The Martian NetDrive Wireless: 40 gigabytes of small, silent, 802.11b filesharing

The above is a paid, sponsored link. Email for more information.

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

Clay Shirky explains why 3G's mindset is permanently wrong, and Wi-Fi is enough by being nearly enough: Shirky explains how the misguided logic of airphones, those expensive seatback calling devices, is also at work in third-generation (3G) cellular data network marketing and deployment demonstrated in the confusion by telecom strategists between ubiquity and utility.

Shirky defines permanets as always available often expensive alternatives to nearlynets, which can be nearly ubiquitous, but are usually cheap and with less or no monopoly control. He uses this analysis to explain Iridium's failure: the people who most need good service already had cell phone access in the vast majority of places they needed, and you can't sell a bunch of expensive satellite-based phones to folks living in the outback. To put it more succinctly, Shirky writes, a permanet network doesn't have to be unused to fail. It simply has to be underused enough to be unprofitable.

Of course, this is Sky Dayton's argument for founding Boingo Wireless. Dayton's message was and is that Wi-Fi infrastructure is a separate business from selling access to the infrastructure. Infrastructure wants to be loaded -- and you're only making money as a physical deployer if your hardware and network are in use at a non-marginal level. This is why T-Mobile's failure to offer roaming or peering with other hot spot networks so far seems baffling.

Shirky explains the attractiveness of nearlynets as providing the right utility and price compared to the assurances of quality and ubiquity: The permanet strategy assumes that quality is the key driver of a new service, and permanet has the advantage of being good at every iteration. Nearlynet assumes that cheapness is the essential characteristic, and that users will forgo quality for a sufficient break in price.

(Interestingly, T-Mobile and Cometa have repeatedly stressed the quality of their Wi-Fi networks (existing and planned), especially compared to community networks which have near 100-percent utility because they're free, but can only have that utility when they're actually working.)

Why spend money per bit or per minute everywhere when you can get a flat rate most places? I can't be as articulate as Shirky, but I've been saying since late 2001 that if Wi-Fi hot spots cover 97 percent of a business traveler's need for access, why wouldn't that traveler wait a few minutes here or there to suck at the big pipe using a flat-rate account? The fallacy that Shirky examines is that even if travelers want access everywhere, they won't pay an arbitrarily high or metered price for it when they have cheaper, only slightly less convenient alternatives. I've said it before: business travelers are more concerned about pricing consistency and predicability than ubiquity. If that weren't true, Iridium might have succeeded.

By the end of 2003, some of my long-time predictions, delayed by market realities, will finally come true: all of the top 25 airports will have a reasonable amount of hot spot coverage; most hotels will have some, at least in public areas and conference rooms; virtually every block in an urban area will have at least one hot spot in a Starbucks, coffee shop, bookstore, or other gaterhing place. When you hit that level, where does 3G come in? As a marginal, expensive option.

In talking to early adopters like myself, Bluetooth plus a Sony Ericsson T68i phone on a GSM or GPRS network bridges the space between Wi-Fi hot spots. I willingly use a 9600 bps GSM connection -- 2G at that! -- in locations where hot spot service isn't available or is expensive compared to my utility. If I need to cheap email for five minutes in an airport that offers $7.00 per day access, I'll use my GSM service. 3G speeds might be nice, but not at a metered price.

(Cingular meters GSM by minutes, and I have a big pool; their GPRS service is by megabyte, and thus not worthwhile even at the higher speeds. Incidentally, Shirky lumps GPRS into 3G, but GPRS is a 2.5G technology; 3G is a catch-all phrase for several newer technologies.)

Shirky writes, The lesson of nearlynet is that connectivity is rarely an all or nothing proposition, much as would-be monopolists might like it to be.

Interestingly, this could be the corporate anthem for Cometa. But remember: Cometa's model is to sell access to their network to those would-be monopolists, the cellular telephone operators trying to sort out 2.5G and 3G data services.