Oh, the NBN..

..and seems to work reasonably well. Better than Mr Turnbull’s
alternative.
The real benefit, apart from PayTV, was the fact that the network
was capable of providing 100 Mb/s internet services from day one
when the planets were aligned, though the more users that were added
to your hub meant that speeds declined as it is a shared service.
But with the latest DOCSIS 3.1 technology, which is now available
and tested in Australia, 1Gbp/s will be attainable. When is the
interesting question?
The other minor bit of fun about PayTVwas that Optus and Telstra/Foxtel
were bidding for the same content at much the same time. The American
content providers, some of the sharpest in the world, recognised
a golden opportunity locking Australians into world record cost
contracts.
But I, as usual, wander.
There are only three components of the internet that we care about:
speed, data and reliability. Given there is nothing faster than
the speed of light fibre optic cable is technically the best option
if you want speed and therefore the ability to download –
and upload – lots of data. You just have to get the optic
cable to the premises which is expensive given the two options;
digging trenches or renting Telstra’s ducts, pipes and pits.

Fibre is intrinsically reliable unless a contractor digs it up.
The technical term for this is Backhoe Fade.
And the lucky million or so users who got fibre are not complaining
at all. Not unlike new customers where a quarter of all customers
are very disgruntled.
Mr Turnbull’s Fibre to the Node is a very second rate option.
All it does is break the exchange into smaller components and move
them closer to the customer. And the Node boxes aren’t always
going to be small as all will have to be powered and many will have
to contain batteries for fail-safe 48 volt power so the phones still
work. Does that fix your copper connection to the node or to your
house? Absolutely not. It was a solution that Telstra floated as
a $15 billion proposition to the government many years ago as a
way of preserving its near monopoly. There was some mischief in
the idea in that it actually cut off most of its competitor’s
customer access as well.
What about wireless? I mean we get excellent data speeds out and
into our mobiles phones don’t we? Well yes and no. There are
a couple of drawbacks the first being that they amount of useable
spectrum is relatively limited and much of it spoken for by hundreds
of useful and needed services. And the amount of data that you put
through a mobiles base station is shared and limited by the number
of antenna and directionality and the overlap with other base stations.
The number of high speed data channels is limited per base station
and the service degrades with the number of users. You have to put
more and more mobiles cells in if you want coverage and speed. The
capacity of radio technologies is limited by available spectrum
and the cost of spectrum will determine the economics of wireless
systems
One new slice of spectrum is being touted for 5G is very high frequencies
– millimetre wave radio at 25Ghz+. Capable of carrying much
more data than the lower frequencies now used. Having been part
of a study of this spectrum someone’s mischievous advice to
management (was that me?) was that you had better sell a chainsaw
with each service if it was going to be deployed in the bush. Easily
upset by foliage, highly directional and didn’t like the rain.

As for the NBN business model? Its silliness will keep for another
day.
So remember, if you want good throughput insist on plenty of fibre.