD in about 10 minutes. I had to point out that is not 100mbit, that is 10mbit. Nice and fast, but same as I was getting on my connection (12mbit at the time).
Especially if you have a high density area like an apartment building, but even if not, it isn't hard to offer Ethernet to the units and that will be 100mb or 1000mb of course. However it is a lot harder to have all the stuff higher up to keep maintaining those speeds.
Also you discover that some function like big WANs. They've got reasonable internal bandwidth, but not a lot outside. Latvia seems to be like this. They rank highly on Speedtest ratings, but it is all people testing to their own ISP's speed test servers. When I test those speed test servers from a large bandwidth site in the US, they get only a few megabits. So you'll see good transfers to others on your ISP, but not so good when downloading from a website in another data center.
An impressive connection not only has good bandwidth to your house, it can back it up at higher levels. I've been happy with Cox Home Business for that reason. It is reasonably pricey ($150/month for 50/5mb and 4 static IPs) but it has good infrastructure. I get my speeds, and to many different sites. It isn't like I get it to their internal test server but nothing else, I can download from Steam and Impulse and so on at those speeds.
Any time you see something with tons of bandwidth for a small amount of money, ask yourself what the catch might be. Remember that lots of bandwidth requires a lot of expensive equipment to make happen, and a lot of connections to other large networks. That isn't free.
It's actually even better than 100Mbps per person sharing it, as long as each of the 10 persons uses it in a reasonably intermittent fashion. Everybody can get more than their share some of the time as long as nobody gets their whole share all the time.
Are you adequate?