sorted by: recent | see : popular

Content Tagged with goog + broadband

Today saw the first of at least two Congressional hearings concerning managing privacy on the web in relation to online advertising. The hearing today involved executives from Google, Facebook, Microsoft and startup NebuAd as well as the Federal Trade Commission and two public policy groups. For a complete listing, check out the hearing, although it clocks in at about two hours and is very, very repetitive.

Everyone present agreed that advertising on the web is not bad because it allows for all this wonderful free content; they similarly concurred that consumers are both uncomfortable with some of the data collection that occurs online want information on how they can control that information. After that, though, there was little common ground to be found over issues including self-regulation, the way NebuAd tracks Internet usage for advertising, and how long personally identifiable data is stored on Microsoft’s and Google’s servers.

I was saddened by the FTC’s unwillingness to put forth any meaningful regulations or guidelines related to behavoiral advertising, or to really even get back into the conversation. I was also (and here’s where the hate mail will start) impressed with Bob Dykes of NebuAd and his defense of that firm’s privacy technology (yes I know NebuAd said it would adjust the technology just yesterday to make it more palatable).

While much of what the firm says must be taken at its word (at least until the audit Dykes promised back in May is completed), surfing habits are harder to pinpoint to an identifiable consumer using NebuAd’s technology than search engine data. That isn’t a ringing endorsement, but it’s something. A larger fear about NebuAd’s technology that wasn’t addressed in the hearing is how the startup secures the data from its ISP partners. I trust my ISP very little, and now even less since they’re close to being granted immunity from legal protests related to them sharing phone calls with the government.

And the threat of government prying was by far the most interesting aspect of the entire hearing. In an age of government surveillance, the personal data such as that collected by Google, NebuAd or even my ISP is frightening. If the government chooses, it can find my web surfing information — perhaps stripped of context, but not of my identity. In the worst-case scenario, my searches could end up as evidence against me before a dozen of my peers in a municipal or federal court.

Think I’m crazy, or maybe have something to hide? I would point you to the brush with the government writer Lawrence Wright had researching an article on the Middle East (bottom of page 11), or the fate of those caught in RIAA’s nets. For those less concerned with government interference and more focused on protecting their online privacy, the next Commerce Committee hearing on this topic will focus on ISPs. I can’t wait.

Speculation that Google is working with French ISPs to build out a Gallic WiMax network has folks at Fierce Wireless wondering if Google may push open broadband overseas by investing in WiMax deployments. They point to a report in a French paper that says Google has teamed up with Illiad-owned ISP Free (Om loves these guys) and Bollore Group’s Bollore Telecom, which own licenses to the WiMax spectrum in France. Google has already invested in the nationwide U.S. WiMax effort Clearwire, has plenty of interest in making sure there’s open broadband in important markets, and money to spend. Allons-y!

The 700 MHz auction ended yesterday, and the $19.59 billion going to the Treasury looks like a lot until you realize the government’s total budget is $2.9 trillion. But now the waiting (and speculating) can begin. What will happen with the failed auction for the D block, which had been allocated for public safety? Who paid the $4.75 billion for the C block — Verizon, Google (not likely) or AT&T? What will an “open network” look like? As in life, the answer to one question often leads to many more.

• Ars Technica: Comcast Tweaks Terms of Service in Wake of Throttling Uproar
• Tech Confidential: Transmeta Hires Piper Jaffray to Examine $190M offer
• EETimes: Alcatel-Lucent Sees Flat Telecom Market
• BroadbandReports: Los Angels Taxes VoIP
• CNet: Is Mobile Really a Sure Thing for Google?
• The Register: Nokia’s N95 Successor Slips Out on Web

Google’s mobile ambitions have by now been widely articulated in the media, and after my initial post, I spent some time on the phone with various people discussing whether or not it made any financial sense for Google to be chasing the wireless dream. Many of the folks I chatted with expressed reservations about Google actually building a network, and felt that the company is using a big stick to get U.S. carriers to get a move on.

The FCC, for instance, has offered Google (GOOG) some measured encouragement in the hopes that the firm’s involvement in the wireless auctions would help push the prices up past the $10 billion mark. Google is, after all, obligated to make at least a minimum bid of $4.6 billion since the FCC agreed to their “Open Access” requirements for the C Block.

Secondly, this is going to be one expensive exercise for them. Assuming they win the 700 MHz auction and it costs them about $5 billion and another $2 billion in network buildout costs — that means a little less than $1 per share in lost income. Of course there is also the question of management focus, something that is much more difficult to quantify. By the way, our good friend, Ben Schacter who follows Google for UBS Research helped with the math here:

Think of it as not getting the interest on $7b and then tax affect it. So assume $7b multiplied by 5% (interest rate), then a 26% tax rate = about $260 million a year in lost income, or well less than $1.00 per share.

I think I am being conservative here. According to UBS estimates, the cost to build out the network is going to be about $25 a pop in urban markets and as much as $40 a pop in suburban markets, adding up to a total of between $8 billion and $10 billion. Of course, the operational costs of maintaining a nationwide network are humungous, never mind the service-support infrastructure.

The more I think about it, the less likely it seems that Google is going to build their own network. My guess is that they are going to try and participate via investments in other efforts. I had outlined one crazy scenario last week. Nevertheless, it is fun to see Google drive the wireless carriers batty with its posturing, and at the same time get what it really wants.

Pair-bonding for faster DSL, 3G wireless, and a beefier backbone using OC-768 (40 gigabits per second) technology are some of the things in the immediate future for AT&T (T), the San Antonio, Texas-based phone company, according to CEO Randall Stephenson, who was in San Francisco last week for the Web 2.0 Summit.

Looking fit as a fiddle and sporting a Texas Tan, Stephenson had just come off the stage after charming the skeptical audience and skillfully dodging some of the prickly questions he was asked by John Battelle. In a quick private chat following his on-stage conversation, Stephenson soldiered on with his charm offensive, and outlined a vision for transforming the phone company into a communications company with an emphasis on broadband and wireless.

The New Network

I started off by asking him when AT&T will boost speeds on its DSL connections and give us at least a semblance of what is considered broadband in other parts of the world. He pointed out that while I might be buying a faster connection, it is unlikely that I’ll get anywhere close to it. “We constantly test throughputs on all other networks and from our tests, median throughput is around 256 kilobits per second,” he said.

The problem lies in the infrastructure used to deliver the content — from the data center gear to the networks that bits have to traverse. AT&T, he said, is building an OC-768-based nationwide IP backbone “that will address the issues of speed and latency and bring content to you faster speeds.” Of course, it means serving your content from AT&T data centers via AT&T pipes.

Google (GOOG), Stephenson pointed out, has put its data centers close to the end customer, allowing it to offer a superior experience; AT&T’s new wholesale content division, he said, will offer the same “closeness.” Not that they are having any trouble selling. The company, he said, is experiencing strong demand for its bandwidth-related services, largely due to a boom in video-related activity.

AT&T is investing heavily in its network infrastructure primarily because the demand for bandwidth keeps increasing. “The early buyers are your big dogs like YouTube,” he said. The consumer traffic (driven by DSL) is up 40 percent per annum, while the boom in wireless broadband has sent the amount of bits being pushed up by a factor of four. Business-related traffic, he said, is up 60 percent or so over the past year.

The U-Verse Factor

The conversation then shifted to U-verse, and why AT&T was not pursuing an all-fiber strategy. Stephenson did an admirable job of defending his belief that his company’s U-verse (fiber to the curb + DSL) strategy was better than the all-fiber strategy adopted by Verizon (VZ). U-Verse is AT&T’s IPTV service, which is being slowly rolled out in different parts of the country and at present has over 100,000 subscribers. Getting U-verse service status to the mass-deployed level will take between three and four years.

AT&T’s IP-based video system is superior, he argued, because it’s able to send only the HD video channel that a subscriber has asked for, while other service providers are forced to pump out HD channels constantly.

“It’s not the question of fiber or copper,” he said. “What happens when you have to send 100 HD channels? The fiber capacity will get used up pretty quickly.” AT&T sends one channel that currently uses up 6 megabits per second.

“Pair-bonding is coming next year,” he said, which will allow AT&T to send out more than one HD stream at a time. (Some have forecasted that within five years, we will have 3 HD and 2 SD streams coming into our homes.) Pair-bonding, which will use gear from Alcatel-Lucent (ALA), will allow AT&T to connect homes to the fiber nodes at over 40 megabits per second. This is not the first time AT&T has brought up pair-bonding as a solution. (Related: Our post on pair-bonding.)

The 3G iPhone

As my allotted 15 minutes were winding down, I asked Stephenson why AT&T introduced the slow EDGE network for the iPhone, which is one of the reasons I am not using it any more. “Steve [Jobs] wanted to have ubiquitous coverage for the iPhone, and you have to remember it was two years ago when 3G networks were not everywhere,” Stephenson said. So they went with the slower EDGE network. So does that mean there will be a 3G iPhone soon? “Yes, there will be a 3G device,” is all he would say.

“As a consumer device, it has met all expectations, and for me personally it has changed how I travel,” Stephenson said of the iPhone. Two of his favorite iPhone activities, he said, are catching up on news via the use of hotel Wi-Fi connections, and watching “24.” He also uses the device to check his personal email, though not his email for work.

“From a business perspective, iPhone has brought in a lot of traffic to our stores and helped with the brand transition (from Cingular to AT&T),” Stephenson said. “So I am satisfied.”

Photo Courtesy: ZDNet/Dan Farber

Meraki Networks, a Mountain View, Calif.-based mesh networking gear maker, which is leading the unique ComMuniFi model of providing wireless broadband access in communities, neighborhoods and small cities, has quietly launched a new three-tier business model that may boost the revenues, but it also might alienate some of its existing customer base. The problem: 2x increase in price of some of its gear from $50 to $150 per wireless router.

As part of the new plan has Meraki now will offer gear and services tailored to three market segments - Standard (for community and individuals wanting to set up free networks), Pro and Carriers. As part of this change, the company is going to increase the price of its gear by as much as 200 percent, a move that impacts its customers in the Pro-tier.

The Pro-tier includes property owners and small network and hot-spot operators, who are currently using Meraki to offer for-pay wireless broadband. The price increase has some of their customers, especially those who are currently operating Meraki-based networks, up in arms. They will now have to pay $150 per Meraki router versus $50 they paid previously.

Many of them were lured by the low cost of Meraki gear, in addition to the superlative technology, and the company runs the risk of alienating some of those customers. “It is no longer the cost effective system it was credited to be,” wrote one poster on the Meraki forums, while another lamented, “[What] a shame … drawn in by a cost effective method just to be slapped in the face by an uncaring company that used us as pawns.”

“This is part of the business evolution of the company,” says Sanjit Biswas, co-founder and CEO of the company that has raised venture capital backing from Google (GOOG) and Sequoia Capital. This is a dilemma that is faced by most projects that have roots in open source but eventually have to evolve into a for-profit business. Biswas justified the price increase because it comes with other benefits such as guaranteed support. The company will also start selling its gear through the channel in addition to direct sales, and needed to boost prices. I am not sure about the magnitude of the price increase.

Biswas explained that company has envisaged lot of interest from Internet Service Providers and carriers, especially those outside of the US, and that is why the company is introducing a “carrier edition” of its products. The price for devices being used to power free community networks remains unchanged at $50, though the company is going to include advertising on the landing pages for these networks.

The advertising move shouldn’t come as a surprise. Last month, Sonic.net, a Bay Area ISP had started offering wireless broadband using Meraki gear, and that offering included advertising as part of the package. Of course, given Google’s investment in the company, it isn’t hard to connect the dots.

“This is still an experiment, and if it works, then we can see advertising revenues subsidizing the hardware costs,” Biswas said. There are plans to share advertising revenues with network operators as well.

Related: Our previous coverage of Meraki.

Update: Dane Jasper of Sonic.Net left a comment saying that this is their own initiative and the equipment is coming from Meraki.

We’re doing this independently, using equipment from Meraki. Meraki and Google have an ad partnership, and any revenues that flow from that will be split with our customers.

Dane says that if the program works well in SF, then it would be expanded to other Bay Area regions where the ISP currently offers the service. (Original post below the fold.)

Despite the best efforts of Earthlink (ENLK), Google (GOOG) and Mayor Gavin Newsom, San Francisco MuniFi project is still stuck in neutral and going nowhere fast. For San Francisco residents, a new option has emerged: a tag team of Sonic.net, a Santa Rosa, CA-based independent ISP and that is using gear from Meraki Networks, a wireless hardware company based in Mountain View, Calif., and is trying to promote an ad-supported MuniFi model. (Its actually more like community wifi, and you can call it ComMuniFi.)

Sonic.net today notified its customers via email that they can get a Meraki wireless mesh router at a subsidized cost, which will allow them to connect it to their DSL line. The wireless router will share up to 500 kilobits per second of the bandwidth available on the DSL line.

Network users will see a Google ad bar at the top of the browser. In the future the ad revenues generated by this ad bar will be split between those who choose to opt and place a wireless router on their connection, and will be credited against their broadband bill.

It could be a rather small credit, so don’t get your hopes too high at this stage; this is still experimental and we are still working out many of the details.

This is a good model for Google to imitate in other regions as well. Google’s had to have known all along that their San Francisco grand plan was going to run ran into political trouble. The big question is why didn’t they roll out A similar service with Earthlink, a much larger ISP with many more broadband customers, would have been a better option for all concerned. I have become a fan of this community-based WiFi plan, which doesn’t need a lot of government dollars, and instead bets on citizen’s desire to share. Independent ISPs such as Sonic.Net are more likely to embrace this model.

Meraki backed by Google and Sequoia Capital, is one of the companies which has been championing a more community approach to free wifi. It recently announced plans to expand their experimental Meraki network to all across San Francisco.

Meraki has been selling its wireless 802.11b/g access point and mesh repeater, the Meraki Mini, for $49 and claims its products are already being used by 2,000 networks in 40 countries. The company also lists an outdoor ruggedized version of its Meraki Mini for $99. Meraki’s business is being built off hardware and software based on MIT’s Roofnet project. The Roofnet Project was previously funded by MIT’s Project Oxygen and NTT DoCoMo.

Related News: Our previous Meraki coverage.

Over the past few weeks, there has been a steady increase in the number of articles questioning the wisdom of municipal WiFi networks and wondering if they really will ever realize their potential. It seems the pendulum has swung from one extreme to another.

With some of the existing MuniFi networks plagued by technological and business model issues, some cities such as Milwaukee, are rethinking their MuniFi plans and taking a more cautious approach, before committing themselves. The trials and tribulations of MuniFi’s biggest commercial champion, Earthlink, haven’t helped matters either.

The San Francisco MuniFi effort is mired in political morass, while Google (GOOG), one of the most visible champions of MuniFi, seems to have shifted its attention on the 700 MHz wireless auctions.

MuniFi, to some extent, has been a victim of putting expectations ahead of reality. MuniFi used to be seen as the panacea of all our broadband problems, giving us the ability to roam free and make VoIP calls while watching the great Meteor shower. Reality turned out to be a bit different.

Andy Abramson laments about the poor performance of Philadelphia and Anaheim networks, especially their ability to handle VoIP traffic.

I have to say that part of the blame is the deployment of a mesh network architecture from Tropos…Tropos officials admitted to me earlier this week that mesh as it currently sits is not really ready for VoIP until the voice 802.11 R standard comes along.

One of the reasons why MuniFi seems disappointing is because the “selling” of the concept involved offering consumers Internet access. Yet a recent report from Forrester Research shows only a tiny percentage of general consumers using muni wireless. Craig Settles, an independent consultant, points out that consumer is the weak link.

If the MuniFi networks focused on public service, government, first responder and educational services, it would be easier to make a business case. Settles conducted a survey of 318 economic development directors, managers and professionals which takes an in-depth look at the actualand potential impact of muni broadband on economic development. (Download Report, PDF file)

His findings are pretty surprising. For instance, in municipalities where networks are in place, “[B]etween two and three times as many feel wired rather than wireless muni networks will have a direct impact on attracting, retaining and improving businesses.”

We are not ready to write off MuniFi just as yet, just like we didn’t see it as the much vaunted third pipe for consumer broadband. Over next few months, technology will improve, and communities will find a happy medium. And maybe we will use our WiFi mobile phones in the park!