Is WiFi part of your product strategy? Let Blue Mug handle the details: integration, interop, roaming, power, performance.

Blow your mind wide open at Supernova 2002, a 2-day conference Dec. 9-10 on decentralization, and the fundamental questions facing software, communications, and media.

The above are paid, sponsored links. Contact us for more information.

Subscribe to essays from this site via email. Email to subscribe, or sign up via your Yahoo account.

Hot Spots and Fragmentation. Both of today's keynotes at Jupitermedia's 802.11 Planet conference dealt with industry fragmentation and hot spots.

Dennis Eaton, the chariman of the Wi-Fi Alliance, talked about how Wi-Fi had united the industry through rigorous testing, and warned against releasing hardware based on draft standards. He also announced a new alliance initiative for marketing hot spots under the name Wi-Fi Zones. This is clearly the precursor of formulating hot spot standards which they could then use as an assurance of levels of service quality.

Sky Dayton, Earthlink and Boingo Wireless founder, pushed a message of fragmentation across two dimensions: fragmentation across networks is bad, resulting in users needing to deal with many settings, many accounts; fragmentation across elements of running hot spots is good (physical venues, infrastructure, aggregation, and end-user branded service). The former fragmentation reduces uptake and increases complexity, while the latter, he says, sorts out the right tasks to the right companies, letting them focus on what's important.

Sky gave out a few pieces of news today. First, the Macintosh Boingo client will ship in the first quarter of 2003. Second, they are focusing on embedding their hot spot authentication integration software into both higher-end and consumer-level access points, allowing a purchaser of standard equipment to become a hot spot with a minimal amount of work. Current Hot Spot in a Box offerings are as low as $500; he expects it to drop to $300 in the near future.

Third, Sky revised his estimate of 4,000 hot spots in their aggregated network by the end of 2002 made earlier this year -- he didn't disavow it explicitly, but they have about 900 hot spots at the moment. But he pointed out that they have over 100,000 hot spots in the pipeline from interested parties who need to take the next step. He estimated 5,000 to 10,000 hot spots during 2003.

The focus on hot spots is interesting in the middle of all the rest of the talk about security at the conference -- but it makes sense. Eaton talked in some depth about WPA (Wi-Fi Protected Access), but WPA essentially solves the near-term security problem. Boingo already offers a security solution: a VPN out to their network operations center as a built-in feature (currently included for free) of their client software.

Raw Notes

Day Two from 802.11 Planet in Santa Clara. I'm sick and didn't sleep well, so notes will be in shorter supply than usual.

Dennis Eaton, chairman of the Wi-Fi Alliance and a marketing director at Intersil, keynoted this morning.

After presenting quite a bit of good background on the Wi-Fi Alliance history and goals, he identified 10 top trends for Wi-Fi over the next year:

10. Lower power Wi-Fi chips: "embedded in all sorts of devices that it isn't today" -- "great stuff, but it really sucks my battery down" is common comment

Wi-Fi will find its way into "communications devices and handheld devices"

9. More multimode products: "Wi-Fi and some other technology," such as Bluetooth, GPRS, 3G, high-speed WAN tech with LAN tech

8. Public access will start to mature globally. "It's been pretty popular in the US, the US probably got an early start in this market, NA did, but we've seen a lot of big announcements from other parts of the world." Asia: Japan, South Korea

7. Even higher Wi-Fi penetration in laptops. Increase in cert already for mini-PCI. Wi-Fi standard, not optional feature

6. More desktop computers will use Wi-Fi. As broadband comes into the home, esp.

5. Wi-Fi will be used in more smaller devices.

4. Dual-band products will ship in volume. "Dual-b products will be very important to many people, espially in the enterprise." Laptop vendors looking for universal solution. Want customers to have wireless connection without concern to infrastructure

3. CE (consumer electronics) devices will start to appear in 2Q03. "Seen demonstrations of CE devices." Interest in manufacturers.

2. WPA will be widely adopted in the industry. "Broad support for it when introduced in October."

1. Analysts seem to agree: Wi-Fi will grow at about 30% CAGR through 2006. "We firmly believe that. We believe there's potential for it to do better than that."

He also presented some of what he sees as challenges for next year.

Simplifying product setup: "Somewhat frustrating at times to products to work." Big initiatives by larger companies already, but "industry as a whole is going to have to address simplification of product setup."

Common nomenclature. Too many names for the same things.

Maintain realistic product performance claims: 11 Mbps is data rate, but throughput is lower than that. 300-500 feet range indoors doesn't match with what consumers find. "It's important that we be realistic about claiming what our products" can do.

Prestandard product "interoperability" is bad for the customer and industry -- a painful lesson

"If you release a product before it's ready and before all the interoperability issues among the multiple vendors have been worked out can lead to problems."

Like Bluetooth -- lack of coordination slowed adoption of "wonderful" technology

Resolving roaming and settlement for public access

"One of the most exciting things happening right now in the Wi-Fi industry is public access"

Education for the retail channel to help consumers

Meet demand

Potential surprises
Wi-Fi will be an asset to 3G rollout
Public access matures globally
Wi-Fi and broadband to the home will accelerate adoption of both technologies by consumers

Initiatives
Lowering cost to deliver Wi-Fi certified products
Aggressive marketing to grow industry worldwide
More certifications for WPA, 802.11g, 802.11e, 802.11h (g almost certainly; e, maybe; h, probably)
Training and initiatives to retail channel
More initiatives in the public access space

Eaton announced that a program kicking off in December would brand hot spots as Wi-Fi Zone. The first zone involves promotions for both members and non-members. (I would anticipate that future phases would use the recommendations of the alliance's WISP committee for service guarantees to receive branding.)

Analyst session

I substituted as moderated for a short panel with four analysts of enterprise and consumer wireless technology, all from major firms. I'll distill what they said later, but it was remarkable how in agreement they all were.

Sky Dayton's afternoon keynote

Alan Meckler comes out to introduce Sky Dayton. At Internet World several years ago, Meckler says, "this kid came in…about 22 years old…with four or five adults." "I'm really going to be a power in this field, and I should be a keynote speaker at Internet world.

"I didn't take him up on the offer" but had the chance in this brand new field. "Sky didn't hold it against me…he sure was correct."

Sky: "The reason I had all those adults with me at the time is that I couldn't rent a car. They drove me around."

A whole new business. Alan was one of the first to hop on wireline Internet in the 1990s.

Talk overview: Ubiquitous. But challenges. How we get there.

Wi-Fi's achieved escape velocity. TCP/IP of wireless. Everyone works on and gets benefits of everyone else's work. Consumer adoption price curve. Price reduction pushed adoption.

Public access: Hot Spot Operators (HSOs). Listed many in business. Targeted airports, hotels, cafes, campuses, convention centers. Grassroots side: barrier so low.

Visiting Aspen and used early version of sniffing software. Opened software in hotel and saw three networks. Connects to AP that's named with his name and phone number. Calls the guy up. Jim Selby.

Sky meets with Jim and by the time he returns to his hotel room he has a message from a guy running one of the other networks trashing the first guy and wanting to meet with Sky.

All previous attempts to roll out wireless used proprietary technology, like Ricochet.

"Wi-Fi offers a price per bit that 3G will never be able to touch." Free spectrum, cheap equipment for hot spot. Obscenely low barrier to entry to Wi-Fi hot spots.

Everywhere: two or more signals. He was in NYC, running the new Pocket Boingo software. While in a cab through New York, he saw signals constantly.

36M business travelers in US; 27M carry laptops. Talked about survey from last year that they conducted for market research: most business travelers they spoke to want high-speed service and will pay for it.

"it's becoming incredibly inexpensive to add Wi-Fi to products." Notes that Dell will include Wi-Fi radios by 2003 with all laptops. At Comdex, HP announced iPaq with built-in Wi-Fi. (Editor's note: Just bought one and have it with me.)

Will show up in cell phones, automobiles. Pull into service station and download maps, new information. GameBoy: not untethered, but play against anyone in a hot spot.

Anything with a battery that uses information: calculator, MP3 player, a watch.

Challenges to mass adoption

Ubiquity. 3,000 intentional hot spots; 1M potential locations. 212 conf centers, 3,032 train stations, 5,352 airports, 53,500 hotels, 72K business centers, 202K gas stations, 480K restaurants, cafes, and bars; 1.1M retail stores.

Fragmentation. Short range. Each venue a different hot spot. Looks like early ATM/cellular days. "You would go to a different suburb and your cell phone would stop working in the early days."

"We need to get to unified roaming as an industry." Multiple operators equals multiple accounts.

Hard to use. Invisible. Lights dimmed unintentionally: even when it's dark, the networks are hard to see. SSID: "I hate this word. Users should never see this word. Service Set Identifier? What is that?" Really hard for users.

Lack of focus. One company can't do it all. Short range means too many spots. Hard to deploy and service users.

From here to there

When Earthlink started, industry chaotic. No one knew who their competitors were.

(Editor's notes: Some of these slides are modified but essentially from their original press briefing a year ago -- the model hasn't changed for them, and their message is the same.)

ISP segmentation: physical (wires: Sprint, AT&T), network (IP backhauls like UUNet), and brands (end users: like AOL, Earthlink, MSN). In early days, companies tried to do the whole thing, like Netcom.

Each company ultimately focused primary energy on particular layer. Competition within layers; cooperation across layers.

Hot Spot Industry segmentation: venues (locations or real estate owners like hotels, bookstores, etc.); HSO (networks: Wayport, NetNearU, T-Mobile, etc.). HSOs contract with venues to build hot spots.

Aggregator layer on top of that. Take fragmented layers and aggregate them into a single service. Boingo, GRIC, iPass. Boingo only one focused on wireless, largest aggregated network.

Aggregators offers to end users as the brand: Boingo, T-Mobile, Earthlink, Sprint, AT&T.

"property owners who are working with hot spot operators who specialize in installing and running hot spot networks" then partner with aggregators which offer service to brands.

Lot of opportunity for smaller companies to get in. Room for one or two major aggregators. Both brand and HSOs want to work with companies that have economy of scale.

No HSO will control more than 10 percent of underlying footprint. Short range, low barrier to entry, venue fragmentation.

Boingo: single network plus great service

Boingo client: available for PocketPC and Windows. Mac client coming in Q1 2003

Demos clients. (Lots of active networks in the room.)

900 hot spots. 32 airports, 400 hotels, 200 cafes.

"Our pipeline today is over 100,000 locations. These are not locations that are deployed yet, but these are locations people are looking to deploy in the next 12 to 24 months." 5000 to 10000 could be in Boingo's network by the end of 2003.

Boingo Hot Spot in a Box (tm); $500 now (from Colubris) but will drop to under $300 and will be a feature built into consumer APs that you can just enable.

"There are very legitimate policy" considerations that have to overcome in terms of the upstream provider. But economics could drive revenue that would make it worthwhile. "Every end point of the Internet: attached to it is an access point that is radiating broadband."

Boingo Ready equipment: turnkey for setup. Announcing today: Colubris, Nomadix (using their equipment at the show), Vernier.

What's the value of roaming to hot spot operators? With Boingo, 500K Wi-Fi cards each month come bundled with Boingo software that installed when drivers are installed.

HP includes Boingo bundled on half of laptops they sell.

Carrier and ISP partners. Earthlink 5M customers. You can buy Boingo service from Earthlink today. Fiberlink: targets Fortune 500 companies.

Incremental traffic and revenue to HSOs that join Boingo-like network. Pie will be increasing for everyone in the industry -- market is completely untapped. Roaming plays important role to drive more users and adoption.

Hot spot success factors

Make hot spots highly visible: put it out on a counter. don't hide it.

Got to get easier to find and connect networks. Boingo software is best of breed.

Universal roaming is essential.

Future

Technology agnostic: b, g, a, doesn't matter. But 2.5G and 3G services will be available in dual-mode in mid-2003 for seamless connectivity.