Mixed messages on Telstra copper complicate FttN NBN case

Mixed messages on Telstra copper complicate FttN NBN case

Summary: Is Telstra's copper network dead or isn't it? Experts just can't seem to agree – nor can those people that were supposed to be Malcolm Turnbull's allies, who are compromising his Telstra renegotiations and his case for rural FttN at the same time.


Despite strong declarations over the years that it is both an absolute disgrace and a shining example of legacy infrastructure, Telstra's copper remains the subject of great confusion – with mixed messages from Malcolm Turnbull, NBN Co, Telstra and others confirming that nobody really knows whether Telstra's pivotal copper access network (CAN) is up to the task of delivering next-generation broadband or not.

Detractors have certainly had their say: iiNet CTO John Lindsay recently said parts of the network were "dilapidated" and many others have followed suit – including recent NBN Co board appointee Simon Hackett, who has previously harangued Telstra's "blackmail" of the NBN; argued for government ownership of Telstra's CAN; said Telstra will ultimately control how the Coalition's policy is rolled out; turned to Clarke & Dawe to explain the Coalition's NBN policy; and in April this year delivered a speech entitled 'The Problem with FttN'.

"I'm getting better". "No, you're not; you'll be stone dead in a moment." "I feel happ-ee-ee-ee". Is Telstra's CAN about to kick the bucket or not? Screen shot: David Braue / ZDNet Australia.

Most famously, a decade ago – under Switkowski's watch, no less – Telstra group manager for regulatory strategy Tony Warren told a Senate committee that the company's network was "at five minutes to midnight" and that the company was already engaged in "the last sweating...of the old copper network assets."

At the time, co-witness Bill Scales tried to qualify Warren's statements by saying that they were looking ten years into the future. Well, that was ten years ago – and here we are, still being told that Telstra's network is struggling – and that it has actually never felt better.

Back in June, Stephen Conroy warned that Telstra CEO David Thodey was hardly going to give away its copper and expected money for any deal on the network, which Thodey was already arguing was fit to support 25Mbps service as per the Coalition's NBN design. However, at that point nobody was interested in such piddling details, least of all what would turn out to be the incoming government.

Over six months later, Telstra is still arguing that its network is fit for purpose, with wholesale division managing director Stuart Lee arguing that it's like a "grandfather's axe" that has been repaired as needed to continue serving its purpose.

If you have somehow lived this long without seeing Monty Python's 'bring out your dead' bit, take a moment to watch it now. This is exactly where we are with Telstra's copper network.

Still not convinced? Just ask Ziggy Switkowski, a onetime Telstra CEO who now heads NBN Co and last week told a Senate Estimates hearing that Telstra's copper network is "robust" and that critics suggesting it was not up to the rigours of an FttN NBN were "misinformed".

A decade ago, Telstra said its network was "at five minutes to midnight" and that the company was already engaged in "the last sweating...of the old copper network assets."....[Yet] here we are, still being told that Telstra's network is struggling – and that it has actually never felt better.

This would all seem to complicate Turnbull's proclamation – made before the election and repeated on many occasions – that Telstra knows its CAN is worthless and will gladly give the government the whole kit and caboodle for free. Turnbull wants this to be the case because his entire argument for a "fast, affordable, sooner" NBN is predicated on it.

Upon reading those words in the morning papers, one suspects the shrill screams emanating from Malcolm Turnbull's kitchen would have not been because he had burned his hand on the sandwich press.

Here, after all, is the new CEO of NBN Co, who was hand-picked by Turnbull to make his copper-based NBN vision a reality – something that will only become possible if NBN Co can get reasonably cost-effective access to Telstra's copper – talking up the value of that copper network.

Turnbull chose him for his alleged network-building expertise, but Switkowski has so far mainly succeeded in giving the upper hand to the very same organisation with whom he will soon be fiercely negotiating to get access to, or ownership of, its CAN. Getting a fair price for that access – or, in Turnbull's ever-dwindling dream world, getting it for free – becomes harder and harder every time Switkowski sings the network's praises.

The new CEO of NBN Co isn't the only one causing problems for Turnbull's copper worldview: also stuffing things up was Alcatel-Lucent's global president of fixed networks Federico Guillen, who recently went on record describing the results of NBN Co's VDSL trials over copper.

The trials were, we were told, a resounding success – and it may well be that VDSL vectoring works as beautifully, over short distances, as Turnbull has been banging on about for months.

Good for Alcatel-Lucent. But if you keep reading, you'll note a qualifier that would probably have seen Turnbull's toast sailing across the kitchen at full speed.

"The landscape is huge but in the end there is a significant portion of the population in the cities. So in that regard it is not that different," Guillen said. "Of course there are rural areas which are quite different, and in those areas I would go with wireless. In dense areas, it is the same. It all depends on the loop length."

I'm sorry, but did the global head of one of the world's largest manufacturers of the technology on which Turnbull's NBN vision relies, just say out loud that it's unsuitable in the rural and regional areas that still – after 16 years of deregulation – represent a big black hole when it comes to broadband connectivity?

Did he really just recommend that regional and rural areas be serviced using wireless instead?

Did the global head of one of the world's largest manufacturers of the technology on which Turnbull's NBN vision relies, [really] just say out loud that it's unsuitable in the rural and regional areas that still – after 16 years of deregulation – represent a big black hole when it comes to broadband connectivity?

That can hardly have been what Turnbull wanted to hear; it wasn't too long ago, remember, that Turnbull was talking up the possibilities of 1Gbps over copper and arguing that the Coalition's NBN model would be better for the bush than Labor's.

The basic conceit of that argument, if you recall, was that the Coalition's FttN model would use VDSL to squeeze the most bandwidth possible out of the existing copper in the ground, whereas Labor's fibre-based model would bypass many small regional areas and outlying suburbs of regional centres.

"If you've got a community with 500, 600, 700 premises within... a couple of kilometres from an exchange building which has got fibre in it, you could easily see that you could readily provide good VDSL either from that exchange, or from that exchange plus a couple of nodes," Turnbull said at the time. "It gives you another string to your bow in terms of providing broadband solutions to the bush."

Apparently, however, that's not actually the way it works: if even someone as technically knowledgeable as Guillen believes regional areas should be serviced by wireless, what hope is there for even FttN in regional areas? Guillen believes the copper network in those regional areas simply isn't up to the task – and that VDSL is not fit for purpose for delivering rural broadband.

Voters have already rejected a Coalition NBN policy based largely on wireless technologies, which residents in rural areas have only accepted grudgingly upon realising they were too remote to be included in the FttP footprint. If technical advice now forces Turnbull to once again spruik a wireless solution for the areas of Australia that need good broadband the most, well where does that put his position?

Just to recap: the technical advocate who was supposed to support the government's FttN claims is now saying they're only half correct, while the status quo is being talked up by the new head of NBN Co – who is preparing to enter into what were supposed to be civil but firm negotiations in which the government would politely offer to take Telstra's supposedly decrepit, underperforming and therefore valueless CAN off of its hands.

If this is progress, I for one can't wait to see where it leads us.

What do you think? Have Guillen and Switkowski compromised Turnbull's FttN hopes? Or, can he push on with his FttN vision despite the ongoing confusion over Telstra's copper? 

Topics: NBN, Broadband, Government, Government AU, Telcos, Telstra


Australia’s first-world economy relies on first-rate IT and telecommunications innovation. David Braue, an award-winning IT journalist and former Macworld editor, covers its challenges, successes and lessons learned as it uses ICT to assert its leadership in the developing Asia-Pacific region – and strengthen its reputation on the world stage.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Caught between a rock and a hard place.

    On one hand Turnbull needs to talk up the copper network to get the people's buy in. On the other hand he needs to talk down the quality of the network in negotiations with Telstra.

    I enjoy watching politicians squirm as they slowly get undone by their own comments.
  • Line Rental

    If Turnbull intends running fibre to nodes, relying on copper for just the last 4-500 metres, will we see a commensurate reduction in the monthly line rental charges.

    • Very good point!

      Been wondering this for a long time. To get ADSL1, I MUST have voice services at $40 per month. WHY???

      We ourselves would certainly not need fixed lines if our mobile reception was better, or were on a decent enough broadband connection to run VoIP.

      So why are we still paying line rental? Is this another reason Telstra so desperately wants the copper to continue? So they can continue ripping us off $25-$40 per month even though a vast number of people don't actually need voice services over their copper?
  • Lies, damned lies and copper

    "Just ask Ziggy Switkowski, a onetime Telstra CEO who now heads NBN Co and last week told a Senate Estimates hearing that Telstra's copper network is "robust" and that critics suggesting it was not up to the rigours of an FttN NBN were "misinformed"."

    "Most famously, a decade ago – under Switkowski's watch, no less – Telstra group manager for regulatory strategy Tony Warren told a Senate committee that the company's network was "at five minutes to midnight" and that the company was already engaged in "the last sweating...of the old copper network assets.""

    If, as Ziggy say, we are misinformed then it would appear to be because Ziggy misinformed us? I am sure that Tony Warren would NOT be making statements to a Senate Committee that were not endorsed by his CEO.
  • What utter rubbish!

    The whole premise that everyone needs 1 Gbps to their front door is utter rubbish.

    It is clear that everyone is moving to mobile and, with most mobile plans limited to 1 to 5 Gb of data per month, the average demand per user is therefore 8 to 40 gigabits per month, or about 3 to 15 kilobits/sec. Yes, kbps, not Mbps! Even allowing for the mobile data plans suppressing demand, the fact is that even if you gave everyone 100 times as much mobile data allowance, then they would still be using only 0.3 to 1.5 Mbps, which is within ADSL 1 speeds, let alone ADSL2+ or VDSL!

    My current ADSL2+ plan has 80 GB data per month, and our entire household (4 PCs/laptops, 2 tablets, 3 smartphones + TV, home theatre etc) never get close to that figure, nor do most people and, at a typical speeds of 8 Mbps, it's more than enough to do whatever most people need. So why do we need to pay $39 billion for 100 Mbps on FTTP? Just so a few online gamers can get what they want? How about making them pay for a high speed wireless network dedicated to those who really want it? Then the rest of us can get what we want: cheap ADSL/VDSL broadband at home and work + wireless for everywhere else.
    • 3-15kb/s

      KLugsden, I am sure you must be trolling us.

      That is an average. People don't use the internet like that. It is used in fits and starts. And needs to be fast enough for the purpose wished to be used for. 4k streaming is on our doorstep and (from memory) requires a 100Mb/s - if no one else in your household is using the internet.

      apart from enteratinment uses, telehealth/education all need not only a good speed connection but rock solid reliability. Imagine a surgery performed over the copper, where it started to rain half way through and the service just drops out.

      It is not clear everyone is going to mobile. It is clear mobile is a complementary service to fixed broadband and that in fact mobile can barely cope with existing demand and that fixed line is an essential plank in our telecommunications strategy.

      I can't be bothered to write a more well rounded response becuase you are being so ridiculous.
      John Hicks
      • 3-15kb/s

        You did not address the critical point of my argument, i.e. that most people do not require very high speed Internet access.
        To suggest that a large proportion of the population is going to demand 4K live streaming TV is patently ridiculous. If anyone wants to do that, then let them pay for it, and not let it be subsidised by the average taxpayer.
        Likewise, any method of Internet delivery can fail, especially fibre which requires very careful jointing to prevent reflections. There are even cases of hairy-nosed wombats chewing through optical fibre cables in South Australia!
        So no surgeon is going to rely on the Internet being delivered via a single medium if it would threaten the outcome of his surgery. Even stock brokers have redundant data streams delivered via DIFFERENT and INDEPENDENT technologies, so why couldn't surgeons do the same?
        • ha

          "You did not address the critical point of my argument, i.e. that most people do not require very high speed Internet access. "

          Because everyone is just like you, right?

          Talk about arrogance and ignorance rolled into one...
        • Because there was no critical point to address.

          Just an assertion backed by verbiage and no facts.
      • Yep

        Surprised you spent this many words on this clown John! We never needed more than 640Kb memory inside a computer either. Better not tell him his smartphone now has 1600-3200x more memory than that now. LOL
        • Clown trumps idiot

          So you are happy to get 100 Mbps per NBN user, whose target is 95% coverage of Oz or about 15,000,000 premises, at 100 Mbps, which equates to a peak of 1.5 million Gbps of traffic in Australia alone? And where is most of traffic sourced? Overseas of course, so where are going to get cables that carry that amount of data?

          Don't you see that 100 Mbps is only for the END USER link. The overall speed is never going to be any faster than the slowest link between the server and client. As we are each sharing a tiny percentage of the undersea cable capacity, no one is likely to get more than 100 kbps for any more than a few microseconds at any time of the day, so who cares if the final link is 100 Mbps or 5 Mbps?

          As for 640 KB of computer memory, that is easily expanded without having to rely on external bottlenecks like undersea cables. You are obviously unable to understand that there is more limiting Internet throughput than simply the speed of the link between your front door and the cabinet in the road.

          Computer networks have completely different constraints to computer memory. Your idiotic comparison is like comparing the power of a car's engine with the number of cars that can travel on a road; 1.6 litre turbo charged engines now produce more power than 1970s V8s, but the roads they run on have not increased much in volume (number of cars carried at any point in time) or speed (how fast the cars are allowed to travel safely). Try to read a book on network design before you again decide to compare computer memory to network throughput.

          Clown trumps idiot again!
          • Source of data

            Klugsden, much content is overseas because the market for locally hosting it was incapable of demanding high bandwidth content due to the last mile throttling caused by copper. I live in the fasting growing regional city in NSW and can get 2 Mbps on a good night, but sub 256 Kbps when it rains.

            The biggest bandwidth component in the USA is video streaming. Netflix is about to open in Australia. Foxtel broadband, ABC IView and other catch-up TV services. All of these and more constitute a large part of domestic-sourced data. Cloud hosting of SAP, Google, Amazon likewise, all of whom have local load balanced servers. And in the meantime there is a steady stream of announcements of new undersea cables for the international content.

            No, as Simon Hackett said a few months ago, if you provide massive bandwidth to the premises, almost every other problem goes away.
          • 1080p streaming requires...

            AT MOST 20Mb/s (greatly exaggerated that figure, Youtube takes 3.5-8Mb/s)... Are you suggesting current/future copper technologies can't handle that? Are you suggesting you require most people in Australia to have access to 100Mb/s or God forbid 1Gb/s just to stream HD content? C'mon guy!
            Andrew Hargrave
          • hd and 4k

            are completely different. Hd 1080p requires between 6-8 mbps depending on compression. So if I lived with 2 or 3 other people and we all wanted to stream different hd programs at the same time. 25mbps of the liberal plan gone.

            Now 4k tv requires 20-25mbps. if 3 or 4 people in a household wanted to stream that at once. even 100mbps would only just cope.
          • Maybe I'm being short sighted...

            But I, as a 28 y.o who spends a fair bit of time on the computer playing games and watching shows, can't see the need to stream multiple HD or 2+ 4k TV shows at a time as justification for the greater internet being supplied to *every household* in Aus. I guess that's a point of contention that no discussing could change either of our minds. I don't see that as a necessity or even a luxury I will need in my life time, where as you do. *shrug*
            Andrew Hargrave
          • Need for fast speeds in both directions

            Increasingly there will be more peer to peer sharing within Australia. I (for one) have a 62 gig iPhoto library which should be backed up off site. As time goes on this amount of personal data will not be unusual. With fast reliable two way speeds and devices like http://www.filetransporter.com this will be possible. It isn't now.
          • Backing up data...

            in no way requires the speeds that FTTP delivers. Once you've uploaded the majority of your library, how large will your uploads then be? Are you suggesting you'll be taking gigs of pictures daily? With these gigs of pictures needing to be uploaded all at once?

            No, 10Mb/s is heaps of bandwith for backup processes for the common end user.
            Andrew Hargrave
          • Shortsighted Views

            Not so long ago I thought I'd never fill the 40GB HD in my new PC. Now my current PC's 2TB drive is almost full despite constant offloads to external drives to regain adequate space.
            Yet my "supposedly adequate" ADSL2 copper line still rarely exceeds 0.12MB/s in good weather & I frequently go out for a cuppa while waiting for a web page to load.
            Upload any worthwhile data? You've gotta be joking!
          • your copper...

            Is not as good as some other people's copper. Good Adsl2+ which some of my friends and their families have been on for years, is far more bandwidth than I can see the vast majority of Australian homes needing for their daily needs for many many years to come. I am still yet to be convinced otherwise. Remember your personal speeds, Sultanabran-, are not classified as 'good'.
            Andrew Hargrave
          • Well

            Where to start. Just by reading your comments I can clearly see it is you who does not understand how the basics on how a network works. Lets start with some basic math. Your assertion that people only really use 3-15kb/s is ridiculous. At those speeds it would take more than 3 min just to load a single 3MB page. I am certain you would complain if this page took that long to load, so that is clearly an idiotic claim.

            Secondly, we don't pay for the speeds to download 100mbps 24/7. If we did that then even the largest data plan available to the domestic user would disappear in about 3 hours. I know, I am on fibre and we managed to go through 250Gb in 1 hour and even that isn't the top speed of our connection. Adding on to that is the fact that we do use 100mbps. When I am downloading new games that I have bought (yes i am a gamer, so are a lot of people) I often see 10mb/s. That is 80 mbps. This is while my room mate is streaming 1080p video and i am playing an online game.

            Thirdly all the experts will tell you that the links to overseas will never be the bottle neck to internet speeds. There are a few reasons for this. Firstly if everyone of the 93% of households connected were to get the 100 mbps plan (nbn co would be super rich if this happened) then not everyone would be a. on the network at once demanding 100 mbps and b. not all will be demanding access to overseas websites at the same time. It is for this very reason it is stupid to say that oh dear we shouldn't build a network that can allow this just in case everyone in Australia signs up to one of the highest speed plans and then decides to, at exactly the same time, demand 100mbps from overseas.

            This brings me to the fourth point (which is kind of 2 points in one). How much we actually use and how much spare capacity we have. Using your 15000000 households figure we would need 1.5 petabits per seconds or 1500terabits. We currently have about 5-6 terabits of capacity from about 8 different links. While this falls far short of a theoretical 1.5 petabits required, the actual internet usage for overseas traffic in 2009 (i.e all traffic not just to the US) was 300gbps. Far short of the available capacity. Putting that into perspective as of 2009 our internet usage could have doubled, doubled again, then doubled again, then doubled yet AGAIN before the CURRENT capacity would be exhausted. On top of all this the network capacity can be easily quadrupled or more just by cheaply upgrading end network points (i.e upgrading your router) on existing networks. Currently they are already building a cable that by 2015 will be able to handle the current capacity of every other cable by itself. Without upgrades.

            So yeah. Research trumps clown!