Right on the money with femtocells and a strong justification post-facto for BT FON/Openzone as well.
What wireless can do is behave as a capex catalyst, reducing the upfront funding required to get service operational whilst aggregating revenues to pay for FttH over time - the NextGenUs model looks for a 6 to 10 year incubation journey for deeply-rural community interest NGA to become fully FiWi.
Given the luxury of "it costs what it costs" capital funding then FttH with a complimentary wireless overlay is the way to go for sure.
What is the exact reference you are looking for Optical Girl?
The focus on fixed line solutions is absolutely the right one.
Wireless broadband is a complement not a substitute for fixed acess.
Wireless networks require high-capacity backhaul, and the more people using it, the smaller the cell size, and the more base stations/wireless access points are needed. Ultimately, I think that a high-capacity national wireless broadband network would look a lot like a fibre to the home network where everyone has a Wi-Fi router -- there would be fibre connections to millions of wireless access points, which would have to be very close to the end user.
Why do you think wireless operators are promoting femtocells -- a device that routes mobile phone calls over your broadband connection?
Installing a decent wireless network is slightly cheaper than installing FTTH, but still expensive. But adding wireless capabilities on top of an FTTH network rollout is an incremental cost. FTTH networks can easily include base stations, which tend to be located on top of buildings. I'm fairly sure that the calculations have been done on this, but don't have the reference. Does anyone know what it is?