Interview: Hugh Bradlow of Telstra

Published on:

Services will be the next transformation for telcos says Telstra CTO

Telstra CTO Hugh Bradlow looks forward to getting beyond the point at which he worries about the last-mile network technology into customers’ homes. He wants Telstra to be a services company

Hugh Bradlow: Take away the worry about the network and you
force us to think about being service providers 
Australia’s national broadband network is the best thing that’s happened to the country’s incumbent operator, Telstra. Take a deep breath when you read that, because the person saying it is Telstra’s chief technology officer, Hugh Bradlow.
Given that Telstra’s former CEO, Sol Trujillo, ran a running battle with the Australian government while the NBN project was being developed, this is a surprising turn. But these days Telstra, under CEO David Thodey, takes a very different attitude — and Bradlow, who has been with the company for 15 years, is clearly in agreement. (See GTB interviews with Trujillo here and with Thodey here)
“I believe in the economic benefit of high-speed broadband,” he says. “It will be an incredible legacy for our kids — but, like any large engineering project, it’s complicated.”
The sort of network that the Australian government is planning “would be a basis for refocusing the economy”, he says. “For Telstra it’s the best thing that’s happened to us.” Though he admits: “It may not seem like it at the time.”
Why is it so good? Because, says Bradlow, it allows Telstra to stop thinking about access networks. Fibre running into every Australian home will take away from operators the need to worry about what sort of broadband to install and how to upgrade it. “We can start thinking about the services: home and business services, such as health and education.”
His eyes gleam with the prospects of what telecommunications can contribute to the economy. For example, Telstra’s home city of Melbourne is looking at what it will need if it expands — as is expected — from a population of 3.5 million to five million in 20 years.
The roadbuilders are looking eagerly at the work they will have to do — but, points out Bradlow, “we need to make tough decisions about fewer roads and better occupancy of cars”. Telecoms “gives you the ability to schedule” and to optimise transport.
And “20 years is a long time in technology”, he smiles, thinking of Moore’s law, devised by Gordon Moore, one of the founders of Intel. This notes that the number of devices on a microchip doubles every two years: Moore formulated his observation in 1965, based on data back to 1958, and the rule holds good today.
“We’re talking about a 4,000-fold Moore’s law improvement in technology,” says Bradlow of what we can expect in the next 20 years. Instead of calculating how many more hospital beds they will need in Melbourne by 2030, politicians should start thinking of telehealth services.
“In Australia we built an 800-bed hospital for $800 million — that’s $1 million a bed.” He shakes his head: “The NBN will be good for Australians. A lot of opportunity will come from it.”
Software and services
The future is about software and services but in the past “the main concerns have been about the network”. Take away the worry about the network “and you force us to think about being service providers”, says Bradlow.
It’s a revealing attitude for someone who is an out-and-out technologist: a graduate in electrical engineering who did research into nuclear physics and later became a professor of computer engineering.
The technology of shared networks is making fixed network operators change the way they work. “We’ve got to be more like a shopping centre,” he says. An operator has certain features such as billing and quality of service and “stuff round the cloud”, he notes, “but there is a non-exclusive ecosystem because you have Google and Amazon and they’re not going away”.
But that’s Bradlow’s law for a common fibre-to-the-home infrastructure for fixed networks. Bradlow’s law for wireless networks is very, very different: “Network sharing is a race to the bottom,” he says.
Network quality is a clear differentiator for mobile, though. “The mobile network is less of a utility than a fixed network”, he says — simply because “a radio network is less easy to design”. Telstra's 3G network, branded Next G, runs at 42 megabits a second.
For FTTH networks, you put the fibre in “and that’s 80% of the cost”, but it costs less to maintain. “It’s less prone to noise and corrosion. It’s in-place inventory. It will last 100 years in its ability to deliver capacity. We’re just starting a journey here.”
Not with wireless networks, though. There are serious issues with spectrum capacity, he warns — even though Australia and other countries are looking forward to the so-called “digital divided” as analogue broadcasting services are booted off the airwaves.
Bradlow and his colleagues have been looking at what spectrum may be available over the next few years and what the demand will be. “We will need more spectrum than we can foresee being available to meet the expected demand,” he says.
And that’s even if the operators can use all the white space between the TV channels and take away as much of the military’s spectrum as they can. “Of course, we don’t know what they use it for so we can’t tell them it’s not important. They have the benefits of secrecy.”
LTE spectrum
Won’t LTE help? Broadcasters have a chunk of LTE spectrum for their outside broadcasts, he notes. “We are trying to make the government recognise it’s not a good use of spectrum.” And LTE Advanced will need 100 megahertz of coordinated spectrum.
With other users cluttering up the spectrum and preventing mobile operators using sensible sections, “it’s a bit like having a broken-down shed in an acre of prime real estate,” he grumbles. “It’s just not economically sensible.”
Yes, there is a raft of new radio technologies, including directional antennas and mimo — or “multiple-input multiple-output” — which will squeeze a bit more out of the spectrum. “But you can only go so far. It’s not an order-of-magnitude improvement.”
The technological target for the carrying capacity of LTE Advanced is five bits per hertz of bandwidth, compared with two bits per hertz for today’s HSPA. So, an improvement of two and a half times — a few months’ growth in data at current rates of increase.
Compare that with what technologists such as Bradlow can confidently expect out of FTTH. He can identify techniques that will boost the capacity of each fibre by 40 or 100 times. “Things are going up.”
And the capacity will be needed, as high-definition screens become common. “The cost per pixel on a screen is starting to fall dramatically. And the number of pixels is going up.” Bradlow can already see the uses now for the bandwidth that FTTH will promise over the many decades that the fibre will last: many TV sets around the home, but other high-definition displays too, all using up bandwidth.
“But the radio network is totally different,” he adds. With wireless, you have to keep investing — and “shared networking is a disincentive to investment”, he notes. Bradlow is proud of the 3G network that Telstra runs in Australia on the 850 megahertz band: “Our indoor coverage is so good that I can keep my laptop logged onto our VPN even when I’m in the lift.”
The network experience “is critical with customers”, he warns. “The network does have a major impact in the way customers feel about it.”
Network quality
At that point Bradlow, speaking during a trip to London, grumbles about the network quality he’s found while away from his home network. “I’ve had more dropouts in the last two days here than I’ve had in the last two years in Australia.” His network experience has not been good.
If there’s a good network in operation Bradlow is happy to talk about the “post-broadband world”, when services and not network quality will be the issue. “The key to running services is being seamless,” he says. And services will work across both fixed and mobile networks. “We’re still slowly working our way there. With time it will happen.”
When? Bradlow offers “2011 or later — I’m not sure which year”, but soon, as he declares that this will allow “telepresence, office communications services, rich presence” and so on: “A rich media connection with presence to give you details of people’s connectivity and their willingness to take calls.”
And the services will have to operate across fixed and mobile networks “if they’re to be of any value”, he notes: “If you’re a consumer you will want to connect with the most convenient system.”
Telstra is starting to deploy cloud services for small and medium-sized businesses, including workforce management systems and compliance software to help business owners follow employment law.
Apart from telepresence and other business services, what opportunities will there be for Telstra in the bandwidth-rich world? “There will be mainstream service opportunities for us, but we are subject to much wider competition,” he says. One of the big opportunities is Microsoft Office.
But one of the biggest is security. “It’s not a technology problem. It’s a people and process problem. Everyone faces the same threats.” But if a small business can’t afford the services that a large company such as Telstra has. “We’ve got to do it and it’s a marginal cost” to offer it to others.
Bradlow is clearly excited at the list of services that Telstra can offer via its wireless broadband network and NBN’s future fixed broadband network. “The core value for Telstra is the service provision, not the network operation,” he says. And he wants to build up the services.
It’s a transformation of Telstra that he’s forecasting — even more significant than the transformation from legacy networks to all-IP networks that is still worrying most telcos. Telstra wants to transform beyond worrying about the network. GTB

Further reading
GTB interview with David Thodey, Telstra CEO, here
GTB interview with Sol Trujillo, former Telstra CEO, here
National broadband network news here and here