michaeln:

 

The main cost of providing fixed access services is trenching. The fibre or copper that you put in the ground is less than 10% of the cost of simply digging up the road/field etc..

 

As a very rough budgetary figure, that's about $100/m. If you are 10 metres from the exchange, that's $1,000 just for the trench. If you are 5km down the road, that's half a million dollars. Alas, economies of scale don't really help here. It's not a whole lot cheaper to trench in the country—sometimes it's more expensive. In any case, the greater distances outweigh any savings.

 

If there are 10,000 people at the end of that 5km then it's not too hard to make a case to spend the money. If there is one family, it's somewhat harder.

 

Of course, one you've built this fundamental infrastructure it lasts a very long time, but that still leaves you with the problem of coming up with the funds in the first place.

 

That's why UFB is only building to 80% of the country. It's not that the need isn't there, far from it.

 

Wireless offers a partial solution, but wireless has its own problems—the inverse-square law is not your friend. Neither are mountains, trees, mist, rain and bodies of open water (which particularly affect the high frequency bands you need to get high bandwidth).

 

I heard on the grapevine that in some parts of auckland, when trenching to put in UFB the field guys had to take into account tree roots and not disturbing the street view (Ponsonby? Parnell? yeah....).... meaning tree doctors had to be called in to protect exposes roots and so on.

Equally smashing the volcanic rock in Auckland used to be $500/m. Guess it's now gone up to $1000/m because inflation. 

Although, I expect Wellington Aerial would be a lot cheaper, even if new poles have required to elevate everything up another 2m..... :-)

Sigh, UFB in my road now slipping to Feb... which will become April.... which will become 2018. Hope docsis 3.1 is worth the noise and the modem does NOT come with 70's faux wood paneling choice...