This issue is on the agenda today at the FCC, and it was a major topic at the recent IEEE DySPAN. The technology is intriguing, and it is hard to argue with using otherwise unused spectrum for new services. Like some others (see, for example this item over at Broadband Consensus and some of Tom Hazlett's articles, like this one over at Ars Technica) I am coming to the conclusion that we're going about this all wrong.
Given that a small percentage of households receive over the air TV, shouldn't we really be discussing doing away with that technology, and encourage broadcasters to change their business model to being pure programming providers, instead of bundling programming with delivery techniques?
The reason this is important to discuss now is that many of the white spaces devices are optimized for the particulars of the TV spectrum. If this becomes unlicensed use, then the existing channelization becomes effectively locked in because it is difficult to coordinate disparate unlicensed users to transition to a new channelization regime. Is this in the best long term interest?