Google's Experimental Fiber Network 363
gmuslera writes "Not enough speed from your ISP? Google seems to go into that market too. 'We're planning to build and test ultra high-speed broadband networks in a small number of trial locations across the United States. We'll deliver Internet speeds more than 100 times faster than what most Americans have access to today with 1 gigabit per second, fiber-to-the-home connections. We plan to offer service at a competitive price to at least 50,000 and potentially up to 500,000 people.' The goal isnt just to give ultra fast speed for some lucky ones, but to test under that conditions things like new generations of apps, and deployment techniques that take advantage of it." If they need a test neighborhood, I'm sure mine would be willing.
more competition (Score:5, Interesting)
Google (Score:5, Interesting)
Wow! (Score:1, Interesting)
Awesome, then maybe you Americans can start catching up with the rest of the world :) Funny, 10, or even maybe only 5 years ago, I would had been very impressed by just hearing 1 Gbit. Nowadays, eh. Sure, I only have 100/100 Mbit, but I have no restrictions whatsoever (and I use my fair share) and I pay $10. I could go for the 1 Gbit package, but why pay 5 times as much ($50) per month for something I don't see myself needing. Maybe in a year or so :)
Re:Oh no... (Score:5, Interesting)
Well, they could. So what? Instead of Comcast, Cox, Charter, AT&T, Verizon, etc, Google can get their stats direct. Yes, there's a much bigger pipe, but you and I are still generating the traceable data as fast as we can.
Of course, I see another possibility for this. You know how many of these ISPs are trying to make providers pay for "preferred" access? Maybe Google is seeing this as a way to ensure net neutrality in the market, or possibly turn the tables. We shall see if it makes it far into the market, and if it ends up making a real difference.
I, for one, would welcome such a bandwidth overlord.
No article, just a blog post (Score:3, Interesting)
Google really understands Slashdot. Everything pertinent in the blog post fits just fine in the summary. No need to read the "article." It isn't an article. It's just a blog post soliciting proposals.
Seems an odd way to go about it. It's missing the one thing that is relevant to market demand: price. I don't know whether my community would care to be part of it because I don't know how much it would cost the users. I know for a fact that people are only willing to pay so much for high bandwidth, and past that, they decide it's not worth the cost.
Worse, most internet hosts throttle or load balance their outbound throughput to any one destination. I had 20 megabit downstream service for a while, and the only way to come remotely close to saturating it was bittorrent. And I never did saturate it. I managed to sustain over 10 megabit only twice, ever, and that was hard to do and didn't last. Even most streaming video sites transmit at no more than 300 KB/s (2.4 megabit), and many, if not most, transmit slower than that.
Sounds to me like the whole thing is going to be a disappointment to them. Truly high bandwidth demands will only emerge when truly high bandwidth (1 gigabit) is widely deployed and widely subscribed to, and when major servers move from truly high bandwidth to absurdly high bandwidth (10 gigabit through to the backbone). All of their scenarios can be satisfied by deploying fiber to just a few premises, like hospitals and clinics, which is a big dumb duh idea anyway. It's not already done? The nebulous "let's see what happens" goal they have depends on lots of people having access to lots of bandwidth. Network effects have to kick in before a network is valuable. Build it and they will come, but there's no way to predict what they'll actually use it for. It will take large numbers of bored programmers fiddling around with their high bandwidth to generate something to use all that bandwidth, and they won't bother if 90% of their potential audience has 1/1000th of the bandwidth.
In short, it's the network, stupid.
IPv6? (Score:1, Interesting)
I'm guessing this is going to be IPv6 to the home? Any confirmation?
When you build a network that fast the use-to-idle time drops very low. What is the cost of operating the links when idle? Maybe at a certain speed it makes more sense to go with a low-bandwidth carrier and invoke (turn on) high-bandwidth when needed?
Re:more competition (Score:5, Interesting)
I see this as a response to the filtering and tiering complaints. Google seems to be attempting to say "Fine, take your toys and go home. We'll just give everyone new toys"
I wish them luck, and hope we're not seeing the founding of the new IBM/Ma Bell empire.
Re:Here is what is going to happen. (Score:3, Interesting)
Re:What is Google's interest? Data Tracking? (Score:5, Interesting)
Because it is damn expensive, that is "why not".
The "why" needs to give them a good reason. My guess is this:
1. Google's main revenue generator -- ads -- are very effective. I know a lot of people who hate Internet ads but don't mind Google's because they aren't in-your-face offensive. Considering their revenue, there are a LOT of people like that.
2. The better your experience on the Internet, the more money Google makes.
3. Google, therefore, rolls out products designed to improve your experience on the Internet.
4. Profit! (Goto 2)
This is the same logic I use to believe that Chrome isn't a threat to Mozilla Firefox. All Google cares about is better, faster, stronger Internet experience. They have the tools, they can rebuild it. Chrome isn't a competitor to Firefox.
Comment removed (Score:3, Interesting)
Re:Google (Score:2, Interesting)
Re:Here is what is going to happen. (Score:5, Interesting)
The big names in networking (AT&T, Charter, etc.) are going to sue Google on antitrust grounds because it is easier to hire lawyers than to upgrade failing and obsolete networks.
Perhaps they will. But consider: this is not a profit engine for Google, in much the same way that Android isn't a profit engine. Google says this service is to test new high-bandwidth technologies, and I don't doubt that's true, but it's probably also true that they're just trying to upset this market because the established cable companies are a threat to their other businesses, both because of their slowness to raise the bandwidth bar and because of their marriages to legacy content distribution.
Because of this, Google probably doesn't care whether they own this service or not. I bet if the big networking dinosaurs sued Google, Google could settle with them by agreeing to spin off the fiber Internet company, yet still accomplish all of the original project goals. It would be like if Google had to cut Android free - it would still satisfy Google's main goal of creating an open platform that's more friendly to their mobile web services than Apple's or Microsoft's is likely to be.
Re:Google (Score:3, Interesting)
For the love of everything in every power, that people do and don't believe in. I want more competition. If google can bring it, I'll sing praises until I'm blue in the face. I have the option of 2 ISP's. Both with 60-90GB/mo caps, this is nothing but good in my book.
Re:What is Google's interest? Data Tracking? (Score:2, Interesting)
Re:Google (Score:3, Interesting)
We're paying something like $750/month each for 3 T1 point to points between our main office and our 3 field offices. Might be per end, might be both ends, all I know is we pay through the nose for this crappy service.
Not to mention all the lost productivity waiting for large files, or even windows updates, to traverse the network.
If google wants a rural test bed, I guarantee they would be well received by state officials in wyoming.
Re:What is Google's interest? Data Tracking? (Score:2, Interesting)
Now the real question is why are they going into providing consumers fiber access.
The bandwidth costs of YouTube alone would justify them deploying their own infrastructure. If they can shift X% of their users to their own network they only have to pay for the overhead of maintaining that network. The rest is savings for them.
6 years and another 1G (Score:2, Interesting)
Re:What is Google's interest? Data Tracking? (Score:5, Interesting)
Among many other reasons, its a net neutrality insurance policy. Google favors net neutrality, but if net neutrality foes succeed (and that's an ongoing threat, because they don't tend to back off even as the FCC reiterates its support for net neutrality principles) it needs its own links directly to consumers as a hedge against other big network providers (particularly those that are also trying to compete with other Google services, whether video offerings that compete with YouTube, phone offerings that compete in some ways with Voice, or something else) -- impairing access to Google's services. If Google can position themselves as a competitive fiber-to-consumer provider, it puts them in a position where such actions by competing service providers that are also fiber providers are riskier because of the potential for retaliation.
Google has a strategic investment in not making the internet into a set of disjoint walled gardens, but ultimately the best way of insuring that is to guarantee that if its competitors try to convert it into such a system, those competitors will lose.
Seattle (Score:2, Interesting)
Re:What is Google's interest? Data Tracking? (Score:3, Interesting)
Except they can claim they're bringing competition to areas that don't have it, such as mine, and encouraging the local content provider to invest Federal dollars in my area as well. The Federal Gov't has already earmarked a large chunk of money for doing exactly what Google is talking about doing privately.
Now, I agree that I wonder if Google is the best one to do this. But, who else is going to try? Another large media conglomerate that owns channels and distribution and newspapers? Like Comcast? Oh yeah, my bad. Talk about a monopoly. Let them complain to the FCC, I'll just mediate and introduce the two: "Pot, Kettle".
Coming Home to Roost (Score:2, Interesting)
A moral lesson for the deregulated.
I'll bet there were a lot of smug phone company executives who thought at the time of the breakup of old Ma Bell that there will never be anyone willing to lay down enough wire to seriously challenge them.
Then they got cable TV and wireless phones, but a lot of the data moving business is still in the hands of the Baby Bells.
This ought to cut them down a notch.
But then we'll need a new search engine. I won't search where my bandwidth comes from!
I'm sure your ISP is recording every move you make, and Google is recording half the stuff that moves on the web. A Google ISP brings the two together and will be a privacy nightmare.
Re:Here is what is going to happen. (Score:4, Interesting)
The big names in networking (AT&T, Charter, etc.) are going to sue Google on antitrust grounds because it is easier to hire lawyers than to upgrade failing and obsolete networks.
Maybe. Except that, if they base their anti-trust claim on the basis that Google would own both the content and the connection to that content, then Comcast's acquisition of NBC would also be subject to this same suit.
I can't see Comcast suing itself.
Re:What is Google's interest? Data Tracking? (Score:3, Interesting)
Personally, I want it (Google's ads are less offensive, and 1Gbps? Yes please).
Re:Here is what is going to happen. (Score:4, Interesting)
When Google offered 1GB mail storage, that was industry changing. They didn't end up with 1GB storage per customer though, most people use far less. The trick was to avoid the pitfall of only getting the heavy users.
Perhaps they can pull the same trick with fiber.