[AusNOG] AAPT / TPG Merge NWB

[Jake Anderson]

On 05/06/14 18:15, Paul Gear wrote:
> On 06/05/2014 05:45 PM, Jake Anderson wrote:
>> On 05/06/14 08:00, Curtis Bayne wrote:
>>
>>> Regardless of any theological thought experiment, the pragmatic 
>>> reality is that the only thing that has delivered reasonable 
>>> outcomes to consumers in Australia is infrastructure competition.
>>>
>> Yes, because there is so much choice over just whose copper network 
>> you are going to connect your dslam to, what with all the 
>> infrastructure competition out there. 
>
> There seems to be an assumption there that consumers using the copper 
> network have a reasonable outcome.  I certainly wouldn't agree with 
> that assumption.  The unreliability of the copper network is the bane 
> of my existence at the moment, and the outcome of a previous thread on 
> the topic 
> (http://lists.ausnog.net/pipermail/ausnog/2014-March/023161.html) was 
> basically that if we're on copper, we're stuck with whatever outcome 
> (reasonable or not) that Telstra deigns to bless us with. Last mile 
> infrastructure competition would be a huge bonus for Australian 
> consumers.
>
> Paul
It sure would, but in the past getting on to 20 years it hasn't happened.
The fixed cost of doing the last mile per potential customer vs the 
diminishing returns as competition happens means its a market ill suited 
for dynamic competition.
IE it costs you about the same amount to roll a network out to cover a 
population, regardless of how many customers you actually achieve. This 
is different to a mobile network for a first order approximation.
Unless you can get 3 players into a market it tends to not work well as 
a free market, look at the local monopolies and duopolies in America, it 
just doesn't really work. Google coming in has given them a shake up, 
but that strikes me as more of an act of god than the traditional free 
market at work.

That's why I liked the NBN, having one public entity own the glass in 
the ground that doesn't also sell retail services seems to be the only 
way to get progress outside very built up areas.
The way they have done everything *after* the glass in the ground is far 
more debatable imho, Ideally I'd like it if nbnco somehow sold bare 
fibres to RSP's with provision for them to co-operate and offer multiple 
services over one fibre.
I can't however come up with any really "good" ways of doing so, so I'll 
accept the GPON network unless somebody can suggest something better? 
Perhaps let RSP's put gear into FSAMs, run their own splitters in the 
FDH (you must have X% of customers in a FDH before you are allowed to do 
that though I'd suggest otherwise you wind up with assloads of splitters)
Of course the CVC charge is a massive stumbling block for doing anything 
creative offering advanced services over that nice last mile network.

The MTM is of course pretty much worse than doing nothing from any angle 
I can see, spending about the same money and about the same time to wind 
up with something you need to start replacing as soon as its done, there 
is no "asset" at the end of it, and the same competition environment as 
before. As an end user its not going to get the big speeds I want for 
smaller business customers. With my operator hat on I'm not going to be 
able to sell the big services dripping with cream that do wonders for 
the bottom line to those business customers.
They would be better served offering industry some small sum (say 4.7$bn 
;->) and letting the big players fill the node cabinets however they 
feel, at least then it'll be cheap when they eventually need to buy it out.