To build the new National Broadband Network in a cost effective way that is future scalable will be a significant design challenge. Details on how all this is going to be done are short but it is obvious that getting the network architecture right, putting in an appropriate regulatory framework and also managing the roll-out in a pure construction sense will be absolutely critical — put simply, good designs will mean billions saved over poor ones.
I’m on the road at the moment and so haven’t not had much time to consider this but, for the moment, I am going to post readings as they become relevant for those interested in priming themselves up to understand these issues.
The first reading is from Slater and Wu, whose paper I referred to in my November 2008, AFR piece. Their ‘homes with tails’ idea is to have an open backbone network but to give customers the opportunity to build and own the connection from the street to their home. I have had similar proposals before.
Core Economics 
I like the idea but I think it would be good to stage.
1st stage would be to build out a conduit system that anyone can have access to. This could begin very quickly as it is laying piping between points something that needs less planning. The conduits are the largest cost in networks.
2nd stage would be to lay the fibre into the conduits. The design of the system can happen as the conduits are being constructed so there is less delay in getting started.
The real decision is whether to use direct fibre (one fibre per customer) or shared (like the current cable system were a line is shared among multiple customers). The direct is future proofed.
The eventual bandwidth is determined by the equipment at the ends rather than the fibre in between. Some careful planning can mean that by the end of the 3 years 1Gbps or greater can be delivered.
LikeLike