Gigabit Internet Just out of Reach in Seattle
On 13 December 2012, the City of Seattle announced a breakthrough in a previously shelved plan to bring fiber-based Internet service directly to homes and businesses. A partnership with a private firm, Gigabit Squared, and the University of Washington will make use of a fiber backbone that the city has built over many years to serve its own needs, and which it also leases for Seattle-based county, state, federal, and school uses. A previous plan had called for issuing revenue-backed bonds to fund the effort directly.
But the service won’t initially cover the entire city. Instead, the program will launch in 12 neighborhoods that comprise about 50,000 homes and businesses, and it will be connected directly to individual buildings: this is called fiber-to-the-home (FTTH) or fiber-to-the-premises. It will also use gigabit wireless (over licensed dedicated frequencies) to beam service from within the coverage area using highly directed line-of-sight transmissions to nearby multi-unit buildings, like apartments, as well as businesses. There’s also a vague description of building a high-speed wireless cloud in covered areas, but it’s unclear at the moment if that’s Wi-Fi or something else. All these
services have symmetrical throughput, offering the same bits-per-second upstream and downstream.
What’s interesting about this proposal, as far as it’s currently defined, is that it’s not a “triple-play” service that includes broadband, video, and voice. Some firms offer a “quad play” that adds a cell service, too. (The “play” is a baseball metaphor here: a company makes a triple play to “win” in the business.) Most of the city-wide fiber efforts to date, whether with private partners or run by cities or appropriate public utilities, have focused on the triple play.
But the triple-play approach requires provisioning services. Instead of having a big, dumb Internet pipe, video gets a dedicated chunk of broadband, as does voice. These provisioned services are designed to provide quality comparable to having a dedicated wire (like cable TV or a hardwired phone line). That’s no longer necessary when you have gigabit service, though, or even, say, reliable 25 to 50 Mbps service — so long as you aren’t beholden to the specific 24-hour-a-day, multi-channel video offerings of cable and satellite. Voice, too, can be easily replaced with Skype, Vonage, and other services. (For more on smart and dumb pipes, see “New App.net Social Network
Aspires Beyond Chat and Ads,” 28 August 2012.)
Seattle’s plan may also be an attempt to avoid too much “disruption” at once. If the gigabit plan doesn’t compete with cable services, Comcast still has a role to play. And, one imagines, Comcast may need to step up its game in terms of pricing and higher-tier bandwidth services, which are offered in some markets. CenturyLink, Seattle’s incumbent phone provider, has no plan for fiber to homes and businesses, and its DSL offerings are slow and erratic. It has a fiber-to-the-neighborhood (FTTN) plan that allegedly brings 12 to 24 Mbps of raw throughput, but it isn’t price-competitive with cable. Landlines are being rapidly shed by households, and DSL speeds have been unable to keep up in practice with cable modems, and can’t
hold a candle to fiber.
(For those having trouble keeping track of who is the “phone company” these days, US West was the Baby Bell that served most of Washington State as well as a bunch of the Northwest and a few other scattered states. Qwest acquired US West, and then was itself sucked into CenturyLink. CenturyLink used to be mostly a rural telephone company known as CenturyTel, but it acquired Sprint’s spun-off landline business, Embarq, and then Qwest. Verizon has been urged to sell its landlines, too. Landlines are a dying business.)
You’d think phone companies would be ideally placed to put in fiber to the home, but they all switched their focus to the high growth and profit in mobile voice and broadband. AT&T has its U-Verse FTTN, which gets decent reviews, but has only 7 million broadband subscribers (out of about 30 million locations to which it’s available) in AT&T’s vast service area, and it isn’t growing. The firm’s CEO just said that it’s more or less done building out the offering. Verizon’s FTTH service, FiOS, could carry gigabit, but doesn’t. And FiOS’s 300 Mbps down/65 Mbps up offering costs $210 per month, far higher than the gigabit services in cities that have already deployed FTTH. Verizon has also more or less said it
won’t build out more FTTH, too, with only about 3 million customers taking the service of 15 million to whom it is available.
The joy of gigabit Internet is unfettered access, of course, but it’s not necessarily about today’s Internet. As I wrote recently in the Economist, citing Cyrus Farivar’s Ars Technica coverage, most Web sites and services currently either can’t keep up with gigabit connections or simply don’t have anything to fill the pipe. If you need only a few Mbps to get the best possible HD streaming movie over the Internet, gigabit service provides you the overhead to ensure you can always get it, assuming the connection to Netflix, Amazon, or others is clear. (For
backups and other huge data transfers, gigabit already shines.)
No, gigabit Internet paves the way for the next big thing: whatever services develop for customers who always have massive bandwidth available, whether it’s super-high-quality two-way teleconferencing for remote workers and remote offices (something that can be done with modest quality today), the delivery of Blu-ray-quality video as temporary 50 GB downloads that take just a few minutes or that can be burned to Blu-ray disc for future use, or high-bandwidth gaming that provides more real-time interaction at higher rendering resolutions. In the short term, always getting the highest possible speed from any site or service probably suffices.
I’m peeved because I don’t live in any of the 12 initial neighborhoods, although I’m not far away: a stone’s throw from the University of Washington and the north end of one of the large neighborhood segments south of the university. I’ll just have to wait and sob over my “mere” 25 Mbps cable connection, itself the envy of many in Seattle served only by a second-tier cable firm or CenturyLink. Many of those areas of the city are, of course, getting fiber first, and they will soon be able to lord it over me. At least Jeff Carlson and Agen Schmitz — the other Seattle-based members of the TidBITS staff — aren’t any closer to a covered area than I am, or I’d never hear the end of it.
Article not relevant in todays internet world. Most web servers don't let you download at 45 megabit let alone anything close to gigabit. High end speed is with internal networks and not WAN networks.
Read third from last paragraph: "The joy of gigabit Internet is unfettered access, of course, but it’s not necessarily about today’s Internet."
Not to worry, Glenn, West Seattle will be the last to see such speeds (as usual). Grumble, grumble... ;-)
Irony or intent that three regular TidBITS writers/editors (me, Jeff, and you) are not covered in initial deployment? I smell conspiracy.
Having had Comcast "service" in the past, I would forego cable TV forever rather than be a Comcast customer/serf again.
I hate them when they were my TV provider, but they've been pretty great for Internet. I have had glitches, like their monthly cap and a complaint about the way they handle business service cancellations. Both both situations were worked out in my favor.
Google's not doing triple play either here in Kansas City. We're getting internet & TV but no phone. According to local reports Google said trying to take all the various local regulations into account was too hard. Of course since it's fiber to the house the copper lines remain in place and you can use the local phone company.
I really think Google's free offering (5 mbps down/1 mbps up) is the most interesting aspect of their offering. Pay for installation ($300 or $25/month for a year) and fiber internet is free after that.
Google's fiber may offer some other benefits such as faster connections to Google offerings like youtube just by not having slow interlinks to other providers to route through.
It's not raw speed i'm looking for. I'd just like my Crash Plan backups to not interfere with my XBox games.
Ah, that's interesting. Although I suppose voice is the easiest thing to duplicate because there are many competing offerings. Video is harder.
The "second tier" cable company in Seattle has gotten quite a bit better over the past year. Customer service is quite amazing, and the internet speeds are very good as well.
I'm also right on the border of coverage, enough so that I've emailed them to find the exact border of the service area...
Alas, the city's fiber runs across the street from me, but my neighborhood isn't among the chosen ones either.
There may be some of the city's notorious property developer politics involved here, particularly in the selection of South Lake Union. For someone looking to sell out a condo, the option for gigabit broadband would be a major selling point.
I'll be looking to see just who might pop up on McGinn's campaign contributor list, and what connections they have to this contractor and/or whoever plans to raise money for this network.
I have a telecom finance background, and will be interested to see the technical specifics. From what I can tell at the moment, they're relying at least to some degree on "millimeter wave" radio, which has quite a dodgy past. Here's the deal: The higher the frequency the smaller the physical size of the radio waves. At some point (I'd have to go check exactly what that point is) the waves become shorter than raindrops, snowflakes, mist, fog, etc., and the transmissions break down.
Several "millimeter wave" startups (Teligent, WinStar, Advanced Radio Telecom among them) were very prominent during the Internet/telecom bubble of the 1990s. They got huge market valuations. And then they went bankrupt because, alas, the s*** never worked. I know what Gigabit Squared (and especially its venture capital promoters) will be telling everyone, but I'll be quite interested to see what actually happens. If past is prologue, these promises and performance will be two very, very different animals
I'm not sure about those firms (I remember some of the names), but you can buy off-the-shelf commercial hardware for both licensed and unlicensed use that allows point-to-point and point-to-multipoint gigabit wireless.
I think you're conflating the operators (as you mention) with the technology. From what I can tell, Gigabit Squared (which is founded as an economic development corporation, so a little different than a venture-backed aggressive high-margin firm) plans to use wireless for fill-in, not for primary connections. It's using fiber for that.
I truly love all things telecom, partly because of how unforgiving the realities wind up being. The project's website is VERY skimpy on the details. I think needlessly so, given the lack of competition. It's impossible to evaluate in detail based on what they provided. The lack of detail is definitely a "red flag moment" for this veteran, anyway. The believers will always spin their enticing stories, but those of us who've been in the manholes, conduits, and pedestals with the engineers are much tougher customers.
As a former VC and corporate director, very few ideas ever made the cut with me, but when they did, at least they worked. I hope this one works too, but "hope" plus $3 gets you a cup o'joe at Starbucks. I've never been anyone's cheerleader, and I'm not going to start now. By the way: In a startup, the operators and the technology are inseparable, period. Also: You'd better look again. Their press release says they want to raise money. They're not running some charity.
I would like more detail, too, but they're probably just using BridgeWave's gear.
An economic development corporation is still for-profit (or can be). If you read about the approach they have, it's a private-public partnership methodology that doesn't produce telecom-scale returns. They'll make money, but their goals are attempting to align with social purposes as well. Lots of companies these days work that way; cf., beneficial corporations (B corps).
It's just money. They want to raise it, and to raise it they will tell a story. That's what counts. The rest is adjectives.
There is a lot of actual telecom expertise in Seattle. This outfit has no competition. There is no reason why they can't release ALL of the technical detail. You know, "open source" it. Let the 1,000 people in town who really know what they're talking about take shots at it and see what survives.
I sound like I don't want it to work. Other way around. It'd be great if it worked. But I've seen way too much money wasted on way too many of these things that not only didn't work, but which obviously weren't going to work yet managed to raise money during a bubble. Go ask Paul Allen how many billions he lost. There's a reason he's in real estate now. Land is easier.
I'm not holding my breath for the details. For now, I regard it as part of McGinn's re-election campaign. Nothing gets the hipsters breathing hard like promising a shiny new broadband toy. Gigabit! Wow!