Taking Hotspots To The Next Level

By Gerry Blackwell

July 18, 2002

Heavy-hitter Bechtel Telecom weighs in with its recipe for turning Wi-Fi hotspots into a mass market proposition: you might not want to hear this.

The embryonic WLAN hotspot market will never go mainstream until service providers can offer a consistent and easy user experience across all sites and deliver high reliability and quality of service.

This is not going to happen in today's over-hyped wild, wild Wi-Fi environment, says Jake MacLeod, principal vice president and chief technology officer at Frederick, MD-based Bechtel Telecommunications.

"Start-up hotspot companies are going to have a lot of difficulty getting through the obstacles to making [Wi-Fi public access] ubiquitous and seamless and high quality," says MacLeod.

This is probably not something the Boingos, HereUAres, WiFi Metros and Joltages of the world -- all would-be Wi-Fi hotspot kingpins -- want to hear. Indeed, they could be forgiven for viewing MacLeod's comments as a big business bashing the entrepreneurs.

Bechtel Telecom is certainly big business. It's part of giant Bechtel Corp., one of the world's largest EPC (engineering procurement and construction) firms -- 50,000 employees, $14.3 billion in revenues. Bechtel's bread and butter is multi-billion-dollar engineering and construction projects.

The telecom subsidiary's interests in Wi-Fi/802.11b-networks are multi-fold: it's a user, some of its clients are users and others are prospective service providers. The company sees broadband fixed wireless becoming a key technology in the telecom and construction sectors.

"Companies that win in [the Wi-Fi hotspot] arena will provide me with a uniform user experience," says MacLeod, "When I land in LaGuardia, the [Wi-Fi] network will automatically know who I am and how much I pay -- just as with cell phones."

The experience will be the same in New York as in Dallas as in Cincinnati, he adds. Each carrier will know his subscriber profile and he'll get one bill. "Who can do that now?" MacLeod asks. "Only a handful of companies have mature OSS's [Operations Support Systems] that can do it -- the ILECs and the major mobile carriers."

It's worth noting that the big telcos and mobile carriers are also among Bechtel Telecom's best customers. For example, it was recently been awarded master contract agreements from both AT&T Wireless and Cingular to build the GSM overlays on their existing TDMA networks, which will pave the way for 3G.

Don't discount the company's analysis of the Wi-Fi hotspot market, though, just because it has some fairly obvious biases. As Scott Semple, Bechtel's vice president of strategy, points out, Bechtel doesn't have any vested interest in seeing the market develop in any particular way. It just believes it inevitably must evolve in certain ways if it's to realize its potential. MacLeod and Semple make a good case for their view of it.

For now, MacLeod says, Wi-Fi hotspot subscribers typically have to re-configure their laptops and re-submit their credit card information wherever they go to get access. "It's a pain in the neck," he says. "People are not going to use it."

While it might be okay for the "10,000 or so early adopters who really need this kind of access," for the market to get to the point where there are millions of subscribers, it will require something much slicker.

It will also require much higher quality of service and reliability, and a little less hype. Semple notes that hotspot companies frequently talk about the 11Mbps of throughput in Wi-Fi networks. In fact, an estimated 80 percent of hotspots are connected to the Internet backbone via nothing more than a 600 to 800Kbps DSL or cable connection. This is hardly sufficient to support a large number of users, or even a few high-bandwidth users.

"As long as it's just you and me sitting there in the coffee shop, it's no big deal," Semple says. "But as soon as Jake comes along -- and he's a power user -- and you start getting the transfer of some massive files, from Paris to Singapore to Calgary, it goes down hill pretty quickly."

When subscribers at hotspots lose service or quality of service because of congestion, most will blame the company they subscribe with -- a Boingo or Joltage -- but the quality of service problems encountered at hotspots in these low-capital, "viral" business models will inevitably lie with the site owners.

Site owners have no idea about rigorously architecting networks for quality of service and reliability -- something Bechtel, with its background in big wireless projects, is all to well aware of, Semple says.

MacLeod and Semple also see Wi-Fi hotspots being yoked to and complementary to 2.5 and 3G mobile networks.

"I think you'll see zones where you have widespread broadband access," Semple says. "Those would be mainly in big metro centers. So out to the Dulles corridor in the Washington area, for example, you might have [contiguous] Wi-Fi hotspot [coverage]. But then as you go over towards Annapolis, you might go back to GPRS or GSM coverage."

In less densely populated areas or at places like golf courses or resorts, subscribers might have to go to a specific location to get broadband access. Yet in major population centers, they'll be able to get access anywhere. Semple points out that Grammercy Park in midtown Manhattan is now "wired for Wi-Fi" as are two parks in Pittsburgh, PA.

Certainly there is some indication that ILECs and mobile carriers are interested in Wi-Fi. Sprint has invested in Boingo. VoiceStream has said it aims to integrate Wi-Fi and GPRS networks in co-operation with sister company T-Mobile Wireless Broadband, which owns the assets of defunct MobileStar. Cingular is also interested in Wi-Fi.

"They're looking at it," MacLeod says of the ILECs and mobile carriers. "But we have not seen any direct evidence yet that they're going to jump into it with both feet. We just believe they will be the ones to bring stability and profitability to [the Wi-Fi hotspot market.]"

Bechtel believes the ultimate ideal is seamless integration between Wi-Fi and 2.5 or 3G coverage so that subscribers can maintain a connection while moving from one network to the other.

This could be done a couple of ways. One is with a Web of roaming agreements, with roaming partners using some kind of common back-office middleware for authentication, billing and reconciliation. Semple says Herndon VA-based Bonsai Networks' Gnorizo platform could serve this function.

More likely, though, is that ILECs and mobile carriers will buy up Wi-Fi companies and impose standards. "It's not that the phone companies are standing at the sidelines either waiting for these guys [Wi-Fi hotspot firms] to get big enough and then buy them or to fail and then buy them," Semple says.

However, it's clear to Bechtel at least that the phone companies, with their long history of managing customer relationships, are the logical ones to provide the kind of umbrella under which universal, seamless Wi-Fi/3G services could evolve.

The current environment, with lots of entrepreneurial start-ups carrying the ball, is perfectly natural, MacLeod and Semple point out, but history says it won't last. The same thing happened with cellular, the same thing happened with the telephone at the turn of the last century, the same thing happened with railways. Lots of little companies jumped in, often with incompatible, non-standard technology. Eventually the cream rose to the top.

"That's the wonderful thing about capitalism," Semple says. "There's always this settling out period. What's going to happen [with Wi-Fi hotspots] is that some companies will build out truly rigorous networks that focus on things like authentication, things like ease of billing. And at the end of the day, they'll be the ones who succeed."

Got a comment or question? Discuss it in the 802.11 Planet Forums

Comment and Contribute
(Maximum characters: 1200). You have
characters left.