Here we go again, this time from Newsweek. We have been hit with the collective realization that for-pay citywide WiFi networks are not getting nearly enough subscribers to support the costs, and that QoS issues when covering large geographic areas and going in doors are driving up costs and undermining the value proposition.
As Rolla Hoff, the new CEO of Earthlink puts it, "The Wi-Fi business as currently constructed will not provide a return." Those planning citywide deployments are asking cities to make upfront commitments on purchasing the services that the network would offer. That's a big change from three years ago, when WISPs were offering municipalities concession fees and revenue shares for the opportunity to build.
What those who planned the large scale deployments in San Francisco, Philly, etc failed to understand about WiFi is this -- it is a local technology, by FCC regulation and given where 2.4 GHz operates. It broadcasts at low power, so the range from any one access point will be quite limited (300 ft). 2.4 GHz doesn't penetrate walls too well, or foliage. Creating broad coverage through a mesh of access points was touted as cost-effective, but throughput, especially around video, is a big issue. If you have to have a connection back to the internet every two hops, as opposed to every four, your costs just went way up.
As the article points out, the local duopoly -- the cable company, the phone company - have the infrastructure and the deep pockets to combat an upstart Muni-WISP. But as we argued here before, if you propose to set up a third provider citywide to go toe-to-toe with the incumbents, you will lose because WiFi can only do so much as low power, open spectrum, and because setting up all the marketing, tech support and customer service functions associated with being a muni-WISP requires enormous amounts of capital.
WiFi is a local technology. Stay local. Light up Main St. Create a wireless environment that supports local businesses and mobile users. From there, something bolder may come. It will still take a while to get to the right device density, for pervasive or ubiquitous computing to truly take hold. Get it right with the user first. The networks will come in response to that.
Leave a comment