Imbroglio by the Bay

Add San Francisco to the list of cities where ballyhooed municipal wireless networks have run into unforeseen obstacles.

San Francisco, in fact, is perhaps the most vaunted city-wide mesh project in the country, thanks to the involvement of Google (Nasdaq: GOOG) and EarthLink Inc. (Nasdaq: ELNK), which won the right earlier this year to build a wireless network that would provide free, faster-than-dialup (though slower than broadband) Internet access to all the city's residents as well as premium service to those willing to pay. (See Google & EarthLink Team Up.)

Google and Earthlink have both presented the San Francisco effort as an act of corporate benevolence and a way to bridge the digital divide that persists in one of America's most technologically advanced cities. Earlier this week, however, Google director of special initiatives Chris Sacca, who heads the muni-networking program for the Web search giant, made some biting comments about the city's "unreasonable" demands in an interview with the San Francisco Chronicle, noting that the project is still at essentially the same point as when the agreement to build the network was announced in April.

"Every meeting is like the first," Sacca told the Chronicle reporter, adding that the city's demand for a share of revenues would make the project economically unfeasible for the two companies.

Speaking with Unstrung Wednesday evening, Sacca tempered those remarks while still expressing his impatience.

"I don’t disagree with anything the city says -- they're just trying get the best deal for the people of San Francisco," he comments. "I applaud them for being this diligent. I just think we could apply this same level of diligence and still get [a contract] done relatively quickly."

San Francisco IT director Chris Vein had not responded to requests for comment as of this writing, and Earthlink spokesperson Deisha Galberth did not reply to multiple emails.

In any case, the city has already acknowledged that the network, originally scheduled to go live later this year, won't be functional until at least mid-2007.

Other cities have seen ambitious wireless network programs delayed or scaled back:

  • Philadelphia, where the project is at least a year behind schedule and the head of the city's wireless initiative, Dianah Neff, recently left to take a job with a network-consulting firm. (See Wireless Philly Loses Head.)

  • Sacramento, where initially selected provider MobilePro Corp. dropped out of the project altogether. (See Mesh Mess Sinks Sacramento Net.)

  • New York, where the City Council passed a measure last year establishing a committee to lead the creation of a municipal broadband network, but no contract award or request for proposals has emerged. (See Where Is NYC?.)

To be sure, some metro areas are moving ahead. In Silicon Valley, where wireless-network leaders explicitly rejected a government-led model along the lines of San Francisco in favor of a private-sector effort, a consortium headed by Cisco Systems Inc. (Nasdaq: CSCO) has begun laying the groundwork for a pay service that will reach some 2.4 million people in the heartland of the tech industry. (See Cisco & Pals to Unwire Silicon Valley.)

Indeed, the twists and turns of the San Francisco project could indicate that an inclusionary, citywide network that purports to reach all residents could be the strategic missile defense of the wireless world. Sacca implies as much, admitting that the original vision for a dense mesh network blanketing the city probably won't work.

"I think what everyone is learning is that the difference is not made out in street," Sacca says. "It's not the radio up on the lamppost that's really going to bring this service to the people who need it -- ultimately the client devices have to be there."

In other words, even residents and businesses using the "free" service will have to spring for some kind of exterior repeater or other piece of customer-premises equipment to bring the outdoor WiFi signal to the user's computer. "Instead of really pushing to attain deep penetration and coverage with the radios on the street," he explains, "it's going to take an ecosystem of repeaters to grab that [signal] and bring it into your home."

Meanwhile, Google is counting on a series of townhall-style educational meetings across the city to convince San Franciscans of the benefits of the municipal network -- thereby putting pressure on city officials to close a deal and start installing nodes.

"That's really gonna generate a lot of interest among the citizens," Sacca says, "plus it'll remind us of who we're building this for."

— Richard Martin, Senior Editor, Unstrung

lrmobile_JerryG 12/5/2012 | 3:39:54 AM
re: Imbroglio by the Bay We have no record of Mr. Martin contacting EarthLink's Corporate Communications department for this story. We have received not a single email, nor phone call, from Unstrung or Mr. Martin himself at any time over the past week.

Jerry Grasso
Director, Corporate Communications
RBMartin 12/5/2012 | 3:39:43 AM
re: Imbroglio by the Bay Not so, Jerry. I sent 2 emails to [email protected] (which according to your Web site is the only way to contact the Earthlink press office -- no direct emails or phone numbers are provided). The first was sent Wed., 9/20 at 1:09 p.m. and had the subject line "SF muni network". The 2nd was sent Thurs. 9/21 at 10:37 a.m. and had the subject line "SF muni network - 2nd request." I would be happy to forward these emails to you if you can provide your email address. For that matter, why not just email me directly so we can take this conversation off the message board?

Richard Martin
[email protected]
wlanner 12/5/2012 | 3:39:40 AM
re: Imbroglio by the Bay Corporate Communications 101 - contact the reporter if you aren't happy with something - you don't start posting on a message board.
phillipwolfe 12/5/2012 | 3:31:32 AM
re: Imbroglio by the Bay Hello Readers: After a long career at an electric utility, and city municipaly work, I got recruited and currently employed by a large manufacturer of wireless devices, point to point backhauls, point to multipoint mesh, radios, repeaters, end user devices, networks, wireless hybrid broadband over powerline and such.

I agree with Sacca that it will take a hybrid approach to solve the digitial divide problem. I also think the "last mile" issues will always be a challenge. It will take a combination of wireless and wireline technologies to make the business model work.

For example, the wireless signal may not penetrate the inner workings of a dense urban multi-unit dwelling. I encountered this problem and we plan to solve this by using hybrid broadband over powerline for multi-dwelling units. We also plan to install mesh radio technology whereby the incumbent city can take advantage of the mesh mobility aspects on 4.9 GHz instead of depending solely on 2.4 GHz 802.11 (x) radios. Finally, there are areas whereby wirless may not work unless you design the system ahead of time. Therefore, we simulate the wirleess design in our wireless planning software prior to real installation. For example, one of my clients spent 6 hours (!) trying to place only three access points! We did the same in about 45 minutes using planning software.

So yes, there are many things to consider.

P. Wolfe
joset01 12/5/2012 | 3:31:29 AM
re: Imbroglio by the Bay Interesting. But doesn't that method make the whole scheme a rather more costly undertaking?

phillipwolfe 12/5/2012 | 3:31:23 AM
re: Imbroglio by the Bay DJ - To answer your question on whether or not it makes the system more expensive. My intitial thought was "yes" but after some real work in the field I realized that many WiFI systems are built without the end-user in mind, nor the local wireless ecology, ie - buildings, Non Line of Sight charachteristics.

Picture a family or student in an apartment complex. Will the WiFi signal penetrate the building? Probably not. But you won't know until 1) you try it with a real access point/subscriber module, or 2) you model and simulate the system on software and account for for the dB levels, signal/noise ratio.

Then you deploy. But even then you might still need a wireline connection to overcome the "last mile" which is really from the curb to the inside. T-Mobile got smart and deployed Hot Spots at every Starbucks and Kinkos/FedEx in the nation. People gladly pay to get access to internet, check email, send messages, etc at 11 Mbps aggregate.

I think relying on WiFi alone will be part of the answer. It takes a hybrid approach to make it cost effective over the long term.

As a matter of fact, the new thinking is that OWA (Open Wirless Architecture) is the real future and everything we see now is still a transition. But it always is a transition with technology.

Which brings me to another point. I've always wondered why entities "purchase" wireless gear to deploy a system. Seems like a "lease" plan makes better sense. Why purchase a technology that is ever changing? Enterprises realized this with desktops and laptops and therefore lease all the time.

P. Wolfre

Sign In