As Stephen Conroy interrogated the incoming NBN Co chief Ziggy Switkowski in last week's Senate hearing into the network's rollout, it became increasingly clear that politics is getting in the way of good policy.
The Federal Parliament's Select Committee on the National Broadband Network latest proceedings, which concluded Friday, focused on the state of Telstra's copper network, the use of alternative technologies and the expansion of the NBN's footprint.
But it was hard not to be reminded of Monty Python's Dead Parrot Sketch -- particularly when Dr Switkowski and Stephen Conroy sparred over the copper. Is copper dead? Is it resting? Will it voom if you put four million volts through it?
At the same time, journalist David Braue of Fairfax Media had a scoop after obtaining a draft analysis from a source in NBN Co, which was apparently used as part of the content of the "blue book" briefing for the incoming government.
Technical problems experienced in the Senate hearing -- specifically the failure of a video link with Senator Scott Ludlum -- reflected the frustration all of us experience with the current condition of broadband in Australia. It was a fitting allegory for the state of domestic internet services.
Dr Switkowski was in the hot seat on Friday, making the clear point that he is focused on expanding the footprint of the NBN rollout to get it back on track, and this is how it should be.
But it is unfortunate that the politics of the NBN is sucking the oxygen out of the task of getting the job done, and the blame lies in equal measure with both sides of politics.
A short history
If you have a long memory, by communications technologies standards, you might recall that the 2007 concept of the NBN was for a Fibre-to-the-Node network proposed by the then opposition ALP under Kevin Rudd.
The Fibre-to-the-Node option was abandoned in favour of a Fibre-to-the-Premises model, in part because it was the best long-term technical solution, but mostly as a means of forcing Telstra's structural separation between wholesale and retail services.
The Fibre-to-the-Node approach would have involved buying back Telstra's copper network, while Fibre-to-the-Premises turned out to be cheaper by rolling out fibre and compensating Telstra as it decommissioned the copper network.
The copper network was widely acknowledged to be in a poor state of repair and unable to keep up with conservative estimates of demand for high-speed internet.
The situation was complicated by the compromises made in reaching an agreement to form government with independents in 2010, resulting in a shift of priority to regional areas such as Armidale ahead of urban areas.
The result was that the initial stages of the NBN rollout were much more about feel-good politics than the rational commercial reality of getting the customers most engaged and willing to pay connected to the high-speed fibre network. That approach would have generated revenue and built up demand.
The aggressive rollout schedule was also a political folly. It was unrealistic, with even more delays caused by protracted negotiations with Telstra and the discovery that asbestos in pits had not been remediated by Telstra.
It became a political hand grenade because of the emphasis on the number of homes connected at each milestone, rather than the proper emphasis on getting the foundations right first.
Of most concern is that the NBN's rollout effectively froze investment in alternative technologies by competing networks. Why would you invest in high speed ADSL when you only have three years for a return on that investment?
As many homes, as quickly as possible
In this context, the Senate Committee heard from Dr Switkowski that the emphasis is now on getting something -- anything — rolled out to pass as many homes as possible as quickly as possible.
Suddenly the copper isn't dead, according to Dr Switkowski.
But the Communications, Electrical, Plumbing Union (CEPU) says the condition of the copper is "an absolute disgrace".
Who do we believe? Both sides of this argument have a political agenda, so the truth, as always, probably lies at some unknown point in between.
Dr Switkowski's contention that ADSL working up to 10 megabits per second is out there, working, and meeting public demand on the existing copper is missing the point.
The average downlink internet speed in Australia was recently reported as around 4.5 megabits per second, and there is a very robust argument that says there will be consumer demand for this to exceed ADSL's 10 megabits per second capability by 2015.
Upload speeds remain cripplingly low, and there is a disconnect in the minds of policy makers between what customers are putting up with now versus what they would be happy with to meet their telecommunications requirements.
The current government's policy to support competing technologies, by competing providers, simply creates competition where there is the most demand.
The Senate Committee heard evidence on Thursday that this, in addition to a shift to lower speed plans and an inability to support services such as multicasting, would dig deep into NBN Co's wholesale revenue projections.
Where to from here? Both policy and action needs to shift toward these following priorities:
- Accepting that Fibre to the Premises should be the goal by 2025, and taking action to roll it out with a realistic and achievable fifteen year time frame. This will mean having to shift the political discourse away from getting fibre "sooner" and instead work on a properly managed timetable to achieve the end goal.
- Meanwhile, demand moves on, so there needs to be an interim and incremental series of solutions which will deliver an adequate job. Fixating on Fibre-to-the-Node is not the answer. Using Fibre-to-the-Node where it makes commercial sense is a much better approach, alongside other technologies. This means shifting the political discourse away from "faster" and towards "what we need, when we need it, where we need it, at the scale we need".
- Fixating on a total capitalisation figure makes no sense -- a successful network with high take-up by customers will, necessarily, need to invest in the network to continue to upgrade profitable services. The rhetoric needs to step away from "cheaper" and focus on "cost-effective".
A previous article pointed out that "Optical fibre is the only known viable technology beyond 2025. The only justification for considering anything else in the meantime is to buy us time".
It is time to buy us time. Is it too much to ask the politicians to tone down the rhetoric and let NBN Co get on with the job?
Matthew Sorell is Senior Lecturer in the School of Electrical and Electronic Engineering at the University of Adelaide. He does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.