One of the biggest challenges NBN Co faces as the rollout of the National Broadband Network continues is learning to manage success, NBN Co non-executive director Simon Hackett told AusNOG 2014 yesterday.
"The base challenge that NBN Co has over the next three to five years is ... surviving, in my opinion, the transition from managing failure, which it has done a lot of, to managing success," the founder of ISP Internode said during an extensive keynote which also touched on his motivations for joining NBN Co's board last year.
Update: Simon Hackett has posted his presentation online.
"NBN Co is having to get used to the idea that as the network starts to finally get built at a decent pace, which is starting to happen now, a whole lot of things that should have happened now, in terms of scaling the organisation up, have not occurred and so they now need to," Hackett said.
"The whole organisation needs to think about end-to-end process management in a way it never has before. You can't hold millions of customers together on spreadsheets. You have to actually automate the hell out of systems. There's an awful lot of systems at NBN Co, but right now there's an awful lot of siloing between them."
"Today the largest complaint you get from an ISP interfacing into NBN Co is that the systems don't work," Hackett said.
"That you ask to provision a connection and it doesn't necessarily happen on time. All the things that we loved to hate about dealing with Telstra as a wholesale partner, NBN Co has managed to achieve those same things in three years that Telstra took decades to manage."
"[NBN Co's] culture has clearly suffered from its own inability to get results out the door in the previous electoral cycle. Nobody within the organisation is happy about that and there's a whole rebuilding process that's having to happen... giving people the confidence that things are capable of being a little bit different," Hackett said.Read more:Broadband projections fail reality test: Rod Tucker
Fibre on demand
Fibre on demand "it is very much still in the modelling and it is still very much being worked on," Hackett told AusNOG 2014.
Malcolm Turnbull, now the communications minister, early last year raised as a possibility having residents and businesses in areas primarily served by fibre-to-the-node able to pay to have fibre extended to their premises.
In NBN Co's strategic review of the National Broadband Network fixed line rollout, fibre on demand was listed as a policy issue that was yet to be resolved.
The interim report of the Senate's NBN committee, dismissed the idea of fibre on demand, saying it would "be too expensive for many small businesses and will entrench widespread inequality in access to infrastructure for Australian households and small businesses".
The difficult question "is not a technology one," Hackett said. "It's what do you charge to do it."
"My personal view about that is a pragmatic one, that what will probably work pragmatically is the same thing the power companies do to a great extent — if you want to bring power on to a rural property and it's not there yet — which is to say the cost is x dollars to get off the main bearer to your farmhouse.
"Pay us x dollars — whatever that is - and we'll connect you up — same thing here right? You can see a world where you would wind up with a quote generated some how to get from the end of your street to your house. It costs x and that's what you'd pay. And if you're prepared to pay you'd get a [fibre-to-the-]premises service."
Hackett acknowledged that fibre on demand has had a "chequered history".
"British Telecom do this in the UK, and depending on whose reports your read or whose opinions you listen to, it's either been a roaring success or an abject failure. Either way the point is it's intended to be a part of the delivery model, so the choice exists for people to do that if they want to."
As the NBN rollout accelerates, NBN Co faces a "heck of a lot of challenges", including adding support for new technologies as part of the government-mandated shift to a multi-technology mix and increasing the organisation's transparency
"It's got to learn walk, chew gum and hum tunes all at once," Hackett said. "And it is daunting for the organisation, because it was born as a one-trick pony — it was born as an FTTP [fibre-to-the-premises] company, so the idea of rolling out a few more types of networks — some people think that's easy, some people think that's hard – they don't have a choice, that's the job."
The shift away from rolling out only FTTP will ultimately help create an organisation that's more flexible in its capabilities, Hackett said. This is important because NBN Co has already had to operate under two different policy settings, and future governments may well end up mandating further changes.
"If you picture a world where NBN Co has hopefully got the hang of deploying all these technologies, if future governments said, 'That's great, but turn off all those other things and just go back to doing fibre', that's a pretty easy thing to do because that wheel's already turning.
"It's not hard to make that transition back in the other direction. I'm not predicting that it will or won't happen. It's important to appreciate, again, that inside NBN Co none of this is about policy – it's all about doing what the government tells us...."
"I think it needs to get better at changing with government whim, because over the long-term history the network, the long-term future of the network, it will get its policy changed lots of times," Hackett said.
NBN Co is building an internal model to assess which technology "makes the most economic and pragmatic sense in each location in Australia."
"The point of that model is not to run it once and spend the next six or eight years building a network," Hackett said.
"It's to run it every year to produce a revised forward view for the next 12-18 months about where you're going to build. This is the important point – as technologies get cheaper to build, as fibre gets cheaper to build, it is rational to assume that the amount of fibre built will [rise].
"If something else turns up that's way cheaper, it's rational to assume that more of that will get built. So there will be a map produced internally of what everyone's going to get, but the only thing I'm going to tell you about that map is that it won't be correct beyond about a 12-18 month cycle time because, again, we will keep re-evaluating that as we go on..."
Follow Rohan on Twitter: @rohan_p