In defence of Turnbulls’ NBN speed claims

300

opinion Those blinded by Labor’s glitzy NBN vision need to rub their eyes for a second and realise that Malcolm Turnbull knows what he is talking about when he says there are few consumer applications which require the kinds of 100Mbps speeds which the fibre network will provide.

The Shadow Communications Minister has come in for an extraordinary amount of criticism this week from all sides, due to a series of seemingly innocuous comments he made on the Patch Monday podcast put together by journalist and commentator Stilgherrian. “It’s certainly very difficult to think of many applications that are of interest to residential users that would not be perfectly well serviced by the speeds I’ve described,” the former Opposition Leader said, referring to speeds between 12Mbps and 24Mbps.

In the ferocious debate underneath the article, as well as a derisive follow-up commentary by Stilgherrian published by The Drum, a number of commenters ripped the Liberal MP a new one. “Shame, shame, shame,” wrote one commenter, while others accused him of simply not understanding optic fibre technology or of having “limited vision”.

On Twitter, not only was the same rage repeated, but a sense of superior amusement had already set in, following the paraphrasing by futurist Mark Pesce of Gizmodo’s rebuttal of Turnbull’s comments into the slogan: “No computer could ever need more than 640K”. – Bill Gates, 1981. “No computer could ever need more than 12Mbps.” – Malcolm Turnbull, 2011.”

Despite the fact that Turnbull never actually said those words, Pesce’s influence ensured the quote was passed around between online commentators like a well-chewed morsel of humour, including between journalists like Risky Business publisher Patrick Gray and iTNews.com.au editor Brett Winterford.

The general attitude was one of amused condescension.

Turnbull’s comments, so the sentiment went, illustrated that, like other politicians before him, the Member for Wentworth fundamentally just didn’t understand technology. While not truly a luddite, many of those who commented on the matter seemed to feel Turnbull had ultimately betrayed his true colours; that he had shown once and for all that he wasn’t qualified to set telecommunications policy and he should stick to matters he understood better … his background in law and business, perhaps, or even his popular stance on climate change.

Thus, it might surprise you to learn that I agree with Turnbull’s statements completely and believe that many of those slamming the politician have missed the fine-grained nature of what he truly told Stilgherrian.

In his commentary on The Drum this week, and previously, Stilgherrian has argued that the application which will soak up the 100Mbps bandwidth promised by the NBN is clear: Multiple streams of high-definition video to people’s residences. He has consistently given several examples to back this argument, ranging from a geologist remotely controlling a mine site or a HR manager teleconferencing with colleagues (both from home), to virtual doctor’s appointments and collaborations over school projects.

In a previous article on Technology Spectator, Stilgherrian has listed the bandwidth requirements which would allow the tasks to take place. And, courtesy of a hyped marketing video produced by NBN Co and promoted by Stilgherrian as a realistic look into the future, we can even have a look at what all of this frenetic video conferencing activity might look like in days to come.

Now, I’m not going to argue with the fact that all of the technology required for all of this videoconferencing — as much as ten HD streams from one house! – is available today. After all, vendors like Cisco and Polycom already offer high-quality videoconferencing gear for business, and in the consumer sphere there are a range of suppliers selling decent kit as well.

But what I do want to raise is the fact that it might not require quite the level of bandwidth which Stilgherrian claims it does – and also the idea that despite it being available, so far most Australians have shown absolutely no interest in actually using it.

In his article on The Drum, Stilgherrian never actually goes into what bandwidth level each of the HD videoconferencing applications he describes requires. However, in a previous article published by Technology Spectator, he does. And it quickly racks up. Let’s go through his list of what could be a typical household:

  • The geologist mother: Remote mine viewing (2 X 5Mbps), plus 3D manipulation of rock strata (10Mbps), plus a videoconferencing session with her boss (5Mbps)
  • The HR manager father: Videoconferencing with his PA (10Mbps, plus a possible further 10Mbps for passing handwritten notes)
  • The student son working on homework with friends (10Mbps)
  • Daughter streaming video online (5Mbps)
  • Another family member’s doctor’s appointment (10Mbps)
  • Security camera monitoring (10Mbps)

Hmm. With a sum total of 80Mbps of capacity being used in the vision outlined here, it’s not hard to see why some people believe that the sooner we get the NBN, the better. The only problem, of course, is that all of this is complete horseshit of the highest magnitude.

Let’s take the geologist mother. Does anyone seriously believe that it’s possible to remotely manage a complex mine site from their home office thousands of kilometers away? Doing so raises so many issues around safety, duty of care and even simple management philosophy that it’s not funny. In the event that there was an accident on site, how happy do you think the authorities would be that the facility was being overseen from Ballarat?

And even if you assume it is possible, why would someone doing so maintain two high-definition video cameras trained on the site at all times? What could they possibly be watching that wouldn’t drive them nuts with security guard-esque boredom? And how could they keep an eye on those two cameras, while also carrying out a videoconference with their boss? This use case makes absolutely no sense.

Lastly, why would anyone carry out all of these tasks simultaneously while manipulating a complex three dimensional geological model? And even if they were doing so, why would displaying that 3D image remotely soak up 10Mbps of data, given the easy availability of remote terminal tools like Citrix, which require quite a bit less than that, and are used every day for access to remote sites?

What Stilgherrian has described in the case of the geologist is the extreme example of a professional working from home but using none of the tools commonly available today to get around issues of limited bandwidth. Tools like … the telephone, for calling your boss. Virtualisation software from companies like Citrix for accessing remote computing resources. HD webcams which squirt an image down the pipe to you once every few minutes … so that you don’t have to waste bandwidth with a constant, needless HD video stream.

The same is true for the other examples listed. There are plenty of HR managers out there who already work from home. But they don’t need HD video to do so – what they normally need is a software as a service application like NetSuite which enables them to work through a browser, as well as a basic Unified Communications suite. Showing their personal assistant a picture of their kids’ sporting trophy? That’s what Facebook is for.

As for the other examples … have you ever tried multi-way videoconferencing? Between more than three people? I can assure you it’s normally a glitchy nightmare, unless you have a whole room set up for the purpose, such as with Cisco’s pricey Telepresence suite – and even then it gets a bit complex. The days when kids will be able to use HD videoconferencing to effectively work on a school project with “three classmates across town”, as Stilgherrian describes, are very far away indeed. Right now, it’s just more efficient to work together in person.

And the numbers of doctors who would consent to consult remotely via videoconferencing – a format which does not let them touch a patient or examine their symptoms as easily as they could in person – are very slim indeed, right now, and likely limited to those servicing patients in very remote areas. And I should know … I grew up in one of the most isolated ‘cities’ in Australia – Broken Hill, in Far Western NSW.

Perhaps the most realistic example Stilgherrian gives is that of a daughter watching a streaming film online (5Mbps). However, again, this does not represent a completely accurate picture of the situation.

I regularly watch 1080p videos streamed online from sources like YouTube. This morning I tested what impact this had on my home ADSL2+ connection. What I saw was that the application grabbed all the bandwidth it could (16Mbps) to download the entire video file and cache it on my PC. That took only a few minutes for the video I was watching. Thereafter – and for the vast majority of the time I was watching the video (a StarCraft II game, if you must ask) – YouTube was using no bandwidth at all from our broadband connection.

Even if you assume constant streaming video as delivered through an ISP’s IPTV video service, such as the FetchTV service being offered by ISPs like iiNet, such ISPs only recommend you have a minimum total connection of something like 4.6Mbps to access such services.

Turnbull, I can assure you, is painfully aware of all of these facts. During a recent visit to South Korea, the world’s global fibre broadband mecca, the Shadow Communications Minister received a demonstration of a high-end Cisco videoconferencing solution which required just 1.5Mbps (symmetric). This is a far cry from the 10Mbps which Stilgherrian has listed – and if you have the Annex M standard set up on your home ADSL2+ connection, with the higher uplink speed it allows, you could probably get two going at the same time.

In short, much of the vision which NBN Co and Stilgherrian are painting also doesn’t gel with current usage of videoconferencing technologies which have in some cases been around for decades.

Vendors like Polycom have long been spruiking desktop videoconferencing as the solution to quickly boosting business productivity. However, in practice, what most employees have found is that technologies such as screen-sharing, integrated unified communications (employee directories linked to VoIP with click to call and instant messaging) have found far higher adoption rates in the workplace.

More often than not, Australian workers simply do not want or need to speak to each other face to face via videoconferencing to get their job done. It just takes far too much attention, and they would rather be showing each other a document on a screen, instant messaging for a quick answer, or having a phone chat while looking at something else on their desktop PC.

Even in the small business sector (which I personally work in), videoconferencing is far from popular. Many small Australian businesses are closely linked via consumer-grade unified communications platforms like Skype; but again the videoconferencing feature is rarely used; it just takes too much effort and attention.

Where videoconferencing is being used, so far, is by large corporations devoting whole rooms to videoconferencing ‘suites’, where a group of executives can have a coffee meeting without flying to the next city. These suites are popular inside national organisations like ANZ Bank, which has a great deal of interstate travel, but haven’t taken off in smaller companies.

Perhaps the only example I’ve seen of a major Australian corporation deploying desktop videoconferencing widely is the Commonwealth Bank, and in a recent case study published by Microsoft about the effort, it was the instant messaging and directories component of the wider unified communications rollout which got more attention – not the video aspect.

Now, the obvious response to much of my argument here is that videoconferencing hasn’t taken off yet in Australia because our broadband hasn’t been fast enough to support it. But I would point out that inside major corporations, this simply isn’t true. Fast broadband to the desktop via internal Ethernet networks, with Internet access and inter-office links often provided by high-capacity links, has long been a reality in Australian business, as well as the education and health sectors, as Turnbull has consistently pointed out.

However, videoconferencing has just not taken off.

With the videoconferencing argument largely out of the way, few residential applications are left which could possibly soak up more than the bandwidth we enjoy today. Video gaming? It’s a very popular pasttime right now. E-health? Online education? Working from home? All of that is already being done on today’s networks. There is no doubt that all of these networks will gradually require more bandwidth, but we can tackle that as it comes … doing so right now with a national fibre network constitutes massive overkill … and represent a solution to a problem which doesn’t largely exist.

In addition, underlying all of these points is the question of who should pay for the next generation of broadband. With most applications being able to be run on today’s networks, Turnbull’s argument, and it’s a good one, is that Government telecommunications policy should set a framework so that the private sector can invest in better services where it makes financial sense to do so — that is, when and where there is demand.

A perfect example of where this has already occurred is in the CBDs of our cities, which have long been wired for fibre, because businesses there need the faster speeds and are willing to pay for them. Where there is demand but not enough demand to make deploying infrastructure in remote areas financially viable, Turnbull has proposed a policy which would see the Government step in. It’s a classic liberal-style minimalist policy — and a good one to compare with Labor’s big-spending NBN alternative.

In conclusion, before you go slamming Malcolm Turnbull for his views on the NBN, ask yourself this: What bandwidth do you currently have available in your home and workplace … and have you ever used it for videoconferencing? Why not?

As an absolute technology nut, early adopter, online gamer, lover of HD video and small business owner who works from home, I would personally love to have 100Mbps broadband connected to my home. But then again … I’m really not sure what I would use it for that I can’t do already.

Image credit: Office of Malcolm Turnbull

300 COMMENTS

  1. Sorry, but I don’t see you writing off e-Health that easily. Are you saying anyone can get an inspection at home from anywhere in Australia with stable HD video for the doctor?

    I see that one of the cases for NBN. More bandwidth is necessary, and it’s never bad.

    • When you say e-health, do you mean videoconferencing with doctors or specialists? Because I covered that quite a bit above. When was the last time you saw videoconferencing being used in e-health?

      In addition, I’d like to see you clarify this statement: “More bandwidth is necessary”.

      Necessary for what, precisely?

      • Heh well, what you covered is a failure of the Australian telemedicine network, not it’s uselessness. I came to Australia from Ontario, Canada – which has the biggest telemedicine network in the world that services many Northern (and unrecheable or with very very long trip times) – OTN, http://www.otn.ca.

        You can’t just say “I haven’t seen a doctor who’d like that”… that’s not how it works. Doctors are specially trained to do those kinds of consultations and they’d be working with the proper agencies for that. What you should think is “Have I seen a patient who’d like that?” and maybe you’re remember some cases from your isolated community when someone would’ve liked not having to head to a major hospital that was very far away?

        There are a lot of cases for more bandwidth in eHealth, and OTN is going out of their way to even to pay for high-bandwidth top-notch business lines at doctors own homes so they can work, perform remote surgery consultations or do their work while they’re on maternal leave.

        That’s just a few random examples just to start. It’s a bit silly to dispute a need for improvement.

        • “It’s a bit silly to dispute a need for improvement.”

          That’s a bit of a motherhood statement ;)

          It seems like most of what you’re talking about can be achieved through soft outcomes such as doctors working with government agencies to be trained in this area. And even much of that can be done over existing infrastructure. Most of what I am talking about is whether the need for billions of dollars of infrastructure investment can be justified by discussing applications which may not need it.

          While I appreciate your comment, I think you need to get into specifics of the argument more ;)

          • No, I’m not talking about what should doctors do if they want to do telemedicine. You’ve said that you’re unsure if they want to do that, I’m saying that they’re already doing it, and that requires bandwidth.

            I also listed other high-bandwidth applications, such as remote surgeries, virtual appointments, senior always-available health support and other cases…

          • No worries Vadi. Thanks for your comment, I appreciate it.

            With this in mind, would you be able to detail what bandwidth requirements all of those individual examples have, so we can work out where they sit in the debate?

            Much appreciated!

            Renai

          • I’m afraid I won’t go to such far lengths for a debate on the Internet, sorry :) I am very in support of NBN though and if there’ll be a national referendum, I know where I’ll be voting.

          • Fair enough mate — but you’ll note that I did go into such specific examples in my article. If you are going to respond, it seems logical you do the same ;)

          • Hi Renai, Video conferencing in Telehealth is already happening extensively in parts of rural Australia. So much so that hospitals are spending money building their own networks as they can’t wait for the NBN. If everyone had NBN then the number of people needing to take time off work and visit the hospital would be decimated – an ambulance costs several thousand dollars to send out. Video consult costs $200. But you need everyone to have fibre for that. Ignore present day applications, think of the future. It might be acceptable for you to wait a few minutes while HD youtube buffers, but it’s wholly unacceptable for the general public using it as their main source of television viewing. There are plenty of households out there where the kids all watch their own TV in their own rooms. You need a lot of bandwidth for that to happen (in addition to other activities). Above all, the current copper network needs replacing – why on earth bother to replace it with copper when we can replace it with fibre? As for Turnbull knowing about this stuff, I’m afraid he’s proved himself fallible after telling everyone that wireless improvements undermined the requirement for a fibre-based NBN. That was nonsense and he got called on it and now he doesn’t say it any more.

        • i think the telemedecine use case is a massive white rabbit.
          1. if i’m in the city and am sick enough, i’d 100% prefer to go see someone in person or call a locum
          2. if i was in the country, i’d much prefer to get to nearest town and see a GP.. if the GP is not qualified to treat my illness, i CAN see the need for a HD video conference to a city specialist.. But i know i’d prefer to do this from a hospital where another GP or nurse could be present… in this situation the country hospital needs fibre, not my farm. (and note, if i’m really remote i’m not getting high speed anyway)

          • People are confusing the primary doctor’s visit with followup visits. Of course patients should see a real doctor when being diagnosed and initially treated, but patients with serious and/or chronic conditions are often seen by specialists every few weeks for what amount to brief “how are you doing?” visits.

            For millions of country residents these visits still require days off work, alternative childcare arrangements, a huge expense in petrol and car wear-and-tear, fatigue from hours of driving, hotel costs, etc etc. and could in many cases be very easily handled via a 15-minute VC from their homes. Telehealth is of course not a complete substitute for hands-on care but it also doesn’t follow that it cannot play an important role.

          • Exactly. I have to travel interstate over recent years at great cost for followup appointments, where the surgeon often only needs a visual and no touch. This sort of technology would save days of travel, lost wages, flight/accomodation costs and convert it to a 15 minute conference.

      • “When was the last time you saw videoconferencing being used in e-health?”

        Exactly. We don’t have the internet speeds to do it. Sure, you can give high-speed internet to hospitals and maybe to doctors’ practices… but without the same speeds in the patient’s remote home, it would be useless.

        • “We don’t have the internet speeds to do it.”

          Hmm perhaps … but I feel like if there was a significant demand for this, we would have seen it in patches already — but we really aren’t.

          • There are plenty of countries and areas where there is fast broadband speed (Korea and Japan)

            I have yet to see any somewhat universal home end application actually needing that bandwidth which doesn’t fall into the area of entertainment

            Hell in Europe the main impetus for such speeds (i.e. the way it is being marketed) is for 3D porn…we already had someone from that industry commentating on this ;)

          • Bit like saying there was no significant demand for Youtube style applications during our dialup days because it wasn’t invented yet. The applications came when the infrastructure could support it. ie ADSL and cable

          • Exactly. Who’s going do translate ideas into applications that can’t run because the infrastructure is not there?

      • Hi Renai, Video conferencing in Telehealth is already happening extensively in parts of rural Australia. So much so that hospitals are spending money building their own networks as they can’t wait for the NBN. If everyone had NBN then the number of people needing to take time off work and visit the hospital would be decimated. But you need everyone to have fibre for that. Ignore present day applications, think of the future. It might be acceptable for you to wait a few minutes while HD youtube buffers, but it’s wholly unacceptable for the general public using it as their main source of television programming. There are plenty of households out there where the kids all watch their own TV in their won rooms. You need a lot of bandwidth for that to happen (in addition to other activities). Above all, the current copper network needs replacing – why on earth bother to replace it with copper when we can replace it with fibre? As for Turnbull knowing about this stuff, I’m afraid he’s proved himself fallible after telling everyone that wireless improvements undermine the requirement for a fibre-based NBN. That was nonsense and he got called on it and now he doesn’t say it any more.

    • Its not about any specific application, people are missing the big picture here – we have reached the limits of the copper network, and we are talking about building a network for the future. Fibre will always be upgradable to faster speeds, copper has reached its limits. We may not see the need for 100mbit now, but in the future we will. Further, there are many people who cannot even get ADSL/Cable, let alone multi megabit speeds, and are stuck paying Telstra’s exorbitant next-g data prices. You dont have to get a 100mbit plan now, but for many people even 5mbit on the NBN will be a major advance from what they can get now where they are. When you look at how important the internet has been to our lives, our education and our work, it is imperative that we look forward to the future so that we are not disadvantaging ourselves as a nation in the future.

      • we have no way in hell “reached the limits” of our copper network. copper technologies are improving all the time and there are plenty of vendors selling tech upgrades that can push the envelope further and further.

        • We are quite clearly approaching the limits of what copper can do. To improve speeds by any sizable amount for 90+% of subscribers we need to dramaticly shorten cable runs (and preferably change to VDSL2) or add pairs (and go with more esoteric technologies.)

          Spend billions to set up nodes within 300m of everyones’ premises and terminate the copper at them? Been there, rejected that.

          Extra pairs? Totally unworkable.

  2. Sigh.

    You know what the biggest problem with the above is?

    It assumes that future needs will be exactly the same as todays.

    Given the radical changes in data usage over the past 10 years can you honestly say that 100Mb/s won’t become the norm in the next 10?

    • Actually I think 100Mbps+ will definitely become the norm, and I’m looking forward to it. What I said, and what Turnbull said, was that there aren’t any examples of residential applications which require those speeds right now.

      In many ways I am naturally conservative about technology. I’ve been through dozens of ‘hype cycles’ over the past decade. Some examples of the ones which have failed … netbooks, “grid computing”, WiMAX, desktop videoconferencing, 3D televisions and so on and so on. Picking a winner here isn’t easy, and while fibre is likely that winner, right now there isn’t much justification for it on an application basis.

      I’ve always said my preferred telecommunications policy would be to split up Telstra, build competitive rural backhaul links, legislate for MDU access to HFC cable and then see what happens. Sure, I’d love fibre, but I don’t think there is a complete justification for it yet.

      • There’s rarely ever a complete justification for anything. By the time you have one it’s too late. Also, I don’t really think fibre is in the same category as some of the “hype cycle” examples you’ve given.

      • There’s an old telecoms industry saying: “What’s the point in having the first telephone?” If you went back in time, do you think you’d be supporting the chorus of voices slamming the expensive rollout of a telephone network when hardly anyone at the time wanted or ‘needed’ it?

      • “was that there aren’t any examples of residential applications which require those speeds right now.”
        So you want us to wait until the apps are released and then constantly play catch up, how about we ready ourselves so we can use whatever becomes available

      • So Renai,

        You state “Actually I think 100Mbps+ will definitely become the norm..” How do you see that happening if a fibre infrastructure ISN’T deployed at some point??? Infrastructure doesn’t appear overnight!!!!! If the NBN takes 10 years to be completed, do you HONESTLY think that based on clear current projections of exponential data growth we have been experiencing for the last decade that 100Mbps will not be necessary in 5-10 years?????

        Turnbull is using the excuse that most people don’t need 100Mbps now therefore there is no point building FTTH. Heck he was advocating nationwide wireless until the universal panning he received from experts now made him turn to FTTN.

        The fact is everyone agrees data use will grow….A LOT… What you and Turnbull fail to understand is that it is not a matter of flicking a switch and woila 100Mbps exists around the country…..It takes an ENORMOUS AMOUNT OF TIME for that infrastructure to get designed and built and what you consider superfluous speed in 2011 will be barely meeting demand in 2021!!!!

        Why is that such a hard concept to grasp? We are FUTURE proofing…if you build something based on your CURRENT needs instead of FUTURE, well then you might as well write off the money and flush it down the toilet!

      • “Given the radical changes in data usage over the past 10 years can you honestly say that 100Mb/s won’t become the norm in the next 10?

        ‘What I said, and what Turnbull said, was that there aren’t any examples of residential applications which require those speeds right now.’

        The answer is in these two points. NBN won’t be all complete for another 7-8 years, whether built in its current form or Malcomm’s patchwork form. So it would have been good to analyse what programs could be available in the future or what other needs might arise. If we built the network based on what’s available now, even dial up can do the job, albiet over a few hours.

        only thing that stands between my vote and coalition is NBN. once my house is wired up, i will listen to everything else Abbott/Turnbull are saying.

      • “Some examples of the ones which have failed … netbooks, “grid computing”, WiMAX, desktop videoconferencing, 3D televisions and so on and so on.”
        Netbooks led to other cut down devices like tablets, video conferencing is alive and well with products like skype. As for grid computing, you’ve clearly never set foot in a data center or heard of any of Amazon’s web services which make extensive use of grid computing to deliver scalable applications.
        Looks like the NBN has a fairly good chance to me.

  3. Journalists and politicians aren’t exactly the most qualified people to comment on technology needs, reading emails, websites, twitter, powerpoint presentation and documents on iPads isn’t exactly demanding IT stuff.

    Anyone who works in a corporate environment knows how poor the experience working remotely via a VPN is in remote offices or at home, let alone small business trying to make do with ADSL which doesn’t even reach anything near 12mps downstream

  4. Is uploading HD video and watching streaming video of events every few days demanding enough, Zac? Because I do that all the time. I’m pretty much the most demanding user of the Internet you could imagine — and I use it about 18 hours a day.

    As for VPNs … sure, they are still important, but I’ve never had a problem using them given my connection, and increasingly they are becoming dated technology, due to the growing usage of SaaS apps which don’t require VPNs.

    • Not sure what line are you using, but whenever I’m uploading anything, downstream speeds get severely affected and everything is a crawl while uploading.

      • Vadi, that’s because your uploading is slowing the return of ack packets for any downloads. It’s a problem with all connections, where if one channel maxes out, the other suffers just as much. It’s very easily solved by throttling any uploads to 90% of the connection limit, leaving that other 10% for ack’s to be transmitted. This is not a limitation of ADSL or cable or any other hardware layer, but of the TCP protocol when saturation is reached – if you had a 100Mbps connection upstream and were uploading at capacity, the same problem would exist.

  5. Renai, sorry – but you’ve missed the point. I think Stilgherrian has, largely, too.
    Constant use of 100Mbit in a household is, today, likely to be very much an edge-case.

    However, it is ‘bursty’ use of 100Mbit that will likely be of the most benefit in the short term.
    In many homes people are using digital cameras to take still images and hd video. They’re often then editing and putting that information online.
    But uploading a dozen 5MB images, or a 10 minute 720/1080P video clip (easily 800MB for most people) takes a long time, as I’m sure you’re aware, on an ADSL link – topping out on average at maybe 1Mbit.

    With a 40Mbit uplink (as you’d get on the current highest level of NBN Connection), the time taken to upload will be massively reduced.

    Cloud file services such as Dropbox or Mozy are becoming much more common too – and while they’re pretty snappy for updating a few hundred KB, they quickly bog down if you need to sync larger files.

    It’s not hard to see services such as Dropbox expanding to providing file sharing for an entire family rather than a single person. In that case, Dropbox’s current approach of being able to download all the files locally won’t work so well, and so you’re back to on-demand needing to download files. This comes back to needing fast network connections so you don’t spend 10 minutes waiting for each file to come down.

    In terms of ongoing high bandwidth streams – well, you point out the recommendation of FetchTV only needs 4.6Mbit. That may be true today, but I’d guess that this is because it’s targeted at the lowest common denominator, rather than delivering the best picture quality.

    If FetchTV wanted to provide a higher quality stream it’s not hard to see that quickly expanding to 30-40Mbit for a 1080p channel. That’s a fairly realistic bandwidth requirement for HD channels based on existing broadcast and cable systems.

    Having someone shout out “Stop using the Internet so I can watch the TV” will quickly become a common scenario in homes with an ADSL link and an IPTV service.

    Ubiquitous high speed bandwidth isn’t all about the ability to use all the bandwidth, all the time. It’s about having that capacity there when you’re actually wanting to do something.

    • “But uploading a dozen 5MB images, or a 10 minute 720/1080P video clip (easily 800MB for most people) takes a long time, as I’m sure you’re aware, on an ADSL link – topping out on average at maybe 1Mbit.”

      I’d actually disagree. A 10min 720p/1080p clip on my PC is nowhere near 800MB — you’d be lucky if it hit 200MB.

      As for Dropbox …. I take your point, but are you really going to argue that we need to build a nationwide broadband network, spending tens of billions of dollars, so that people can back up their data? Why can’t they just (as most people are) use an Apple Time Capsule or other networked hard drive? Or a hard drive they periodically back up to and store off-site?

      “If FetchTV wanted to provide a higher quality stream it’s not hard to see that quickly expanding to 30-40Mbit for a 1080p channel. That’s a fairly realistic bandwidth requirement for HD channels based on existing broadcast and cable systems.”

      I’m sorry, but I watch 1080p TV all the time streamed from YouTube and other sources, and it doesn’t require anything like 30-40Mbps. I have no idea where you’re getting that figure, but current evidence doesn’t support it.

      “Having someone shout out “Stop using the Internet so I can watch the TV” will quickly become a common scenario in homes with an ADSL link and an IPTV service.”

      iiNet, Internode and others have solved this on FetchTV with Quality of Service. And I should also point out that plenty of people have been watching TV through HFC cable — while still accessing the Internet — for more than a decade now.

      To be honest, I can see where you’re going with this argument, and I don’t disagree with your overall philosophy. But this debate is *all* about concrete examples, and the examples you have delivered don’t mesh with my personal experience as a big consumer of the products you’ve described.

      • Re filesize of a 720p video:
        I think you’re looking at the files after compression. Most folks are going to take the files as the camera gives it to them. Meaning much larger file sizes.

        It’s the same as people emailing the giant version 12megapixel version of some happy snaps – they don’t know any better.

        Re Dropbox/Apple Time Capsule:
        Yes, I could spend a few hundred dollars buying and using local NAS (I do). Or I can spend $5/month with, say, Backblaze or Dropbox and get automatic networked backups. The kind that ‘just works’, even if my house gets robbed, and has no storage limits. (Ok, so Dropbox does, but Backblaze doesn’t).

        Local in-house storage of files will quickly become a thing of the past for everyone except uber-geeks like me. Even then, if I had reliable 100Mbit and no need to worry about bandwidth caps, well… I’d seriously consider moving more of my content offsite.

        Re FetchTV/TV Channels:
        I’m not able to dig up hard references at the moment on actual HD Channel bandwidths, but what you’re looking at for Flash video isn’t the same as for MPEG-TS that’s done for most Digital TV systems.

        If I can use my own HD TV recordings though for rough guestimates: a OneHD recording is 22GB for 3hrs 18mins. This gives about 15Mbit, for a 1080i@25hz recording (afaik).
        If you wanted a 1080p@50hz stream – something much better for sports. This would roughly quadruple that to around 60Mbit.

        I’ve been told by a broadcast engineer that HD 1080i@24hz channels are somewhere around 6-8Mbit depending on the broadcaster and their total channel capacity. So, if I can abuse that and also scale 4x, we’re still around the 32Mbit mark. Maybe some smart compression would help here to let it squeeze onto an ideal ADSL2 connection.

        And on the QoS side:
        QoS only helps if you’re actually able to deliver other traffic in a reasonable period of time. Having the internet die or come to a crawl because someone switched on the TV isn’t an ideal situation.

        • “Most folks are going to take the files as the camera gives it to them. Meaning much larger file sizes.”

          Will, are you suggesting that people upload uncompressed video to the web? Because that just doesn’t appear to make much sense. Surely all uploaded video should be compressed in some form.

          “Local in-house storage of files will quickly become a thing of the past for everyone except uber-geeks like me.”

          I’m sorry, I don’t agree — and again, I don’t think it is a persuasive argument that the Govt should build a multi-billion dollar fibre network so that you can spend $5 a month sharing files over Dropbox.

          As for the TV argument … mate I suggest you have a look at what’s being broadcast through FetchTV and YouTube right now, on existing ADSL networks, without disrupting the rest of the house. Arguing that IPTV is going to require 30-40, or even 60Mbps as you just wrote, isn’t going to get you anywhere with me — I already use IPTV and I’ve only got 16Mbps.

          We need to look at current commercial models here and extrapolate … talking about uncompressed video online is not going to get us anywhere at all ;)

          • Fetch TV is relatively poor quality streaming, and YouTube uses highly compressed Flash video, their ‘1080p’ streams are not really 1080p at all.

            ‘real’ HD, or BluRay, can use 40Mbit/s on scenes where alot is happening, and usually needs 20Mbit/s minimum.

            Also whilst Fetch TV itself cannot saturate a connection, see what happens if you try and watch it while some other PC is downloading a windows update, or stuff from iTunes, watching HD videos online – I have a 15Mbps connection, and ‘can you stop downloading that, I’m trying to XX?’ is a very regular conversation.

          • Relatively poor to what, its the same quality (it uses lossless compression with the same quality standard as what is released on physical media)

          • “Will, are you suggesting that people upload uncompressed video to the web? Because that just doesn’t appear to make much sense. Surely all uploaded video should be compressed in some form.”

            it’s not uncompressed. Most consumer camcorders encode in AVC, but the bitrates are high so as to preserve quality so the file sizes are much larger than say a 1080P scene-standard .mkv.

            With Canon consumer cameras for example, 10 minutes of 1080p footage at 25FPS will be around 1GB.

            You don’t resize all your SLR photos when you upload them to Flickr and Picasa, the backend of the system you’re uploading to does that for you and keeps the source files intact, I don’t understand why you’re suggesting that people would pre-transcode all their home recorded video when uploading video in similar ways.

          • A thing many many people don’t realise about video encoding. It takes a lot of bandwidth. And like all things there are many choices and tradeoffs.

            The main ones:
            Timeliness (with 2 sides, Encode and Decode)
            Quality
            Bandwidth.

            Youtube excels at none of these facets, but is acceptable in each. (except encode). Though youtube as a company have grown to the point where the encode delay is hidden from view, or money has been thrown at to compensate for.

            Re deteego below, do you even know what lossless means?
            Nothing uses lossless compression. Not even bluray.
            Additionally, if you read: http://en.wikipedia.org/wiki/VC-1 the wiki on VC1 (bluray video codec) you’ll see that at its highest quality setting (1920×1080 x 60hz, Advanced L4 profile) it can use up to 135 megabits / second.
            Though I suspect (guessing) most HD blurays are encoded at Main profile High (max 20 megabits, 1920x1080x30hz).
            None of these are lossless.

            I’m not claiming we require this quality over our internets. But guys, you are not informed.

            Additionally, my comment on broadband is not “we can’t do what we want on 12 or 25 megabits”. We can. But not everyone has it, and no solution from the liberals has answered how they will do it while at the same time safeguarding competition. Their current track record on that is “give telstra a vertically integrated monopoly for 10+ years and every now and then the ACCC might rap them on the knuckles if their competitors complain enough”.

          • I’d agree with will that a lot of people ARE stupid enough to upload uncompressed video to the web. In fact there would be a majority that would say “what is compression?”
            Most people have no idea at all and will upload whatever their camera gives them – usually rather big files.

          • Actually, because no one can agree on a video standard, the smartest way to put video on the web is push up a single version as losslessly as possible and let the web-host deal with the conversions, FLV for Flash, h264 for Apple, WebM for the rest of us, etc. It’s also likely if you’re pushing up 720 or 1080 video you’ll want to encode several sizes, making the total uploads much larger than the file with low compression.

            Look at youtube, it offers videos in 3 formats, at multiple resolutions. Space is cheap and abundant, bandwidth needs to reflect that.

          • Watching “1080p” videos on youtube (which is pretty low quality as far as HD goes) requires me to pause the video and let it buffer for a fairly long time. It also prevents anyone else from doing anything useful on the internet.
            This is on an 8Mbit adsl connection, which is the best that is available in my area.

      • “I’d actually disagree. A 10min 720p/1080p clip on my PC is nowhere near 800MB — you’d be lucky if it hit 200MB.”

        At lossy bitrates that may be, 800MB for 10 mins is exactly the kind of sized video file output by most consumer camcorders and a perfectly legitimately example of the kind of transfers that an ordinary consumer could be doing when moving around videos that they’ve recorded.

  6. I remember, 12 years ago when everyone was saying 2.5meg per second was too much bandwidth, it would never sell and never be used. It’s only 12 years later and now we are talking 5 times that. It’s not the things we use today that you build for. If you are building for what we use today you are years behind from the start. It’s what hasn’t been invented yet that you are bulidng for. And for me 100 Mphs barely makes the bar. The truth is any bandwidth bar set will ultimately be used and exceeded.

    • They also said that same thing about CPU speeds … and yet today most PCs are not limited by CPU speeds, but by graphics card processing power and the speed of hard disk storage, if they are truly limited at all in terms of the applications they use.

      Today, much CPU development is going into low battery usage cases such as the iPad … not higher processing power. This is not just a straight line into infinity. It is based on use cases — not sheer speeds and feeds.

      • That’s not strictly true. CPU speed increases over the last few years just haven’t focused on MHz.

        • Well, CPUs are mainly heading down the path of parallel processing … however, of course software hasn’t kept up. I don’t think giving desktop CPUs more power (or desktops in general more RAM as well) is going to solve any bottlenecks at the moment. Our hardware is running ahead of our software.

          I used to upgrade my gaming PC every six months or so. Now I don’t need to bother — because hardware these days is so great, the software hasn’t yet caught up.

          • I was mainly speaking to your second paragraph. There are still a few CPU bound applications but you’re right that in the main we are GPU/HDD limited.

          • CPU is actually a very good example of fiscal limitations of hardware. The top CPU speed of a single core has barely moved in the past few years (we hit like 3.4 like 4 years ago).

            Instead mass parralellism is what has happened, and there are very few applications (due to how incredibly difficult it is to make multi-threaded applications that are non trivial) which actually use all of the cores

          • “CPU speed of a single core has barely moved in the past few years”

            As I said, CPU speed isn’t all about MHz. There’s a number of things they do to increase CPU speed outside of clocking things higher. More cores, more cache, better designed cores…

            “due to how incredibly difficult it is to make multi-threaded applications that are non trivial”

            What on earth makes you think that it’s incredibly difficult to make multi-threaded applications? There’s nothing difficult about it whatsoever. The key is identifying where parallelism can be used in an existing algorithm or developing new algorithms that incorporate parallelism at their core. The biggest issue with this at the moment is that existing codebases were written without parallelism in mind and large companies are loathe to change things until they have to.

          • Try doing multithreading in non trivial programs, most consumer level programs are still single threaded, and at most do predetermined threading (so they only use a couple threads)

            Clearly you havent done multithreaded programming, because semaphores/mutex’s/locks is what makes it so hard (and using such methods completely breaks abstraction in programming)

          • “Clearly you havent done multithreaded programming, because semaphores/mutex’s/locks is what makes it so hard”

            I’ve been doing it commercially for the last 15 years.

          • There is a lot more to parralelism then just recoding algorithms to use multiple cores, because most programs are not just algorithms (things such as format converters are what I would call ‘trivial’). Programs are not just algorithms (although some are)

            And reason why companies don’t change their code bases to do parallel programming is because of this difficulty.

          • You originally said “multi-threaded applications that are non trivial”…

            Anyway, this is all OT.

          • “I used to upgrade my gaming PC every six months or so. Now I don’t need to bother — because hardware these days is so great, the software hasn’t yet caught up. ”

            This is a straw man. The reason you don’t need to upgrade your gaming PC at the moment is simply because the current console lifecyle is imposing a ceiling on game performance capability.

            The current generation of consoles has already lasted longer than any before it and because of the weight of popularity of console game sales, all games are currently created with the lowest-common denominator of PS3 and Xbox 360 in mind.

            So a current gaming PC needs only be as powerful as a PS3 or Xbox 360, which even a 4 year old machine can be.

            This is the reason ‘software hasn’t caught up” with respect to gaming and its days are numbered. PC performance potential is almost at a breaking point where developers will soon stop holding back.

            Battlefield 3 potentially marks the start of that movement, lets see if you still think you don’t need to upgrade to play that.

          • “Battlefield 3 potentially marks the start of that movement, lets see if you still think you don’t need to upgrade to play that.”

            Yes that is a massive percentage of the population pushing those technical boundaries, the game most people play is Freecell, but never mind.

          • “I used to upgrade my gaming PC every six months or so. Now I don’t need to bother — because hardware these days is so great, the software hasn’t yet caught up. ”

            back when the upgrade cycle for a gaming rigs were half yearly or yearly directx was also released in a similar manner, this was because the programmers would utilise every last ounce of power they could from current systems.

            Whats changed since then?

            Well its not that those coders can’t nor don’t wish to keep pushing hardware rather they are no longer permitted too. The lowest common denominator in hardware atm is the current generation of consoles and since their take off (starting around ye olde XBox) they have to deal with a fixed point in time in regards to hardware (publishers love to release on multiple platforms).

            Did you update your gaming rig around the time the second generation of consoles were released perchance? there was definitely a major update in the DX version.

  7. You’re both sort of right. It’s really about the ‘network effect’ of everyone having a consistent and reliable high speed, and that CANNOT be delivered by any other technology than fibre, taking into account upstream speeds. With that in place, applications will come, without question.

    Today’s broadband is like peak hour traffic on Victoria Road and the Iron Cove Bridge. Only ever as fast as the narrowest point.

    • “everyone having a consistent and reliable high speed”

      Hmm. I appreciate the comment and what you’re saying, Dan, but again, would you be able to please give some concrete examples of applications which require the sort of sustained throughput you’re talking about? I think it’s important to keep this debate to concrete examples.

      • I don’t actually think that looking backwards in technology terms is the right way to look at this. Here’s why:

        – In the first few years of broadband availability, the vast majority of people couldn’t see why they’d need it, because dial-up was perfectly fast enough for their emails and web use of the day (which probably meant internet banking once a week and a few Alta Vista searches a week).

        – Before home VoIP became possible (adequate broadband speeds and provider availability) people were happy enough paying 40c a minute to call from Sydney to Melbourne.

        – Before ADSL2+ became commonplace, most people were perfectly happy to traipse down to the video store to rent a video and pay the associated late fees.

        – Before the iPad came out, there was negligible demand for tablet competing. (See how spectacularly unsuccessful all Windows tablets had been, for example.)

        – Before the iPhone came out, the vast majority of people loved their Nokias and couldn’t imagine anything better.

        – Back when I started at APC in 2003, there was a prevailing view that PC CPU’s were ‘fast enough’ — you could run Microsoft Office at a good clip on them, and if you had a decent graphics card you could run the latest games.

        – Before HSPA+ Telstra Next G, people had no idea you could really work effectively from a wireless connection at speeds close to home broadband. (Yes I realise the irony of that example, in the context… but no wireless network is a competitor for the rock-solid reliability of a modern wired network.)

        My point is… when it comes to technology, the general public, outside of the tiny technology enthusiast field, do not cry out for new enabling technology. But they sure do appreciate it when it arrives, and when it is done right.

        It’s the “done right” part that I think is the key thing about the NBN. It’s what the iPad is to Windows tablets.

        I don’t pretend to be able to predict the future of technology (and I do think many of the examples given in the NBN promotional video are a bit absurd), but my personal hopes for the impact of the NBN are:

        – That given it is a government project, there will be an impetus for the government to make all government services available online, including face-to-face consultations with government workers.

        – That population stress will be eased on major cities because it will be easy to telework from any remote location using a reliable, non-fluctuating connection. (Wireless _does_ fluctuate, and ADSL speeds are a crapshoot on a premise-by-premise basis.)

        – That more two-way internet applications will become available, taking advantage of consistently low-latency, high upstream speeds. (I am personally disappointed that the basic NBN connection is limited to 1Mbit/s upstream.) Currently, the internet is by necessity architected around one-way download applications due to the highly asymmetrical nature of connections to end users.

        I also don’t agree that any current broadband technologies are capable of providing adequate performance when it comes to upstream speed. Optus’ 100Mbit/s cable, for example, can only provide 2Mbit/s upstream to each user.

        My biggest hope, really, is that the NBN will provide consistent broadband speeds right across Australia. Without that, it’s prohibitive for “IP workers” to move to country areas where their broadband options will be limited and variable. As a result, we’re all forced to live in city areas and pay ridiculous house prices, or live in dowdy regional areas like Geelong or Bendigo where decent ADSL2+ is likely to be available, but you’re living in a ‘mini-CBD’ anyway.

  8. Personally, I could make good use of that bandwidth just on my own. The idea that I could have my HD TV using video on demand to output a 1080p picture in the background while I work on a client server as if I’m connected to their LAN, while at the same time I can have downloads running in the background has me frothing at the mouth.

    What is also being ignored is the fact that the need for increasing bandwidth has happened at a fairly constant rate since the 80s. Are the NBN detractors really putting forward the idea that we are now at a point where our need for data will no longer increase, simply because a network is being built that they don’t like? It beggars belief.

    I highly recommend this video of one of the NBN engineers busting some of the myths around the NBN.

    http://t.co/8dKiatQ

    • Hi Brad,

      “The idea that I could have my HD TV using video on demand to output a 1080p picture in the background while I work on a client server as if I’m connected to their LAN, while at the same time I can have downloads running in the background has me frothing at the mouth.”

      Appreciate your comment, but just curious — what sort of server are you thinking about this being streamed from? Because I currently stream 1080p to my TV from YouTube almost every night (StarCraft II matches, for example), while doing a whole bunch of other stuff on the net. And of course, most client/server stuff is optimised these days, requiring limited bandwidth.

      It’s not a problem for me — just curious as to what internet connection you currently have that doesn’t allow you to get all this done?

      Cheers,

      Renai

      • I have a very noisy ADSL line (due to AM radio interference, we briefly discussed it on one of your other articles) so I barely get over 2Mbps. I’m actually just out of the area marked for the NBN Seaford second-stage build, so I’m hoping the 3rd stage extends that area northwards before the Libs get in and wreck my chances of getting half-way decent internet in the next decade.

        Selfish, I know. But the only valid way I can objectively assess the various plans is how they will affect my ability to do my job and how I want to use the net recreationally.

        • theres the kicker aint it. You have a 2M connection. What if you had a 12M connection? or 25/2 connection? I bet you that would be enough for you. So the 100/40 is NOT required, it would just ‘be nice’.
          Now dont get me wrong, i want the NBN to happen, and i use the net a lot, but not for the kinds of things so far mentioned. The only reason id go as high as 25/2 is for the extra upload speed.
          I just dont really think that we need the 100M speeds. Not for a long while yet.

          • How long is that smurf? The NBN won’t be finished for 8 years. Will you need higher speeds by then?

        • This is not really an argument for the NBN, your problem would be fixed with the coalitions plan as well, which would mandate a minimum 12 download speed

      • Renai, I actually agree with you in principle that spending x billion is very expensive.
        However:
        a) the current technology DOES NOT stack up. If you are lucky (like I used to be, in Mentone), I was able to get 16mb off my IInet ADSL2+. then I moved house to Mulgrave (so no, im not “miles from anywhere”), and I now sit 3.4KM from the exchange…with a great 6mb. I also have a Telstra NEXTG ultimate for work (their large square blue one) which notionally has 24mb+ (I think its actually over 50mb from memory), except at peak times when it drops back, with increased packet loss and time outs, and (if I was paying for it) is so prohibitivly expensive that its not feasable to use on a personal level. that 1080p Starcraft II match you talked would eat the monthly download on most plans. the announced NBN max speed pricing is cheaper than current telstra next g pricing….

        b) Out of interest Renai, what connection (and if DSL, what distancespeed do you pull)?

        c) Having Turnbull come out and say “yes 24MB is sufficient”, yeah he is right…except 80% of people dont get anywhere near that, and he doesnt seem to realise that just because they advertise 24mb, doesnt mean we need it….. I would personally prefer if we had gone with a Fibre to the node (like originally proposed by Labour, but blocked by Liberal)…now Liberal’s have dredged that policy up and claiming it as their own….. (note, Im a swing voter, I think both parties are terrible, I just happen to agree with the NBN as a policy for some action…if they had gone with NBN to the node I would also have supported that..the fact liberal has now reversed their policy concerns me)

        If all metro population could get 24mb I would concede the NBN is not needed. the fact is most people cant get it, they can get “up to 24mb” which is code for “nowhere near that, but it makes the advertising look better” (and yes I know the tech reasons why with ADSL does this)

        Humerous Side note: I note that laptop turnbull is using has a ethernet cable in it. Im betting he hasnt restricted that down to 24mb, and is actually running it at 100mb or 1gb….I wonder why. Could he possibly have departmental applications that need faster speed……..

        • also side note: Re: your comment above about limitations on CPU’s development cycles etc.
          I remember playing PC games over the net on an old 28.8K modem, getting game demos and such in hard format (Disks, then CD’s the dvd)….The only time I saw “matches” was at LANS, or when a mag happened to include a copy of them…

          that was 1997, pre the days of Optus Cable and mainstream DSL and that was the norm. We eventually ended up with a duel line ISDN for work purposes….I still would not trade that back for my ADSL…as I will say the same when I get NBN…..

          I wouldnt care to go back to those days….and that was only 13-14 years ago…What will we be using in 14 years time?

      • But you’re not spending any more money than today. You already pay money for broadband, but it mostly goes to Telstra’s profits. Instead now you’ll pay the same exact prices to build an infinitely faster network that will be owned by Australians. How can you call that “expensive”? You’ve seen Exetel’s pricing. Why are you being so stubborn?

  9. As I’ve said before upload is much more important than download and with less than 1mbps there is not a lot you can do, I could FTP files or video conference but I cant really do both at same time effectively. So then you have to ask the question what is the bandwidth you need to do both effectively, Turnbull will no doubt say 5-10mbps (Funny how that number will coincide with the technical limits of FTTN) but is this any reason to aim so low? 40mbps beats 10mbps, saves everyone time and money, means my computer is on a quarter of the time uploading thus saving electricity. It is the same with downloads I think the criticisms of Turnbull here as just as valid.

    • “upload is much more important than download”

      On what basis do you make that claim, given that far more traffic is driven through downloads, than uploads?

      • On what basis do you make that claim

        Upload speeds have always been a limiting factor. It’s not like someone made a commercial decision to make ASDL2+ 1mbps or slower that is simply the limit of the technology and the copper…

        given that far more traffic is driven through downloads, than uploads?

        Amazing that by some bizarre twist of fate that the copper connected to our houses manages faster download speeds in comparison to the upload speeds and more traffic just happens to be driven through downloads.

        • “It’s not like someone made a commercial decision to make ASDL2+ 1mbps or slower”

          Actually I’m pretty sure the creation of the various ADSL standards was precisely aimed at finding a commercial way to deliver affordable bandwidth (providing the primarily download-oriented services which most people use) to the masses over the existing copper network.

          • So if it were a commercial decision to limit the max speed to 1mbps then what was the real max achievable? (and yes I’m aware of Annex M, I’m talking plain vanilla ADSL2+ here) Also if that is the case why not limit the upstream to 512kbps or even 256kbps? We are just downloading right so really we only need just enough to make those requests for the download, just like dial-up…

          • People predominantly download that’s why the mainstream broadband is asymmetric. You’re correct in some respects though that upload seems like the problem in that some people would find the upload speed a bottleneck with no alternative option. Max achievable is dependent on how much of the total bandwidth you want to assign to uploads but would probably require different hardware in the exchanges. At least that’s my take on it…

          • People predominantly download that’s why the mainstream broadband is asymmetric.

            Yes I understand that Justin, my original point was you are dealing with the technological limits of the copper first not the commercial considerations.

          • There is no demand for upload on residential services

            Proof?

            HFC is able to offer upload speeds far greater then the current 4-8mb on the Telstra/Optus speed packs

            Telstra/Optus however don’t offer the service, because there is very limited residential use for it.

          • There is no demand for upload on residential services

            Proof?

            HFC is able to offer upload speeds far greater then the current 4-8mb on the Telstra/Optus speed packs

            LOL this is your proof? Sorry but you cant claim there is not demand for it base on this, especially since people simply don’t have the choice, they look at the plans available and make a choice from the what is offered not what they cant get and another thing “use for it” and “demand for” it are two different things so don’t try to obfuscate the issue.

            Oh and btw due to this new insight we can disregard the coalitions patchwork plan as well since there is no demand for more upload speed 5-10mbps would be pretty useless right, so they should come up with a new spec that decreases that upload speed and improves the download speed.

          • Hubert: You can divide a DSL line up between upload and download any way you like – your total of upload and download will stay the same. (e.g. you have 25mbps total you can divide that as 1 up 24 down, 5 up 20 down or 24 up 1 down)
            ADSL is just a standard way of dividing a DSL line in a manner that was considered useful for most residential users. There at least used to be SDSL available as well (have not seen it advertised anywhere lately but have not been looking either, I would assume they are still offered) that divides a DSL line equally between up and down speeds. In conjunction with a VPN, businesses use SDSL for branch offices sometimes.
            Technically speaking, the download speed vs the upload speed on ADSL or ADSL2+ is arbitrary, there is no technical reason this ratio need be used. It was selected by whoever it was who came up with the standard as being the most useful upload/download ratio for most people. Obviously the most useful ratio for most people is not going to be the most useful for some other people.

          • Deteego, only recently has DOCSIS 3 come into existence, and cable ISPs have yet to perfect upstream channel bonding. It will still take another year or so before they can deploy it, and even then it will not even be half as fast as download. Furthermore all upload and download bandwidth will be shared among hundreds of users on a single cable node.

          • Current services are download-oriented because that is what the technology supports. It is not the other way around.

            ADSL2+ has a 1M upload limit because it was advantageous to use the same upstream band as ADSL. ADSL has a 12000/1000 limit because AT THE TIME it was assumed that downloading would have more demand than uploading. Since then, the services in use have simply followed that technological limitation. Changing a standard once it is in place is very, very hard.

          • You have no idea what you’re talking about. Leveraging copper to send data makes asymmetric speeds an absolute necessity, with downloading being the dominant side. It’s a technology limitation, not a purposeful decision.

      • “On what basis do you make that claim, given that far more traffic is driven through downloads, than uploads?” Sorry mate but that is circular logic, the *REASON* there’s more download traffic than upload traffic is *mostly* because there’s a asymmetric network capacity to residential (ADSL, wireless) users. There has not yet been any sane sensible argument fornot backing the deployment of the NBN. YES It’s a lot of money and YES we ned to be careful the money is spent well. But EVERY argument is either (a) blatant lack of foresight (ie there are no needs for such speeds *today*) (b) lack of hindsight (by the same arguments of why we don’t “need” the speeds of the NBN, we didn’t “need” anything more than dialup speeds, in fact we didn’t “need” the internet at all) (c) blatant pandering to existing industry players (seriously, wireless? HFC? anyone with a clue join me in laughing out loud – one will *never* scale to speeds that a fibre infrastructure will reach, and the other *fails to fully support a saturated customer base*)

      • I think you screwed yourself on that one, because my download is someone else’s upload. If their upload is slow then my download can’t be any faster than their upload – simple maths. If I pour 10 litres a minute in to one end of a hose I can’t get more than 10 litres a minute out of the other end, as I said simple maths.

        And as for not needing 100mb bandwidth, let me see – I suppose you only have a 100 Mb hard drive – NO. When did you find out you needed a bigger one, before you purchased it or after you filled it.

      • Applications are designed for the network that exists. How is it difficult for you to understand this incredibly basic concept? How can applications that make use of high upload bandwidth be popular on a highly asymmetric copper network? Yet even despite all of those limitations P2P applications have become incredibly useful to homes and businesses alike. Seriously what is wrong with you?

  10. I think it’s important to keep this debate to concrete examples/RL.Isn’t the point no one knows,it’s all about the possibilities.History is littered with naysayers and conservatives,well I say blah.In a few years people will look back at this and wonder what the fuss was about.The sooner the thing is built and future generations can get their teeth into it the better.And further more Renai mate there is starting to be an element of the trojan horse about you.

  11. Let me just throw up some credentials. I first installed Fibre Optic Cable in 1984 for a couple of niche projects… So , unlike 99.99% of the population, I actually have experience with the technology.

    Secondly, I have in the last designed Data Centre for a living (and about to do it again). I have also been a designer of Video on demand systems, and telephony systems, and as such have a better than average feel for demand on networks.

    The issues that are likely to face us are not bandwidth – The main concern is latency, and the NBN does nothing to address that. In fact, it makes it worse. Latency is of course, the delay in getting information from a to b – If you are running say, telemedicine or videoconferencing, getting signal jitter hs nothing to do with Bandwidth you can switch from 10Mbps to 1Gbps and still see the problem. It is about the round trip time. And don’t give me that “speed of light” stuff. Signals do not travel in Fibre at the Speed of light, the travel at a speed determined by the internal refractive index of the fibre. Fibre needs to have repeaters every 50 kms, so an intercity trunk, will actually beslow because of all the stops, into amplifiers and on again. Of course, once you get into the “burbs” it becomes a “rats nest”. with a gazillion patches and breaks, all affecting latency.

    While on the issue of “Fibre” NBN are currently trenching across Australia at a rate of knots to lay inter-city fibre trunks – poblem is, Telstra, Optus and others have already done that. Last time I looked, only 10% of the Telstra fibres between Sydney and Melbourne were “it” i.e. active. It should also be noted that the NBN is being laid in the ground without conduits – BAD move – ground heaves, subterranean animals, errant council diggers, and you have sliced the fibre – every time that you slice it, you splice it. Every splice reduces bandwidth and latency.

    I had to laugh, last year there was a Dog and Pony Show about the NBN, and Queensland Health (there is a hub of expertise…) tried to explain how the NBN helped in remote medicine, someone had fallen in a country town, and needed diagnosis. They actually used a Mobile Phone to do that. They then went on to “pretend” that they were dispatching a helicopter, for this urgent patient… Problem was, that said town was about twice as far from the coast as the range of the government helicopter ambulance… (And the round trip, including two refuel stops) would take over 12 hours…. At least the government could try to do credible examples….

    Of course, having the bandwidth is nice. But already the pricing models are showing that it willbe expensive. Then, if you look at any new housing estate, the builders are being TOLD to include it, at a cost of around $5000 per house. Add to that, your carbon tax increases in running a house. But wait, we, the taxpayers will still be paying off, in our taxes the $50Bn or more that it will take.

    Oh, any by Conjob Conroy’s own admission, it could take as much as 15 years to roll out.

    It would make fr more sense to have FTTN, and offer extra services for those that want them (and are willing to pay). FTTH is the expensive bit (well, as well as the redundant trunks (given the amount of fibre already running interstate), and the ridiculous salaries of the NBN executives, and the need for at least three government ministers to turn up, every time they roll out a village of 2000 people.

    • “The main concern is latency, and the NBN does nothing to address that. In fact, it makes it worse.”

      I’m sorry, but you are technically incorrect … latency is in fact much better on the NBN — between 3ms and 10ms, according to early testing. It’s one of the network’s strong points. The fact that you do not appear to understand this undermines the rest of your comment.

      • Its due to network topology, nothing to do with the medium

        All circuitry, as an example inside your laptop, is copper, because the latency is actually a lot lower then any fiber optic equivalent (among other reasons)

        • Network topology aside. 10ms >> 25-75ms you get on ADSL.

          The circuitry inside your laptop is a bad comparison example. We can’t make chips out of optics. If we could I suspect we’d see more optics involved internally. As it stands, changing from copper to optics and back again for ethernet would be similarly delayed).

    • NBN is not trenching any inter-city fibre. Some new long-haul links are being built by Nextgen Networks under the Regional Backhaul Blackspots Program – where there is no competitor to Telstra. The maps are on the Nextgen Website.

    • [If this comment appears twice, blame the server or something. Went missing first time around]
      Peter – NBN Co is not laying any inter-city fibre. That’s outside its brief, except in the circumstance where (a) there is only a Telstra trunk and (b) NBN Co is unable to negotiate an acceptable price.

      Nextgen Networks is laying new long-distance fibre links under the Regional Backhaul Blackspots Program. This is not NBN-owned, and exists to provide competitive links on otherwise non-viable routes – and has already resulted in new DSLAM deployments by retail ISPs.

      None of the RBBP / Nextgen fibre is on existing routes.

    • [If this comment appears twice, blame the server or something. Went missing first time around]
      Peter – NBN Co is not laying any inter-city fibre. That’s outside its brief, except in the circumstance where (a) there is only a Telstra trunk and (b) NBN Co is unable to negotiate an acceptable price.

      Nextgen Networks is laying new long-distance fibre links under the Regional Backhaul Blackspots Program. This is not NBN-owned, and exists to provide competitive links on otherwise non-viable routes – and has already resulted in new DSLAM deployments by retail ISPs.

      None of the RBBP / Nextgen fibre is on existing routes.

    • I’m sorry but despite telling us about you “credentials” your post soon degenerated in clear political anti Labor point scoring (FUD keywords “50 billion”, “carbon tax”, “conjob” etc) so nothing that you mention can be taken seriously as you seem to be mostly politically influenced.

      As Renai correctly also pointed out for someone who apparently “Knows fibre” claiming that the NBN will make latency worse is totalLy incorrect. check out any ping tests people have done on the NBN and compare to ADSL..it’s a fraction of ADSL! intra Australian pings well under 10msec! Please try that on you ADSL..

        • Router: route-views.ii.net
          Command: ping http://www.waia.asn.au

          Sending 5, 100-byte ICMP Echos to 218.100.43.10, timeout is 2 seconds:
          !!!!!
          Success rate is 100 percent (5/5), round-trip min/avg/max = 48/49/52 ms

          I’m on ADSL2+ as well in a country regional centre (pop 21000)

      • ” intra Australian pings well under 10msec!”

        I tested your theory, pinging from an iiNet core router plumbed with fibre from Sydney to WAIX’s web server in Perth (all connected via fibre).

        Router: route-views.ii.net
        Command: ping http://www.waia.asn.au

        Sending 5, 100-byte ICMP Echos to 218.100.43.10, timeout is 2 seconds:
        !!!!!
        Success rate is 100 percent (5/5), round-trip min/avg/max = 48/49/52 ms

        Try it for yourself, http://looking-glass.iinet.net.au/lg

        • You have to realize that most people that comment on the benefits of fibre probably have no actual technological understanding of what they are talking about

          fiber makes no difference on latency, it is in fact in some cases slower than copper because the light that travels through fiber usually has a speed of ~0.6c, where as with copper its around ~0.8

          There are obviously other overheads, such as routers and whatnot, but the biggest factor is distance.

          • Speed of light is a massive massive waste of time in these arguments.

            I have done the maths already, it is absolutely meaningless to mention the speed of light for a distance of 3km. (your house to the exchange). More than meaningless.

            speed of light:
            299 792 458 m/s lets call that
            300,000 meters per millisecond.
            lets call that
            300 km per millisecond. (in a vacuum)

            Lets call your copper perfect. it gets 300km / ms.
            It travels from your house to your exchange (the only part that’s copper) in 1/100th of a millisecond.

            Your fibre (which in worst case is about .6 of the speed of light) but lets be real nasty and call it 0.1 times the speed of light. This is horrible horrible fibre DESIGNED to slow light down.
            So, instead of getting between your house and the exchange in 1/100 of a millisecond (adding 0.001 to your ping time)
            It adds 1/10 of a millisecond (OH MY GOD).

            Now your ping time of 50.001ms on copper is now 50.1ms !! !

            now tell me, has the speed of light in fibre got even an iota of an effect on your ping time? seriously?

  12. I think you’ve skipped over a few things here Renai. Let’s take HD video for example. You can’t really equate HD on youtube to HD in general. Youtube significantly compress their HD video. Current video technology would already saturate the upload speeds of the maximum NBN speed and some would saturate the download speeds.

    A low end video camera these days produces HD video at around 15-20Mbps. Video capable DSLRs produce HD video at 40Mbps+. Blu-Ray has a bit rate of around 54Mbps, is it inconceivable that someone viewing HD video through some streaming service wouldn’t want blue-ray quality video? Then going to the extreme (for now) you have video cameras like the Red One which produce 4k video at 225Mbps-330Mbps.

    As for the structural separation of Telstra, well really it’s too little too late. It was monumentally stupid of the Libs to trust the market to get major communications infrastructure for the nation right. Our country’s size and population just won’t support non government telecommunications infrastructure, the market will always cherry pick and leave many areas with little to no access to decent connections. You already have the case now where areas in capital cities have no access to anything but NextG and it’s rip-roaring 1-2Mbps connections which periodically drop.

    • Good comment!

      “You can’t really equate HD on youtube to HD in general”

      Why not? I used to watch a lot of TV, now I watch more YouTube. It’s the same type of content … why can’t we equate them, as it’s the same use case?

      As for uncompressed video — as I posted earlier — I don’t think it’s useful to make an argument based on the fact that we should be transferring uncompressed video around on the internet. Why should we? It’s a waste of bandwidth and doesn’t represent a productive use of networks. All video online should and almost is compressed.

      “As for the structural separation of Telstra, well really it’s too little too late.”

      Why so?

      “Our country’s size and population just won’t support non government telecommunications infrastructure, the market will always cherry pick and leave many areas with little to no access to decent connections.”

      I agree, and this is why both sides of politics agree that for the minority of the population which lives in rural/remote areas, there should be subsidies and government-sponsored infrastructure to help out.

      “You already have the case now where areas in capital cities have no access to anything but NextG and it’s rip-roaring 1-2Mbps connections which periodically drop.”

      My apologies, but I haven’t had this experience with Next G — I’ve almost always found it to be excellent, especially compared with Optus and Vodafone. Of course, I don’t use it as my primary broadband connection.

      • HD Video

        They might be the same type of content, video, but the use cases are different. As has been pointed out by a few people YouTube video at 1080p is heavily compressed. I’m not suggesting that uncompressed video be used, all HD video is compressed to some level. The compression used is chosen based on a number of factors, for you YouTube it’s storage and bandwidth for their biggest use cases. But YouTube’s requirements would be very different to say an online video store that wants to provide the best quality video and audio to its clients. In this case they’d be looking at something closer to Blue-Ray. They’d do it now if the bandwidth was available.

        What about Photography/Video enthusiasts who want to store their videos/images somewhere remotely as a backup. In this instance further compression isn’t really an option, these are effectively their negatives.

        Then there’s the medical uses, here you wouldn’t want some compression artifact creeping its way into your files.

        Basically there are many different use cases for transferring video at higher bitrates than YouTube offer, and will most likely be more in the future.

        Telstra Separation

        Basically it should have been done from the get go. Yes, it needs to be done now but it’s not enough. We’ve had 11 odd years of having Telstra stall advancements to get the most out of it’s ageing copper network. Completely understandable for a corporation but not what’s best for the country’s telecommunications infrastructure. The separation alone is not going to be a magic bullet that will fix things, we need more of a plan than that and so far the opposition has offerred very little. Their plan prior to the election was a complete farce. While the current one is better I still feel that leaving this particular thing to the market isn’t the best case in our circumstances.

        While I like the idea of small government, there are certain things that simply need to be kept as publicly funded infrastructure. That will vary country to country but in Australia I believe telecommunications infrastructure (backbone at least) is one of those things that simply can’t be left to private organisations.

        Blackspots and Wireless

        As I said it’s not just rural areas that are in blackspots. I’m about 10km from the CBD in Brisbane and there’s plenty of areas around here that are relegated to wireless. Thankfully I’m not one of them, having said that, the line into my home is terrible and consistently plays up with wet weather. As for wireless I can’t say I’ve ever seen better than 4Mbps and it tends to drop intermittently. Latency is terrible. I accept that there are areas that are much better, I just haven’t seen them :-) Add to that the fact that it’s shared and the moment school finishes in those areas struck by wireless only the bandwidth drops dramatically.

        • All of the examples you can up with a fringe/nich ones that don’t justify an NBN.

          And even with (less) compressed version of video, it can all be done on things like VDSL2 and HFC

          Furthermore we are about to hit a graphical wall regarding both video content and/or games, just as we have with audio, so its unlikely this form of media will continuously increase in bandwidth as people are saying it will

          • Hardly fringe. How many people own HD video recorders these days. How many own DSLRs with video. How many watch blu-ray?

            What on earth makes you think we’re about to hit a graphical wall regarding video content?

          • The human eye has a limit to how much quality it can perceive, just the same with audio (and any other sense)

            The same reason is why DVD-A/SACD is basically a dead market (only an extreme minority purchase such formats), simply because most people cannot hear a difference above 44000/32. Hell most people are happy with MP3’s….

            Same deal with graphics, most people are happy with HD, and hell if you don’t have a big screen there is probably no reason currently to go above HD, which just leaves bluray (and one higher)

          • “Same deal with graphics, most people are happy with HD, and hell if you don’t have a big screen there is probably no reason currently to go above HD, which just leaves bluray (and one higher)”

            Quick! Call Apple and tell them to stop production of the iPad 3.

            We can also tell all those display manufacturers to stop making high res 27″ and 30″ monitors. They can just revert back to 1920×1080 and 1920×1200.

          • “Same deal with graphics, most people are happy with HD, and hell if you don’t have a big screen there is probably no reason currently to go above HD, which just leaves bluray (and one higher)”

            Quick! Call Apple and tell them to stop production of the iPad 3.

            We can also tell all those display manufacturers to stop making high res 27″ and 30″ monitors. They can just revert back to 1920×1080 and 1920×1200.

          • That doesn’t prove anything, I was talking about quality, not the size of the screen

          • That doesn’t prove anything, I was talking about quality, not the size of the screen

          • But there is so much more to a HD channel than a basic video bitstream that venders are likely to utilise. What about all the other value adding things such as 3D, multiple viewing angles for sports events, multiple commentaries/ audio tracks, dynamic targeted advertising, in channel purchasing and others that we couldn’t even imagine.

            This is why its a bad idea to only build something that is going to be adequete for todays needs.

  13. Renai, Comon, many people complaining already about the lack of bandwidth available NOW, statistics also back this up with the rise of data (ABS and so on),

    You and Malcom need to get of the drugs, and look at this logically, logically, you would want extra room, because if you don’t, you’d be stuck in the rock and hard place. Otherwise we are back to using the Road Network as an example of poorly designed networks. As Jamie Hyneman (of Mythbusters) says, a complex design is a mark of an inferior designer.

    Using any other technology is very hard way to archive efficient movement of data.

    A pure fibre based network is fast, simple to design, it can be built in stages (or modular design), fibre is fast and versatile, and secure.

    And no I don’t work for any Network company, i just understand the technology!

    People do not want to be in the same situation as RIMS, we need to remove these obstacles in stages.

    • “many people complaining already about the lack of bandwidth available NOW”

      Interesting comment — can you provide some data to back that up?

      • Gosh your blinded by a bat, how about a number of studies including the OECD stating that the Australia is falling behind rest of the world in terms of Internet?

        ABS stats state that bandwidth is on the rise.

        Also, the amount of complaints in regarding to the limitations of the current copper network, RIM, Too Far From the Exchange, Inability to signup, etc.

        Instead of asking for bloody proof, do your own bloody homework!

      • You’re a journalist and you don’t know of the complaints? Are you purposely being obtuse or are you now on the dole?

  14. This was a reply to a response above but will get lost probably, So you want us to wait until the apps are released and then constantly play catch up, how about we ready ourselves for the future so we can use whatever becomes available.

  15. Finally some realistic balance to this topic, thanks Renai.

    Today’s push is for mobility. Many, many that are and aren’t techically savvy are dicontinuing their household landlines altogether and rather than going ADSL or another hardwired tech are simply moving to wireless. Yes speeds aren’t that good and if you want to download media streams you’d be pushing it, but the fact remains that many day to day users of the Internet and related tech aren’t interested in huge bandwidths, but are keen for mobility.

    Yes, there are businesses that want highspeed connections and there will be a percentage of households that will want the same, either for necessity or for those who want bragging rights with their techy mates, but the vast majority couldn’t really care less for these super high speeds.

    Software advancements, codec developments and better compression for lossless degrading are all making things run efficiently and smarter using bandwidth we have.

    Let’s face it. We will never have anything run as fast as we want or as much as we want… Who’s recently stood in front of their microwave and started getting impatient because it is too slow. I know I have.

    I’m an IT administrator at a mine in the outback. The weak link is in these rural areas as I see it (we only have a Satellite link doing phones Internet access, and data replication to the city), Will there ever be fibre coming to our site? No, but there is the technology to bring a wireless terrestrial link with higher bandwidth. Do we need 100Mbps, no. 10-20 will be super!

    Then there’s LTE, and other new wireless tech making huge leaps and bounds – this is where I see the every day aussie will want a piece of this consistant reasonable wireless connectivity.

    • Cheers!

      I think what we can agree on here is that it’s the applications and end uses which are important — not the technology itself, as great as that is. I want fibre as much as the next man — but I can’t really think of a stack of things I could do with it that I couldn’t do now anyway in a slightly different form.

      In any case, I think perhaps the wider argument is not whether fibre is great (because it is), but what government policy in telecommunications should be. I’m still a favour of a minimalistic approach. Splitting Telstra, legislating for MDUs with HFC and subsidising the bush would appear to set up a framework which may result in quite a lot of further fibre down the track anyway.

    • I hear the claim many are ditching their household lines in favor of mobility thrown around alot, but have seen absolutely no proof to suggest this is the case, every single year, more and more DSL connections are lit up – look at the ABS statistics. In the past 6 months there has been a increase of nearly 250,000 DSL customers, DSL customers are not reducing, the are growing, every year.

      • Actually I am one of those few people. We use the Vividwireless through their gateway modem at home, and I also have a vivifi wifispot to carry around Perth. Both devices are paid for on the same plan, in fact I can have 5 devices linked to the one plan. While not for everyone this is the type of flexibility that wireless can offer. I have not had a home line for over 4 years now.

        I might take up the NBN if it comes to the area but it would have to offer a better plan along with the flexibility provided by wireless roaming, I suspect Telstra will be able to leverage of its existing 3/4 G network in this regard.

    • The claim that “many, many” people are switching from ADSL to wireless doesn’t stack up. ABS statistics show otherwise. Yes, there’s a huge growth in wireless but that’s expected, it started with a lower base. Wireless is complimentary technology.

  16. We didn’t even see the possibility of YouTube 10 years ago. Nobody thought of it until some years later. Now we cannot imagine a world without it. This is why we need the extra bandwidth now – we cannot even begin to imagine what it’s going to be used for in the future! Just like when building roads, we must not build for current demand, because as soon as the construction is over the infrastructure will be obsolete. Anything else will fail, much like Sydney’s roads are at present.

    • You’re right, but to be honest I think it’s a bit difficult to draw a direct analogy between the transport and telecommunications industries. Technology changes a great deal faster in the telco sector — only five years ago mobile broadband was a joke, now it’s an increasingly necessary daily item, for example. Drawing a road analogy, that would be like if flying cars suddenly appeared in five years.

      Like many people, I know that we need to invest in the future. However, as mentioned previously, I’ve seen too many hype cycles boom and bust in the past to do this blindly.

      • This isn’t a “hype cycle”. It’s a historical trend with 20 years of observational data to correlate with and future applications within sight, outlined by many posters on this thread. You’re being disingenuous to call people’s desire for greater upload speeds and control over their internet connections “hype”.

  17. “Let’s take the geologist mother. Does anyone seriously believe that it’s possible to remotely manage a complex mine site from their home office thousands of kilometers away? Doing so raises so many issues around safety, duty of care and even simple management philosophy that it’s not funny. In the event that there was an accident on site, how happy do you think the authorities would be that the facility was being overseen from Ballarat?”

    It might Surprise you, but companies are looking into this more and more. My brother is implementing such a system for a large resource company at this moment. They did look at making it accessibly from the home, which is very financially viable, but bandwidth for the process is just not available. I am sure if the NBN goes through this wil be re-assessed. This technology controls 30 sites remotely at their control centre, but alot of the functionality could have been used from remote login.

    Recently my brother was diagnosed with cancer, and has spent significant time at home while undergoing treatment. Had this technology been up and running, and the NBN in effect it would have saved the company and my brothers insurance thousands. In reality if we had this capability – without the common known boundries of copper – ie a fibre network that would be infinately upgradable when next gen tech comes along – it is not hard or that expensive to replace the tech et each end – just imagine the possiblities, and where money could be spent elsewhere when facing downtime due to illness, family needs or just to get cars off roads for the a day a week.

      • Mining teleoperation is old hat mate. They first starting doing that in about 1990. It’s actually more safe than you might think. While I’m sure you could Google it – basically it means noone dies when the truck crushes a vehicle, landslide wipes out the road, roof collapses on the mine, truck drivers who’ve made this trip thousands of time and are bored/drugged/tired/drunk/hungover. You don’t actually control everything in real time – for example basic load movements are fully automated, some slightly things are supervised, and anything tricky/out of ordinary is real time controlled. You certainly don’t mix humans and robots in close proximity on a heavy industry environment.

        It’s also helping to think outside of your box (ie content consuming internet user) – there’s still far more applications out there for SMEs to get into distributing/generating content. eg work at home graphic designer/architect/musician needs to offload content or receive content quickly as possible.

        • From the mining sector at University (and one of my colleges is actually doing heavy research into remote mining area) there are numerous implications in actually having the person do it from home (legal and otherwise)

          In all cases, the remote mining is monitored from some station owned by the company in the city, with all the professional equipment that they require (as my example, my college is doing research into having a full 3D virtual room for the mining exploration, no way in hell is that going to be done from home)

          • But NBN does not equal home user. It’s everything from home to SME to E. Think of it this way – instead of “Mining Co” flying in their latest crop of transporters from all over Australia (hell all over Asia Pac) to some remote end of Australia – you just get flown to a bog standard local “office” (which uses the NBN last mile) where you do your processing. Essentially they’ve reduced the mining job from a bunch of sweaty blue collars to a bunch of part timers who work around the clock. The NBN last mile enables them to setup cheap offices whereever – rather than flying people everywhere.

          • And I am saying these mining companies already have remote stations connected up to fiber for these situations. Anything above just streaming HD/3D (which does not require fiber) would not be done from home for various obvious reasons

    • Related to this – years ago (5’ish) I sat through a talk given by a person from Air Services Australia. He said they now have the technology to manage the European flight zones from the tower in Brisbane. If they can do that, why is it so hard to believe that a mother could view (where did it even say manage?) a mine from her home office?

  18. Imagination is the hard bit of this debate. Just about everybody argues from what they’re familiar with – “I don’t need 100 Mbps, therefore nobody does” – which is just as unimaginative as saying “I could use it, therefore everybody will”.

    Is it impossible to imagine innovation and ingenuity coming up with an application that suddenly changes peoples’ demands? – In the time-frame of the rollout, or the probable 30-50 years of the fibre’s life?

    To argue that today’s requirement is valid forever is to say that “everything that can be invented on the Internet has already been invented” (to paraphrase Charles Duell).

    It also ignores the huge gap between specification and reality. If 5 Mbps discount on the NBN is worth a storm, what about the typical gap between 24 Mbps and real speeds (my min-max sync last week were 3.5 Mbps and 7 Mbps, respectively from Petersham exchange). ADSL2+’s theoretical max line rate is not a good starting point for defining what we “don’t” need.

  19. Haven’t we discussed this stuff enough already? Let’s sort out issues like live animal exports while the NBN is being built. Did Malcolm foresee the iPad? Who can foresee the future for one-tenth let alone all of the life of the NBN? Why do we obsess over CURRENT applications? Bill Gates’ first attempt relied on a command line not WYSIWYG.

    Having a reliable system in our streets that does not fall over when it rains buried owned by a common access wholesaler sounds like progress to me.

    • I agree although there are many things to do with the rollout that are of concern. Labor have really made a hash of something that should have been their shining light.

      I think the focus on HD applications is a bad idea as it lends weight to the argument that the NBN is for geeks and those that like porn. Most people I know already have more than enough HD content downloaded to watch should they find the time.

      Reliability and infrastructure upgrading is the where the focus should be, over head cables are a big no-no IMO

  20. You suggest you watch 1080p videos on YouTube, but I think what many don’t understand is that is not real 1080p, real 1080p takes upwards of 20Mbps to stream, BluRay can take up to 40Mbps, and if you were to stream something in that quality, to one person, you would need 40Mbps upstream.

    You also mention Annex M, it is ‘up to’ 3Mbps just like it is ‘up to’ 24Mbps, I had Annex M for a while and only had 1.3Mbps or so upstream despite living rather close to the exchange.

    I feel you put too much emphasis on downstream speeds, and not upstream, same with Malcolm.

    But, one important point I’d like to raise is consider how long the NBN is going to take to get built, or even FTTN, 5 years ago noone knew what Facebook is, now huge amounts of people use it, upload videos, pictures – same story with Twitter. Even YouTube has only taken off recently.

    If you asked Malcolm 10 years ago, I’m sure he would suggest a 256kbps ADSL connection is hogwash and noone will ever need it, now pretty much everyone considers it as bad as dialup.

    One other thing I’d like to mention is you already have 16Mbps internet, that is far greater than the majority of Australians – the average is 8Mbps, you have DOUBLE the average, many people have connections that make yours look much faster. I am one of the lucky ones, but it is still a decent bit slower than your connection. To suggest that most applications existing today will run on our current networks is jumping to conclusions a bit, as ‘todays networks’ can be anywhere from the guy stuck on 56k dialup, to the guy who has the exchange 1 house over getting 26Mbps with a high speed profile enabled.

    • This is one of the MAJOR benefits of the NBN, it *guarantees* that any customer covered by the fibre footprint can have (assuming they’re able/willing to pay for it) 100Mbps down and 40Mbps up. There is NO other technology either in place or even in the process of being deployed today which will achieve such guaranteed capacity (ie *actual* not “in theory, when sliding down hill, on a greased pig, down wind and assuming you are three inches from the exchange/tower and *nobody else in your suburb* uses the same service”)

  21. The case for 100mbps is more prosaic than video conferencing etc. Storing and accessing data outside your premises. Backup, Netboot, fileserving. Maybe only 1 in 10 might greatly benefit or want 100mbps but you can’t predict where the 1/10 person lives or wants to live.

    Disruptive technology like the ipad and Facebook can happen within a year or 2. We’ve seen the hopelessness of rolling out what is currently needed rather than what WILL be needed in the laptops for schools program. Rolling out fibre reactively would be a massive fail given the long time it take.

    Anyone who administers a corporate network can instantly see a 100 small use cases for the NBN. None of them are compelling by themselves but the sum of the parts will produce a radically different society.

  22. The whole point of building infrastructure is to build for the future. Also, anyone in IT knows that with expanded capability comes expanded use and you don’t have to be IT proficient to fill that extra bandwidth.

  23. I have Telstra Cable 100Mbps at my old house.
    One night i actually managed to choke it.
    That involved the PS3 downloading an update
    The Apple Tv downloading a movie in the other room
    Me downloading a tv show off iTunes on the iMac
    and My sister on her laptop using Youtube.

    We moved houses and now have Telstra ADSL2+ with a sync rate of 17Mbps.
    Now if one person is downloading off iTunes, Sony, Xbox, Microsoft or anyone of the big names it hogs the connection and makes everything else slower.

    In a house with 4 people. a Ps3, Xbox, Apple Tv, iMac, 2X Macbook Pros, 2Xipads and 4 iPhones its impossible for all of us to be online at the same time accessing rich media and with Youtube offering HD and more and more online offerings of HD 25Mbps is no longer going to cover it.

    Another thing is the upload. Our current upload is a joke and with 25Mbps the best upload we can hope for is 20Mbps which with the way were moving to cloud computing and online backups is not good enough.

    Id like Turnbull to come to my house with his iPad and try and use it when the rest of my family get online.

    • Id like you to justify where your (clear) entertainment binge for such a price on the taxpayers.

      I have no idea why the government should be subsidizing your game/video watching habits, if you actually gave an example of you saturating a link for work reasons (and you are forced to do it from home) you might have at least a somewhat convincing case

  24. > why would anyone carry out all of these tasks simultaneously while manipulating a
    > complex three dimensional geological model?

    The reason a lot of things such as this aren’t done now is simply because the bandwidth to do them is not available Give users the bandwidth and they will start figuring out ways to do them.

  25. I think every one has missed the real point. The copper system is past its use by day, People now steea copper because of its $ worth. In replacing the error fault pron copper we may as well upgrade to something that is much better, and I am sure we will find lots of things to fill up all the extra bandwidth/speed that comes with being direct connected by long lasting none corroding glass instead corroding /failing copper.

  26. You deleted my last comment you poor sensitive soul an do it to this one too,’cause let’s face it you’ve been rolled,you liberal trojan horse.What blows me most is I came here often because I saw this as a balanced place where I could learn.Oh well.

  27. how come turnbull isn’t using wireless in that photo?
    i thought wireless was supposed to be the best thing since sliced bread?

  28. “Does anyone seriously believe that it’s possible to remotely manage a complex mine site from their home office thousands of kilometers away?”
    Its already being done.
    It will end FIFO, so its very carbon friendly.

  29. The irony is Malcolm Turnbull in the picture using a “free” taxpayer funded FIBER OPTIC connection.

  30. FFS Renai. You are just as bad as every other fucking moron that fails to read the whole article, twice, before getting your knickers in a twist and writing some bullshit ‘rebuttal’.

    Turnbull claimed that there were no applications TODAY that needed 100Mbps. Now you are claiming that “Stilgherrian has argued that the application which will soak up the 100Mbps bandwidth promised by the NBN is clear: Multiple streams of high-definition video to people’s residences”

    Today? Yes. In the future? No.

    Stilgherrian said that there are things, like the HD video example proffered by the NBN marketing video, that we have TODAY that would use up that sort of bandwidth. He then went on to say that to think that is the end of our requirements lacks “vision”.

    Hence the comments that have been re-tweeted to death comparing Turnbull’s lack of “vision” with Gates’.

    QED

  31. One killer application that everyone will eventually want to use is online backups. Everything is digital these days: photos, music, movies. And backing them up is essential. I know someone who recently lost all their baby photos of their 1 year old when their laptop was stolen from their house. You can use external media but an offsite backup is much more secure. I have ADSL2 and get about 750kbps upload and have 2Tb of data on my server. It would take 247 days to back up this data using ADSL2 but only 4 days using fibres upload speed of 40mbps. In fact a single Blu-ray would take 5 days on ADSL2 and 23 minutes on fibre!

  32. You get pissed on from a vast height but me an I guess a few other people get moderated?So it’s just about clicks and your ego?Meow.

  33. Given the value of some of the “3d rock strata” data, and associated values of energy exploration data, why the hell would anyone let their staff thick client it? It’s in the region of millions for even know where some companies are looking, and licensing of sites are in the billions.

  34. I suspect a troll. I hope your editor is happy with the increased traffic these negative NBN articles are receiving.

    You claim the example applications provided by the NBN are inaccurate but admit that there will eventually be other applications that will make use of the full bandwidth … fine, keep going with the build guys nothing to see here.

  35. In 1996 I sat next to a Telstra exec who stated that the maximum speed needed for the internet would be 64 kpbs. Some years before they predicted a mobile penetration rate of 6%. Do I need to repeat the prediction IBM made about the number of computers in the world? Why should you build a network that can only cater for what your (limited) imagination tells you is needed now?

  36. Interesting article. Makes a lot of sense in 2011, but unfortunately none of us have access to a time machine to see what bandwidth we’ll be using in 20-30 years time. That said, I’ll put my money on us needing 100M fibre by that time, and I’ll also bet that we’ll be thanking an ageing Julia Gillard and co for having the balls to build it. Technology moves much faster than our imaginations these days, and whilst it might be all well and good, and correct, to write an article criticising the current broadband policy for its promised speeds based on what we use today, I’m pretty much positive that it’s not going to apply in years to come. If you look back 30 years up until today, extraordinary things have happened with computing and networking. And whilst the line on the graph depicting progress in this area might not climb with the same gradient into the future, I’m sure it’s still going to be on the up and up for quite some time into the future.

  37. Why should you build a network that can only cater for what your (limited) imagination tells you is needed now?aussiearnie/bingo.Giddy up trojan horse.

  38. As is so common these days, everyone wants to give their opinion on matters which they simply don’t know enough about. I mean just look at the Carbon Tax, just about everyone has a very strong opinion on whether the Carbon Tax is a good thing or a bad thing, yet 99% of those people are incredibly ill-informed, and often look at things in a very short-sighted, narrow-minded, non-holistic manner.

    With the pre-amble out of the way, I’d just like to clear up some of the more significant factors being raised in this debate, revolving around bandwidth requirements of modern day applications. First of all, video.

    Video is one of those things that still has a lot of technological potential for growth. The future will see higher spatial, temporal and spectral resolution video. That means more pixels, higher frame rates, and more colours (commonly referred to as bit-depth; the amount of data allocated to storing colour information). Then there’s things like 3D which demand (in its current implementation) double the data as 2D content. So please don’t suggest that we’re anywhere near the plateau of the development of video, like we may be for audio.

    The word compression is also being thrown around a lot. First of all, there are two types of compression. Lossy and lossless. Practically all video is compressed in a lossy format. Every digital video camera on the market uses lossy compression, as lossless (let alone uncompressed) is still way too big for practical use, even by major film studios who. So considering all video content is lossy, there are a number of factors that contribute to the size of the video stream. The obvious one’s you may think are resolution and frame rate, but indeed the only important factors are the compression algorithm, and the bitrate, which is the amount of data required for an arbitrary period of time. The higher the bitrate, the higher quality the result, but the larger the data stream. The compression algorithm is also significant. As a general rule of thumb, the slower the algorithm, the more efficient, which results in a smaller video file. Video cameras and film studios generally use fast algorithms with a high bitrate, where as generally speaking distribution formats go for a slow algorithm to achieve lower bitrates while maintaining an acceptable level of quality. That brings me onto my next point, which is that what may be acceptable to you, but be unacceptable to me. You might be perfectly happy the quality of youtube, where as I think youtube’s quality is crap for anything other than sharing home videos and other such throw-away content.

    If not clear, my point on video is that you really can never have enough bandwidth for video (well, there is a plateau point, but we’re talking many terabytes per second). Let’s also not play down the significance of video. Video is the digital reproduction of our most important sense, sight. It’s used in all aspects of computing, so don’t think it’s only something for entertainment or video conferences.

    Another factor that often gets mixed into these debates is computers and computing power. A few of the comments in this article for example imply that processor cores aren’t get any faster, and so therefore chip manufacturers simply bundle multiple cores to achieve higher speeds. They then go onto say that this doesn’t translate to real-world performance improvements as multi-threaded/parallel software is hard to write.

    First of all, processor cores are constantly getting faster. I really hope you’re not using the operating frequency (e.g. the Ghz) as a measure of a processors performance, as that’s more an insignificant property, rather than a determining factor – it’s like using the colour of an athletes skin to determine how fast they are. I’d also like to point out that having multiple cores is very advantageous. Not only is parallel processing more efficient, it’s also much more practical from a manufacturing stand-point. Monolithic single-core processors are not impossible, they’d just be expensive, hot and slow, due to lower yields, higher power requirements and diminishing returns as the single-core processors incur more overhead and complexity, and spend ever more time waiting on data to process (such as fetching data from memory, etc).

    As for writing multi-threaded software, again, that’s not something inherently difficult. There are generally two factors which probably contribute to this wide-spread belief. First, many of the commonly used tools DO actually make writing multi-threaded software difficult, though these tools are normally decades old. More recent programming languages such as Clojure build parallelism into the core language constructs, meaning you almost can’t avoid making your program multi-threaded. The other factor that hinders the development of multi-threaded software is that some calculations simply must be calculated serially (e.g. in-sequence). Many compression algorithms for example are serial, though that’s only due to design decisions. If compression algorithms were designed to be parellel rather than serial (which a lot already are), then the requirement for serial processing would greatly diminish.

    Now that I’ve clarified those two things, here’s a list of applications which could always do with more bandwidth.

    * Video (of course) – Includes video on websites, IPTV, video conferencing, screen sharing, etc.
    * Remote Computing – This includes remote desktops(on which you may watch videos, play games, model in 3D, etc) and rendering (the resulting data of which can often be very large), among other things.
    * Online Backup – The safest form of backup is always remote backup. It protects you against both electrical faults and environmental hazards like fire and weather, which are the major weaknesses of local backups.
    * Data Sharing – For both entertainment purposes (video, music, audio, gaming, etc) and more professional applications such as transmission of complete databases, software (e.g. ISO’s of Windows 7 or OS X), etc.
    * Peer-to-Peer – This includes all of the above, but instead of client talking to server, or vice versa, it’s pc-to-pc. This will be huge when the NBN is completed. People hardly both with peer-to-peer communications simply because the upload bandwidth of DSL is so poor. Even sharing a high resolution photo peer-to-peer on an ADSL2 connection can take minutes; that’s just for one photo.

    Let’s also not forget those applications which can’t be foreseen, and not forget that the NBN fibre is able to carry a lot more than 100mbit/s – it’s only the devices at each end of the line which determine the maximum speed. Copper is obviously a different ball games and has many more hurdles to overcome in order to progress.

    • video, remote PC, online backup, P2P….. waffle and fluff.

      and exactly what proportion of the population is interested in these kinds of activities and how much are they prepared to pay?

      no other country in the world is spending this much money per capita on broadband upgrade because there is no generalised demand for these specialist, niche apps outside of the geek community.

      look at Japan, they have fibre everywhere, yet average monthly IP traffic is a measly 10GB/mth per subscriber

      • Urm, more, because no other country in the world is as sparsely populated as Australia. I’m willing to bet our road networks are among the most costly to run per-capita as well.

        Just because no one outside of the geek community is interested in them now, doesn’t mean they won’t be. 5 years ago most of the world didn’t want a smartphone, once Apple introduced one for the mouth-breathers, suddenly it’s commodity. What happens when Boxee, or GoogleTV (or whatever) take off, and can’t launch in Australia because our networks just can’t handle the load.

  39. Renai,

    There is only one argument for Mr. Turnbull’s proposal. And that is the now dubious claim that it will be better value. If this were not so, no one would want an outcome that is technically flawed, limited in reach and has a short use-by date.

    Otherwise you wouldn’t even waste your time trying to argue about usage. More so, you’ve avoided putting any real time frame on your claims. If you’re saying that most people, right now, don’t need or won’t use 100Mbps you’re probably correct. But that alone is unremarkable. If what you’re proposing is that the limitations of xDSL won’t come into play in the next 5 to 10 years, then you’re barking up a tree. And if you’re seriously ignoring the obsolescence built into a FTTN network then the game you’re really playing sounds like one of those infamous “no one will ever need more than..”

    We’re facing a stark choice.

    On the one hand, the government borrows money and invests it in a business that generates revenue. Thus generating an asset or alternately an income stream. In the time frame of 15 years the NBN pays for itself. That’s long, long before the physical network ages (which comes in maybe 60 years time). Yes, its a big step to take. Yes it has its risks. But it delivers a world class network. One which is future proof without any doubt. One that also delivers on ubiquity and apparently also delivers on value to the end user. But the key point is, the money borrowed for the investment cannot be borrowed purely to fund non-yielding spending. Or to put it very plainly, the NBN does not compete with other projects. Saving money by cheapening the NBN simply leaves us with something that isn’t as capable and probably does not have the same business case.

    Now for the alternative.

    Turnbull proposes to take what the tax payer owns (functioning parts of the NBN) and hand them to Telstra. He’s also going to find Billions in tax money and use that in further bribes and in other schemes and subsidies applying band aids around the edges. You can account for it any way you want but to build a FTTN network essentially requires half of the capital outlay of a
    FTTH network before you get to the Nodes. These require several extra Billion dollars. Then, there’s the big smelly elephant – the cost of the copper. Again, take away all the obfuscation and you’re looking at around $15 Billion – that’s Telstra’s own estimate.

    What once looked like “good value”, is likely to cost directly and indirectly most of what it would have cost to build a fully FTTH network. So, exactly what problem are we trying to solve anyhow? Its simply an exercise in inventing big scary numbers to scare voters with – people who don’t understand the basics of how the NBN is being funded and people who are easily fooled into thinking the alternative is “real cheap”.

    Given that Mr. Turnbull’s proposal – if it every got built as promised – would amount to a temporary network. One that at most defers some aspects of the cost of going to FTTH initially. Given that its likely to hand back some if not a lot of pricing power to Telstra and that this amounts to poor value for users (higher prices, lower speeds choose which). What’s the point?

    We can argue all day about usage but the reality is that in the future – whether that’s 5 years from now, or 15 – 100Mbps will be commonplace, and we will look back and recognise that copper was better abandoned sooner rather than later.

    Mr. Turnbull will never come clean about the “cost benefit” of his scheme. The reason is simple. He and the Coalition in general have no credibility. They must put something forward. Something superficially workable. Something that will prompt writers like you to step onto the bandwagon, perhaps because you’ve not through it through in detail.

    And if that hasn’t made the short lived argument over usage look pretty pointless, consider this. If the unthinkable happens and Abbott gets into power, and assuming we take Turnbull on his word (and that’s pretty laughable given his leader thinks any kind of truth optional) then what? 6 months for a CBA. Another 12-18 months to renegotiate with Telstra. Then Optus. Then the rest of the industry. A bullied, restructured or dismantled NBNco will take time. Now this all of course assumes a co-operative Senate. Not bloody likely. And for constitutional reasons they can’t pull a double dissolution until 2015. The most plausible scenarios go that they get nowhere and offer their policy as an election promise in 2015, or more likely they get some of the groundwork done and then promise to start building after an election in 2016.

    Why put years of hard work down the tube. And why risk going back to the situation where Telstra can name its price. As it stands, by 2018 Telstra will be effectively separated and Telstra will be just another retailer under the NBN. Turnbull, however nicely he talks about it, is still basically proposing to rip up not just the NBN, but all the progress made in the industry itself.

    In the end, Renai, for whatever motives you publish yet another take on the usage argument, its pretty much irrelevant, and from the point of view of someone standing looking back from 2030, just plain rubbish.

    • I would normally address your points, but your post smells so bad of pulling FTTN FUD out of thin air and anti Telstra hate (even though NBN will be the exact same as Telstra if it ever gets built, which it won’t) that there is no point

      And coalition getting into power is not unthinkable, hell most political commentators (even former Labor powerbrokers like Graham Richardson) are on the point of saying its basically inevitable

      Continue living in dream world, and I laugh at you claiming the coalition has no credibility, especially in comparison to the current Labor government

    • 1/ please learn to differentiate btw the needs of the average household and the average tech geek. the average household has ZERO need for fibre internet. as Malcolm mentioned in his interview, Telecom NZ is even struggling to up-sell from 10Mbit to 20Mbit.

      2/ in the real world, people do not build “future-proof” infrastructure, instead you build what you need for TODAY and incrementally upgrade as demand arises. this is how infrastructure investment works all around the world. you NEVER jump for the most expensive option. no other country in the world is spending anywhere NEAR the sorts of money on broadband that Labor is proposing.

      3/ NBNco is not a financially viable company. this is obvious as the Fed Govt failed to attract private investment in the NBN as was originally intended. research produced by investment analysts from the leading stockbroking houses have long assigned a negative (as opposed to low) NPV value to the NBN.

      4/ on top of this, NBNco is deliberately underestimating the true costs of building the NBN. the Analysys Mason report commissioned by the UK Broadband Stakeholders Group has shown that the cost of FTTP is five times the cost of FTTN. Gibson Quai, a local specialist consultancy in telecommunications engineering has put the real cost of the fibre network (capital works alone) at $60-80bln.

      5/ the physical longevity of white elephant infrastructure is completely irrelevant. all that matters is what VALUE the said infrastructure can generate to the populace in the immediate period. outside of tech geeks, the vast majority of Australians have precious little room in their household budgets to fork out more money for luxuries such as high-def entertainment or “faster internet”.

      6/ the faster NBNco is dissolved, the less taxpayer money will be wasted on this super-expensive white elephant political project.

      the Government shouldn’t be spending $50bln+ of precious taxpayers’ money on fibre extravagance to fellate the egos of self-styled broadband evangelists and the vanities of “public tech intellectuals” whose delusions of self-importance lead them to falsely belief that taxpayers should be funding their grandiose dreams.

      the vast majority of taxpayers do NOT want or need fibre broadband. hence, we shouldn’t be funding the selfish luxuries of a very small minority.

      the desperation of these well-known “fibre evangelists” and “NBNco cheerleaders” to justify squandering of public funds of INSANE proportions is obvious judging by how far (and low) they are prepared to go in twisting Malcolm Turnbull’s words in a vain attempt to discredit his reasoned and rational analysis.

      • “”no other country in the world is spending anywhere NEAR the sorts of money on broadband that Labor is proposing.””

        China Telecom have spend $150bn in the past 5 years rolling out FTTH all over China, they are spending ‘tens of billions’ more to get it rolled out to every city in China.

        Also, the NBN business plan technically suggests the NBN will cost $0 in the grand scheme of things.

        “”the vast majority of taxpayers do NOT want or need fibre broadband. hence, we shouldn’t be funding the selfish luxuries of a very small minority.””

        I would not reccomend putting your details into the following website, as some people feel it is being used to collect marketing data, however this has a petition that has been doing the rounds on facebook/twitter where you can vote if you want the NBN or not.

        http://www.nbnpetition.com.au/

        The statistics currently show 94.1% of people want the NBN, and for 60.5% of people it is a large influence in who you vote for.

        Also, consider that the NBN pretty much won Labor the election, if they had no NBN, chances are the independents would not have sided with them and maybe the Liberals would be in power.

        • “China Telecom have spend $150bn in the past 5 years rolling out FTTH all over China, they are spending ‘tens of billions’ more to get it rolled out to every city in China.”

          By FTTH you actually mean FTTN.

          “The “Broadband China, Fibre Cities” project will see the deployment of a mix of PON technologies connecting homes, office buildings and remote “nodes.” The nodes will then support fiber-to-the-node (FTTN) delivery of broadband services, wherein the final distribution of broadband signals to uses existing copper wires. Alcatel-Lucent will supply its Intelligent Services Access Manager (ISAM) product family, which includes a mix of GPON, EPON, and DSL technologies.”

          http://www.ict-oase.eu/?page=12&action=article&article_id=221

          I’ll repeat this bit in case you don’t get it.

          “wherein the final distribution of broadband signals to uses existing copper wires”

          “Also, the NBN business plan technically suggests the NBN will cost $0 in the grand scheme of things.”

          I meant to ask all the contractors and the NBN staff all the way to the top how they are surviving doing it all for nothing.

          “The statistics currently show 94.1% of people want the NBN, and for 60.5% of people it is a large influence in who you vote for.”

          ooh yeah that would be poll that represents a fair cross section of the population, would not be skewed by tech geeks by any chance would it?

          Let’s look at the polls based on proper population sampling, why is the Coalition leading in the polls by a substantial margin then?

          “Also, consider that the NBN pretty much won Labor the election,”

          No it didn’t from the voters point if view, the election was a dead heat with a hung Parliament with each major party getting 72 seats each.

          “if they had no NBN, chances are the independents would not have sided with them and maybe the Liberals would be in power.”

          We have covered this to death before, suffice to say the NBN alone did not get Labor over the line, there were many enticements Gillard brought to the negotiating table for all of the Independents.

          • In regards to China Telecom’s rollout, they are only doing FTTN in areas where FTTH does not make sense, in the large high-rise city buildings, it’s all going to be FTTH, this is where the majority of the Chinese population lives.

            Quoting a little further in the article:
            “This project is part of China Telecom’s plan to extend its FTTH coverage”

          • Yes but you said they were rolling out FTTH, they are actually rolling out FTTN, hence the reference to the use of existing copper, what is your reference that the majority of the China rollout is going to be FTTH?

            Be careful, they mix FTTN and FTTH terminology, they tend to call a fibre to the node solution fibre to the home, which it is in a way but not as in the definition of a fibre run all the way inside the apartment or residence to a ONT box, like the NBN here.

        • look at the chart at the bottom of this page:

          http://www.analysysmason.com/About-Us/News/Insight/FTTH_roll-out_Insight_Aug2010/

          you’re looking at USD600 per household max.

          also, go read this report:

          http://www.analysysmason.com/about-us/news/Press-releases/BSG-publishes-costs-of-deploying-fibre-based-superfast-broadband/Costs-of-deploying-fibre-based-next-generation-broadband-infrastructure/

          the cost of FTTH is approx. FIVE TIMES the cost of FTTN. we already know from the previous FTTN tenders that FTTN for Australia costs $10-15bln depending on penetration.

          the so-called “NBN business plan” is quite simply rubbish, both in terms of cost assumptions and revenue projections. it’s a 100% political exercise and (Alcatel interest-aligned) corporate gravy-train.

          *The statistics currently show 94.1% of people want the NBN, and for 60.5% of people it is a large influence in who you vote for.*

          go set up a website poll on how many people want Govt-funded and i guarantee you the results will be exactly the same.

          if people were so desperate for superfast broadband and pay-TV, the HFC networks wouldn’t be so underutilised.

          • HFC is mostly underutilized because;
            * Hardly anyone knows if they can even get HFC, for some, they have no idea it exists.
            * Even if HFC passes your premise, Telstra and Optus are picky in who they will wire up, apartments and units especially have huge problems.
            * Optus has a terrible international network, and Telstra are hardly competitive, if TPG or iiNet was allowed to resell HFC, AND advertise the fact that HFC actually exists and provide a easy way to check, I bet more would be interested.

            In regards to the site you linked to, it is talking about the UK, how are those statistics generated?

            It specifically mentions FTTC which is something British Telecom are doing, is the cost for FTTC including the fact BT already have the infrastructure to peoples houses needed, whereas the providers rolling out FTTH suchas Virgin Media need to build everything from scratch? (they primarily do HFC, but are also rolling out some FTTH)

            I’m sorry but the article you link to does not explain how the statistics are generated at all, it’s just a pretty picture with no real technical explanation.

            Does the FTTC include DSLAM costs? Is this PTP FTTH which drives cost up far higher?

          • *HFC is mostly underutilized because…..*

            oh, c’mon, that’s lame… you’re telling me Telstra and Optus spent billions of dollars building their HFC networks and couldn’t be bothered trying to maximise the network value by grabbing high-value subscribers where they can? that’s not even remotely credible.

            *In regards to the site you linked to, it is talking about the UK, how are those statistics generated?*

            go read the report. it’s done by a specialist consultancy that has international expertise in telecommunications.

            *It specifically mentions FTTC which is something British Telecom are doing, is the cost for FTTC including the fact BT already have the infrastructure to peoples houses needed, whereas the providers rolling out FTTH suchas Virgin Media need to build everything from scratch? (they primarily do HFC, but are also rolling out some FTTH)*

            they model all kinds of scenarios including savings from duct-reuse. even after factoring in savings from reuse of infrastructure, the cost of FTTH is still five times the cost of FTTC/FTTN.

            *I’m sorry but the article you link to does not explain how the statistics are generated at all, it’s just a pretty picture with no real technical explanation.*

            well, i’m sorry all the expert consultancy evidence that has been presented to you destroys all the arguments in support of the NBN (NBNco).

            *Does the FTTC include DSLAM costs? Is this PTP FTTH which drives cost up far higher?*

            the FTTC and FTTH costs include everything from equipment to civil works. the report also models PTP FTTH. most relevant is that GPON FTTH is five times more expensive than FTTC/FTTN.

            go read the report. it is entirely consistent with Gibson Quai’s comments that the capital works portion of the fibre network alone will cost $60-80bln.

      • “2/ in the real world, people do not build “future-proof” infrastructure, instead you build what you need for TODAY and incrementally upgrade as demand arises. ”

        I see. So this is why they design freeways with major structures notionally designed to last 50 years but in practice serviceable for hundreds.

        The reason you should build a network that’s good for 50 years, and serviceable well beyond that is, because you can. And because the reality is that to anything less is false economy.

        • well, is the Federal Government building freeways to service a nation with a population of 50mln (which we will probably eventually reach) as opposed to current 20mln?

          false economy? more like false analogy.

      • “2/ in the real world, people do not build “future-proof” infrastructure, instead you build what you need for TODAY and incrementally upgrade as demand arises. this is how infrastructure investment works all around the world. you NEVER jump for the most expensive option. no other country in the world is spending anywhere NEAR the sorts of money on broadband that Labor is proposing.”

        Oh god, that is a horrid horrid horrid paragraph and completely nullifies anything else you posted – infact I stopped reading your post after that paragraph as it fully shows your lack of understanding.

        Yes, people/companies pro-actively “future-proof” (or at least try too). They don’t install what they need “today”, because tomorrow it will cost them even more.

        E.g. Companies installing CAT6a and CAT7/CAT7a when its not required but doing it for “future-proofing”.

        There are many many other examples…

        Sigh.

        • *Yes, people/companies pro-actively “future-proof” (or at least try too). They don’t install what they need “today”, because tomorrow it will cost them even more.*

          no, they don’t. companies (which don’t go bankrupt) always select the least cost, minimum requirement option in evaluating infrastructure alternatives. overinvestment is the surefire way of going bankrupt.

          *There are many many other examples…*

          except you failed to provide a single decent example.

    • Hi ungulate, I see you are back, got all a bit hard over in the discussion ‘Turnbull’s new NBN policy is 90% win’ and you left a few points hanging unanswered, never mind let’s have a look at your latest ‘effort’.

      “There is only one argument for Mr. Turnbull’s proposal. And that is the now dubious claim that it will be better value.”

      It’s not dubious just because you say it is, it is just as as dubious as the multi billion dollar NBN without a CBA, but that’s a better sort of ‘dubious’ I take it because who cares we are building it anyway?

      ” If this were not so, no one would want an outcome that is technically flawed, limited in reach and has a short use-by date.”

      The basis of this statement that it is technically flawed is what, and what use-by-date do you have in mind and why?

      ” And if you’re seriously ignoring the obsolescence built into a FTTN network then the game you’re really playing sounds like one of those infamous “no one will ever need more than..”

      That coming from you is rich, you stated in that other Delimiter discussion mentioned above that FTTN cabinets cannot upgraded to FTTH, when shown examples that it can and overseas telco’s are rolling out FTTN with that upgrade path in mind if consumer demand requires it you disappeared from the discussion.

      All the millions of residences using FTTN all over the world, and telco’s that are continuing to roll it out must be really ‘suffering’ because they have FTTN which has a shorter build time.

      Australia in August 2011 is still tech tyre kicking trial use of the FTTH!

      “On the one hand, the government borrows money and invests it in a business that generates revenue.”

      It hasn’t borrowed anything yet, don’t forget you said ‘NBN bonds were available now’, when I asked for a link for the website you disappeared from the discussion.

      “Thus generating an asset or alternately an income stream. In the time frame of 15 years the NBN pays for itself.”

      Yeah right sure it will, all revenue predictions are based on 70% takeup of a service and 7% return on the investment.
      It is also based on the incredulous prediction that by 2025 only 16% of residences will be wireless only, as at 2010 that rate was at 13% and rising.

      ” But the key point is, the money borrowed for the investment cannot be borrowed purely to fund non-yielding spending.”

      What tortured semantic rubbish is that? – it means nothing.

      ” Or to put it very plainly, the NBN does not compete with other projects.”

      No if you give billions to Optus and Telstra to shut down their networks to ensure everyone is FORCED onto the NBN that’s one hell of a way to eliminate competition, and make sure no one can compete.

      “Saving money by cheapening the NBN simply leaves us with something that isn’t as capable and probably does not have the same business case.”

      Probably, maybe, so it may have a good business case but you are not sure?

      Never mind you have concluded that it will cheapen the NBN anyway even after admitting in the same sentence you are not sure, love the logic ungulate.

      “Now for the alternative.”

      This will be good.

      “Turnbull proposes to take what the tax payer owns (functioning parts of the NBN) and hand them to Telstra.”

      No he doesn’t he hasn’t said that is definitely what is going to happen.

      “He’s also going to find Billions in tax money and use that in further bribes and in other schemes and subsidies applying band aids around the edges.”

      More emotive clap trap bribes is it now, you do talk rubbish.

      ” You can account for it any way you want but to build a FTTN network essentially requires half of the capital outlay of a FTTH network before you get to the Nodes.”

      Where did you get that figure from?

      “you’re looking at around $15 Billion – that’s Telstra’s own estimate.”

      Estimate for what?

      “Its simply an exercise in inventing big scary numbers to scare voters with – people who don’t understand the basics of how the NBN is being funded and people who are easily fooled into thinking the alternative is “real cheap”.

      Oh the basics of how the NBN is being funded are not understood, that coming from you who said ‘NBN bonds are available today’ is hilarious.

      I assume you are including yourself in the ‘people who are easily fooled’ category?

      “Given that Mr. Turnbull’s proposal – if it every got built as promised – would amount to a temporary network. One that at most defers some aspects of the cost of going to FTTH initially.”

      Yes and that’s bad thing why?

      “Given that its likely to hand back some if not a lot of pricing power to Telstra and that this amounts to poor value for users (higher prices, lower speeds choose which). What’s the point?”

      Well the point is you don’t have a point, because the pricing power rests with the ACCC not Telstra on a monopoly network anyway, just in the same way the ACCC will determine pricing on the NBN by the NBN Co, in the exact same way the ACCC will determine pricing when the NBN is sold off, as Labor has stated it will be.

      “We can argue all day about usage but the reality is that in the future – whether that’s 5 years from now, or 15 – 100Mbps will be commonplace, and we will look back and recognise that copper was better abandoned sooner rather than later.”

      Yes we can argue all day about how the majority of residences don’t use the speeds that are available to them in 2004, 2005, 2006, 2007, 2008, 2009, 2010 and 2011.

      We can also argue why the highest speed infrastructure available today HFC has most residences giving it a miss.

      We could argue all day that if you have HFC available to you since 1994 why keep using 1500/256?

      “Mr. Turnbull will never come clean about the “cost benefit” of his scheme.”

      You mean exactly like Conroy and the NBN?

      “The reason is simple. He and the Coalition in general have no credibility. They must put something forward.”

      Yes that’s why if a election was held next weekend Gillard and Labor would be rolled, obviously the Coalition have no credibility!

      “And if that hasn’t made the short lived argument over usage look pretty pointless, consider this.”

      It hasn’t.

      ” If the unthinkable happens and Abbott gets into power, and assuming we take Turnbull on his word (and that’s pretty laughable given his leader thinks any kind of truth optional) then what? 6 months for a CBA. Another 12-18 months to renegotiate with Telstra.”

      Seeing we are still waiting for Labor NBN CBA, and the Telstra shareholder vote in October to ratify the Telstra/NBN deal your ball faced hypocrisy is wonderful to see.

      “And for constitutional reasons they can’t pull a double dissolution until 2015.”

      Huh what? where did you get this rubbish from?

      “The most plausible scenarios go that they get nowhere and offer their policy as an election promise in 2015,”

      The next election is due in 2013, your 2015 date as being the earliest of the Coalition could gain Government is wrong.

      “As it stands, by 2018 Telstra will be effectively separated and Telstra will be just another retailer under the NBN.”

      It will be just another retailer under the NBN irrespective of when structural separation takes place, BigPond could provide trial NBN plans tomorrow.

      “Turnbull, however nicely he talks about it, is still basically proposing to rip up not just the NBN, but all the progress made in the industry itself.”

      Turnbull never said he was going to rip up the NBN, only Gillard has said that’s what the Coalition would do, Turnbull has stated he will not at least twice.

      “In the end, Renai, for whatever motives you publish yet another take on the usage argument, its pretty much irrelevant,”

      Not as irrelevant as your no fact rant is though eh ungulate?

          • yes, ungulate, i’m sure you’re “honoured” to know you’re the biggest NBN FUD spreader in Australia. and that post of yours that 3 posters responded to was a word-for-word direct copy/paste job or cross-post from another website.

  40. I’ve worked in the IT & Communications arena for mining companies for about ten years now, I can’t see a whole lot of benefit to mining companies from the NBN.

    This whole remote operator mining is a joke, mainly because it’s already been done, not in the sense of a pimply faced youth sitting in Perth driving a 200 tonne dump truck down the ramp, but modern machinery is equipped to operate in dangerous area’s with a remote operator, there’s no reason why we can’t relocate the operator interstate if we had to, but at the end of the day for the last 20 years whilst this technology has existed the operator is generally within close proximity to the equipment being operated. I could run a single mode fibre down the hole tomorrow with 10gigabit of bandwidth, hook up the bogger’s operator console to it and control the bogger from Berlin, unfortunately the mining company is still going to insist that the operator stands 20 metres away from the machine they’re operating in the safe zone.

    Given the cost of other infrastructure mining companies build such as roads, rail, power lines, processing plants etc. if fibre to the site was really a requirement they would have done it by now, and may have, it shouldn’t be something payed for by the government, and if mining companies wanted to extend the reach of the network into employee’s home’s they’d do that as well if it was of benefit.

    At the end of the day, they have no shortage of funds, if something is of really such great benefit like enabling employee’s to operate from home they’d pay for it.

    And well you know what, I live in an area where I don’t public transport, roads are falling apart, I don’t have the option of reticulated gas for my house, I can’t even get a foot path out the front of my house, the essential services like police, hospitals, schools are all seriously under-funded in this town, my cost of living is increasing at a silly rate, I’m getting taxed more, my mortgage payments are increasing. I would rather see these area’s fixed before we roll out high speed internet to the nation.

    Also given the reason state of economic affairs I’d be one to save my pennies if I was a government in power.

  41. Renai – the Cisco Teleprecense solution in multi-party 1080p mode requires a hell of of a lot more bandwidth then the 1Mbps stated, refer to their own documentation (links at the end of post) where they suggest a minimum of 12Mbps of symmetric bandwidth to support 3 x 1080p video streams. I can’t imagine this type of video conferencing working over that ADSL2 connection you seem so fond of.

    I would also point out that just because you’re happy with HD video streaming from Youtube, that doesn’t necessarily satisfy everyone’s requirements. Last time I checked a Bluray movie is upto 50GB in size, and I personally would like the unfiltered video stream sent to my expensive HD TV… not the crumby compressed one that Youtube and other similar providers insist on sending. Again, try doing that on ADSL… you’ll be sitting down to watch your Saturday night move on Sunday night.

    Finally, I’d like to add I actually have the NBN (100Mbps/25Mbps). I was hooked up last Friday in Brunswick as part of the Iinet trial. The killer application for me is working from home (tele-commuting). I work for a large enterprise that doesn’t have a great remote access solution (no Citrix etc), and it’s just the ability to access documents on network shares as if you were sitting in the office. It’s not fancy, but it’s just one simple example of a task that’s normally painful (handing large powerpoint, Viso’s etc) that’s become a lot easier.

    And to tie all this back to the Coalition’s plans for Broadband in this country; the reality is that they have provided no plans as to how the country can move past ADSL2, or the 1Mbps upload limitation. The private sector “incentives” you mention would have to fairly monumental for any company to risk rolling out an alternate transmission mechanism (ie. not wireless) down your surburban street, with Telstra still owning the copper (and last time I checked they’ll continue to own the copper if the NBN gets cancelled, unless Malcolm is planning on buying it back for what will obviously be a huge amout of money). Quite frankly, I think Malcolm’s proposal is ludicrous (at least for the inner-city areas)…. and only appeases the liberal conservative voters who get scared when they say the $27Bn dollar sign attached to the NBN. Never minding the fact the spend is over 10 years, and the network will return an enormous revenue stream, and the basic fibre infrastructure is slated to last 40-50 years. Oh, and the annual defense budget is 27bn annually…. and no one seems to whinge about that.

    James.

    Cisco links:
    http://www.cisco.com/en/US/solutions/collateral/ns341/ns524/ns546/ns670/ns672/net_implementation_white_paper0900aecd805bbda0.html
    http://www.cisco.com/en/US/docs/solutions/Enterprise/Video/tpqos.html#wp1045102

    • You do realize that he did mention FTTN, which combined wtih VDSL2, easily provides speeds above 12/12, and VDSL2 is the same type of upgrade from ADSL to ADSL2.

      Your last paragraph is also complete tripe, some company has to own the copper, I have no idea why you have a problem with Telstra ‘owning’ it.

      • Last time I checked Malcolm hasn’t committed to anything above 12Mbps downstream under the coalition, so your VDSL2 example is a hypothetical scenario for a proportion of the population, and certainly not the basline.

        And If I have to explain the problem I have with Telstra continuing to own the copper network under a Coaltion broadband future, then it’s obvious you wouldn’t bother listening to the answer anyway.

        I’ll get back to enjoying my NBN connection, and you enjoy your 3G connection which I’m sure you’re very satisfied with.

        • Its about as hypothetical as moving from ADSL to ADSL2

          The only reason VDSL2 is not used in Australia is because it never got standardized for use in Telstra exchanges when NBN MK1 fell over

          And I am not using any 3G connection, in fact I am using a stable 30mbit wireless at uni

          • It was perfectly stable before it moved to the new Z system (and at that time the network was being over saturated since there were no usage limits)

          • I think you and I have very different definitions of “stable” deteego. Stable to me means I get consistent bandwidth within a reasonable margin of error. A saturated network would not be stable under this definition either now would it?

          • As I said, it was ‘stable’ according to your definition before UNSW changed to another IT system, at which point the wireless started fluctuating massively in both terms of speed and performance

      • Referring back to Telstra’s own schemes, they weren’t planning to install enough Nodes to enable VDSL2 type speeds except in some limited areas. The fact is that to enable VDSL type speeds rather than ADSL means using exponentially more Nodes (shorter distances). Something Telstra never intended to do and something Malcolm certainly will not fund, even if he is spending our tax money on it.

        In the end though, its a dead end. Like I said, if the author thinks that the limitations of xDSL won’t bite in the next 10 years, he’s barking up a tree.

        • Malcolm will have to place more nodes if he is actually going to upload his claim of 12 mbit download, increasing to 24 mbit in a couple of years after that

          • The math is a little complex, but if you want to double speeds with node technology you’ll need to add an exponentially higher number of nodes.

          • The math is a little complex, but if you want to double speeds with node technology you’ll need to add an exponentially higher number of nodes.

  42. Renai – I find your comments rather hypocritical considering that you said you were going to be on the 100mbit plan when the NBN rolls your way.

    http://delimiter.com.au/forum/national-broadband-network/95-what-speed-will-you-get.html

    If there are no applications as you say, then what are you going to use the plan for? And why bother having 100mbit LAN connections since there are no applications which will use it. May as well go back to 10mbit since this is just adequate for today’s needs.

  43. Renai,

    You love to draw a crowd don’t you. First off you support the NBN then you… kind of support it? There is no need to justify the NBN. When you used the Internet five years ago how much did you get done then? Now when you use it, how much do you get done now? Faster Internet means you get more done, in less time. So in the same amount of time you spent five years ago you now can do three times as much. The Internet is more responsive, quicker to access and you can get through more information than you ever could before. That is your productivity/cost benefit or whatever other label you want to put on it, right there.

    The NBN will have dramatic effects on the Internet based productivity of this country. Since pretty much every business will have a presence on the Internet then it will affect those as well. Especially since the access costs for them to have a permanent presence will drop significantly as a direct result of its construction. I know what this is like. I am one of the very few who have access to a 100MB/s cable connection and I can tell you our household habits have significantly changed having easy access to large amounts of bandwidth. My son watches ABC iView while my wife VPN’s to work and I download 7-12GB files. It all works, easily, without the issues you would face on ADSL. The NBN is not just about single user Internet activity because it is no longer just single use. We might have been the forefathers of high speed Internet but now its the children as well, at the same time, in the same house.

    Not some theoretical pipe dream in the future but right now.

    Kevin

    • you don’t need fibre to watch ABC iView. it works just fine on my “crappy” (but affordable) 6Mbit ADSL connection.

      • Sigh. That was but one of three concurrent activities that he listed. The point was, he has no bandwidth contention and doesn’t have to choose who gets to use the internet at a particular point in time in his household. Fairly basic point. On my “crappy” 5Mbps ADSL connection I *do* have to juggle bandwidth usage from time to time, and it is becoming more frequent.

        • worldwide evidence from the US to NZ tells us that few consumers will pay extra to upgrade beyond 10Mbit. this is why Verizon has halted roll-out of FTTH at ~5% and AT&T has halted roll-out of FTTN at ~60% of their respective markets.

          if you can’t multi-task sufficiently with Malcolm’s proposed 12Mbit mandated minimum, i would suggest you’re in the very very small minority of power users that should self-fund their personal digital luxuries instead of expecting taxpayers to pick up the tab.

          • What taxpayers? I thought the NBN had to pay its own way (despite the fact that the govt will reap returns from economic benefits). Besides, from what I’ve seen of NBN retail prices to date I’ll be more than happy to make the switch (and dump the PSTN phone for good), and I am currently on TPG ADSL so I believe I am one of those price sensitive consumers you’re so worried about.

          • You know what common-sense also says, it is impossible to predict the future because there will be factors beyond our control or understanding.

            Also, common-sense dictates that if a government project starts to be unsustainable the government will be forced to revoke the project, either of their own free will, or by being voted out.

            So stop throwing around statements like “permanent drain on taxpayers” because we both know that is complete bullshit, drawn from genuine concern about the possible unsustainable nature of the project, but still bullshit.

          • Thanks for the comment, but I suspect that it is more politics than policy that you are concerned about. Why else would you not be complaining about the taxpayer having to indefinitely fund broadband for regional/rural users as per the coalition plan? I have no problem with either approach by the way (govt subsidy or NBN cross-subsidy of non-city users) but your arguments don’t seem to hold up. How much do you think the coalition’s plan is going to cost anyway? Turnbull isn’t sure and that $10 billion I saw mentioned in the media is a joke.

          • Few consumers will pay extra to upgrade beyond 10Mbit? That’s odd because in South Korea it seems that there a far more than a “few” customers on 100Mbps.

            According to this article there are 6.8 million 100Mbps subscriptions, or about 14% if you assume one connection per resident. Now when you consider that these connections will likely be shared between 2-5 people this is quite a significant number of people, conservatively about a quarter of the population, and growing at approximately 14% a year according to that article.

            And last time I checked 100Mbit is beyond 10Mbit.

          • the crux of the issue is whether consumers are prepared to fork out more on the margin for faster broadband.

            define $X as the baseline cost of providing broadband over depreciated legacy network. what “premium” over $X can telcos achieve from marketing faster speeds? this is the crux of the issue.

            also, that $X (or starting point) will vary across different countries. there’s no way in hell that NBNco can provide 100Mbit at $30 (which is what that article seems to suggest is happening in Korea).

            please consider cost relativities.

          • Moving the goal posts now Tosh? You said specifically 10Mbps. My point was that you can’t just draw a line in the sand and decide “this is what consumers want for bandwidth”.

            It’s fine for you to talk about marginal cost, and the fact that consumers will not opt for higher speeds if the marginal cost to upgrade is too high, but don’t be foolish and draw a line at 10Mbps, or 12Mbps like Mr Turnbull has.

          • Thats because 100mbit is the same price as HFC/ADSL2 in south korea

            Hell in Japan fiber is CHEAPER then ADSL2 in a lot of areas, due to the way regulations regarding financing and costs for Telcos (ISP’s in Japan have to pay for the power costs in exchanges, FTTH offloads these costs to the consumer, making FTTH cheaper)

          • Your point deetego? Did you read my reply to Tosh? If he wants to talk about marginal cost, fine, but don’t draw a line in the sand about what consumers want for bandwidth.

          • *but don’t be foolish and draw a line at 10Mbps, or 12Mbps like Mr Turnbull has.*

            the market is drawing the line based on consumer affordability and (lack of) interest:

            1/ go listen to the iTnews/Phil Dobbie interview with MT, he talks about Telecom NZ struggling to upsell from 10 to 20Mbit

            2/ also this:

            http://newtstuff.blogspot.com/2011/05/at-stankey-u-verse-build-virtually-over.html

            “They are going from 1-3 meg DSL to 10 meg cable, but the 50 and 100 megabit cable isn’t selling because of the $100 typical price.”

          • *facepalm*

            The market drawing a line that just happens to fall around 12Mbps does not, in fact mean, that consumers don’t want faster speeds. South Korea proves this, their market there has drawn the line at a higher bandwidth tier and people are migrating to it in spades.

            You’re confusing the market dynamics with arbitrating the best product. What the market is doing is resulting in most people wanting about 12Mbps, but as prices fall, this will likely rise.

            That does not mean that people only want 12Mbps. That just means that right now, that’s all they can afford. Do you see the difference? If you want to arbitrate the best speed you shouldn’t be looking at the market but instead surveying both current and theoretical use cases of consumers, and then finding the best trade off.

            I’ve seen people talking about use cases but no one actually doing a wide scale survey of consumers to see what they want.

          • *The market drawing a line that just happens to fall around 12Mbps does not, in fact mean, that consumers don’t want faster speeds*

            yea and i want a Holden HSV too.. as opposed to my crappy Commodore.

            *If you want to arbitrate the best speed you shouldn’t be looking at the market but instead surveying both current and theoretical use cases of consumers, and then finding the best trade off.*

            false dichotomy. the market is already telling you how much it costs to provide certain services and how many consumers are willing to pay for those services.

            *I’ve seen people talking about use cases but no one actually doing a wide scale survey of consumers to see what they want.*

            you don’t get it. NO OTHER COUNTRY IN THE WORLD is spending this much money on broadband. we’re TOTALLY OFF THE CHARTS. there’s plenty of market evidence from all around the world that consumers do not need superfast broadband, e.g. telcos halting FTTH roll-outs at 5% or FTTN roll-outs at 60%, lack of interest in 100Mbit HFC, etc.

          • You have to draw a line in the sand somewhere, else you don’t get anywhere

            The whole point of drawing a line is to find the minimum benchmark required for a decent standard, and in the case for the internet that is 12mbits (which is actually highest out of any country in the world, I cant recall any country that has a minimum benchmark of 12, USA for example has 5, NZ has 5, and countries like Japan and South Korea don’t even have a minimum).

            That figure of 12 may be too high, it may be too low, and you can debate around that forever. However most sane people will agree that almost every end user application for today standards (and remember Turnbull said that he would double that minimum speed to 24 in two years) can be used effectively for that 12 mbit

            Of course the minimum has no relevance as to what speeds people will be able to access above that line, most likely in around 5 years time, people will easily have access to VDSL2 like speeds for around 70% of the country

          • yea and i want a Holden HSV too.. as opposed to my crappy Commodore.

            You don’t get it do you, currently in South Korea they are offering HSVs for the cost of the “crappy Commodore.”

            What does that tell you? That everyone in South Korea just happens to be richer? That the value of their economy just goes so much further? No, it tells you that they have already invested in the billions needed to install FTTH, and the marginal cost is coming down. The same thing will happen under the NBN.

            false dichotomy. the market is already telling you how much it costs to provide certain services and how many consumers are willing to pay for those services.

            So no, you don’t see the difference. You need to understand something Tosh: the market is not good at doing a few things, namely long term spending. This is because humans look for immediate benefits, because immediate benefits are far more tangible to them.

            For example: I can spend 3 real dollars on a coffee now, or have 12 real dollars in retirement thanks to saving those 3 real dollars over the next 30 years.

            Broadband is a long term, 10-20 year, investment. Countries like South Korea and Japan who have put in this investment early are now seeing the benefits, because now everyone has HSVs.

            you don’t get it. NO OTHER COUNTRY IN THE WORLD is spending this much money on broadband. we’re TOTALLY OFF THE CHARTS. there’s plenty of market evidence from all around the world that consumers do not need superfast broadband, e.g. telcos halting FTTH roll-outs at 5% or FTTN roll-outs at 60%, lack of interest in 100Mbit HFC, etc.

            Bullshit. First of all, there is plenty of interest in 100Mbit when the marginal cost comes down which takes time Tosh. Secondary, South Korea are investing just as much as we have in their network. They are spending a further $26 billions on expanding to FTTH from their FTTB networks. That’s on top of the tens of billions already spent. Thirdly, see my point about the market being bad a long term investments.

          • Your comparisons are laughably fallicious

            South Korea has a population that is around two and a half times the size of Australia, and that doesn’t even take into account the vast differences in population density

            Malcolm Turnbull is correct, there is not a single country in the world that is spending as much, per capita, as Australia is on broadband. We are the only country that is doing this

            Take the wool off your eyes

          • South Korea has a population that is around two and a half times the size of Australia, and that doesn’t even take into account the vast differences in population density

            So? Oh great South Korea has 2.5 times the people as us and a higher population density, obviously that is the explanation of why the marginal cost of 100Mbps services is so low. It couldn’t possibly have anything to do with the fact the network investment has already been completed.

            Malcolm Turnbull is correct, there is not a single country in the world that is spending as much, per capita, as Australia is on broadband. We are the only country that is doing this

            In PUBLIC spending, get your metrics right. The fact we’re doing it mostly all public is unique, and only a bad thing if you’re foolish enough to think that Private Enterprise somehow always gets it right.

            The United States has seen capital expenditure of over $250 billion dollars since 2008, $71 billion of which was for wireless based services. That is about $145 pp/yr on wireline services and backhaul. Please note that this was without the subsides the National Broadband Plan will be providing to the US.

            The NBN, by contrast, is projected to be $36 billion in capital expenditure, over about 10 years. That’s a projected spend of about $165 pp/yr. Comparing capital expenditure to capital expenditure, and noting that investment is likely to increase due the the National Broadband Plan in the United States, and that after the NBN completion, investment wireline in Australia is likely to stagnate, it seems that the United States and Australia are likely to spend about the same. The difference being that America’s investment is going to be primarily from private enterprise.

            Take the wool off your eyes

            You too. I am not just blindly following the NBN plan as you seem to think.

          • @NK

            Korea is a very unique, “nationalist” economy where you have highly-incestuous links btw govt and industry and domination of the economy by chaebols which are “privately-owned” yet directly implement politically-determined industrial policy.

            the $30 price of 100Mbit is artificial govt (or administrative) pricing:

            http://www.infodev.org/en/Document.934.pdf

            “However one of our interview respondents commented on the Government’s “administrative guidance” on pricing and another commented that “in terms of pricing for BB services, the government recommended a level of price that would not be a financial burden for Korean citizens, around KRW30,000 (US$30)”.”

            it doesn’t reflect underlying economics or anything like that.

            *the market is not good at doing a few things, namely long term spending.*

            so, Rio/Fortescue’s billion dollar iron ore expansion projects and Woodside/Origin Energy’s billion dollar LNG projects which take decades of planning to bring online and involve massive financial risk in the form of upfront investments in surveying, engineering, construction, etc aren’t “long term”?

            venture capitalists who back, incubate and fund highly-risky ventures that take years to bear fruit isn’t “long-term”?

            the difference btw govt and private industry is that experienced industrialists know the diff btw a smart bet and a dumb bet (93% FTTP NBN).

            *For example: I can spend 3 real dollars on a coffee now, or have 12 real dollars in retirement thanks to saving those 3 real dollars over the next 30 years.*

            but the crucial diff is there is no demand for that “full-cup of coffee” now, but rather demand for “quarter cup of coffee” which is cheaper to purchase and the balance of the cash can be saved instead. (a bunch of fibre-greedy internet geeks posting on internet forums does NOT reflect generalised demand for fibre extravagance.)

            *Secondary, South Korea are investing just as much as we have in their network.*

            go look at the chart on this page:

            http://www.analysysmason.com/About-Us/News/Insight/FTTH_roll-out_Insight_Aug2010/

            the highest projected spend is USD600 per household. we’re TOTALLY OFF THE CHARTS when you divide $50bln by 10mln households.

          • @NK

            *The United States has seen capital expenditure of over $250 billion dollars since 2008……. that is about $145 pp/yr on wireline services and backhaul.*

            go read the report the figures are derived from:

            http://www.ustelecom.org/uploadedFiles/Learn/ReportsAndStudies/052711-Research-Brief.pdf

            those are gross capex numbers which include “regular maintenance or replacement capex” on existing legacy networks which can be funded from ongoing revenues. you must differentiate btw this and “upgrade capex” which enlarges the capital base and requires HIGHER subscriber revenues to justify the upgrades.

            look at the chart on p4. compare Australia’s $211 vs US’ $249 capex per capita over 1997-2007. note how similar it is. all those figures you quoted are gross capex numbers, and not just NGN upgrade capex.

            *The NBN, by contrast, is projected to be $36 billion in capital expenditure, over about 10 years. That’s a projected spend of about $165 pp/yr.*

            go read this:

            http://www.analysysmason.com/About-Us/News/Insight/FTTH_roll-out_Insight_Aug2010/

            for North America, projected cumulative NGN capex per household of US$300 over 2009-15 works out to US$43 per annum (which is less than a third of your $165pp/yr number).

          • Hmm. And what are we doing over the next 10 yeats Tosh? Oh that’s right : replacing one network with another. Which means the ongoing maintenance costs of the existing network will steadily decline as the rollout of the NBN continues.

            Then as stated the NBN will cause investment to stagnate, bringing the average spend down further.

            It’s more expensive, I get that, I’ve always gotten that. But not as much as you seem to think.

          • the $30 price of 100Mbit is artificial govt (or administrative) pricing

            Beside the point. South Korea have still managed to justify it economically, and despite the obvious huge cost overheads that will be irrecoverable continue to invest in the infrastructure.

            Doesn’t sing well with your “FTTH is too expensive for anyone to build because there is no need for it” assertion now does it?

            venture capitalists who back, incubate and fund highly-risky ventures that take years to bear fruit isn’t “long-term”?

            You seem to be confudsing the idea that an entity is bad at something with the ability to be able to do it at all.

            I am bad at dancing, I can still dance, and do not avoid dancing, but it is hard for me to do.

            the highest projected spend is USD600 per household. we’re TOTALLY OFF THE CHARTS when you divide $50bln by 10mln households.

            So yeah. That $250 billion spent on Broadband in the US over the past four years. You know, where I calculated spending of about $145 pp/yr that is set to increase and then compared it to the projected cost of the NBN ($165 pp/yr)? That’s totally not relevant at all is it. Not at all.

            Really Tosh, really? You know that this debate is going to go on for ages. Why don’t you give up while you’re ahead. I’m just undermining a few of your assumptions here. I haven’t proved we need the NBN anymore than you’ve proved we don’t.

          • @NK

            *Beside the point. South Korea have still managed to justify it economically, and despite the obvious huge cost overheads that will be irrecoverable continue to invest in the infrastructure.*

            it’s not beside the point. you were trying to argue that “$30 for 100Mbit” is a market outcome due to Korea building FTTH years ago and capital has already been repaid. this is clearly not the case. the $30 price point is a government-determined administrative measure.

            you’re obviously not familiar with the Korean economy. South Korea is not your typical industrialised nation. command of capital and investment is highly-concentrated within several “chaebols” or conglomerates whose investment activities are closely-coordinated with government policy. Korea is a very nationalist, top-down administered country.

            they have paid a very high cost to build up their huge chaebols (high import tariffs, etc) and some have even gone bust in the past. they certainly don’t get everything right. and so far, they have very little to show for their FTTB investments.

            http://www.ustelecom.org/uploadedFiles/Learn/ReportsAndStudies/052711-Research-Brief.pdf

            pg. 5: “South Korea is a bit of an outlier due to extraordinary amounts of residential online video downloading. The massive usage gap between South Korea and the rest of the world virtually disappears when compared based on, for example, business traffic.”

            most of the billions poured into fibre networks is not being used productively at all. it’s just home entertainment, and hasn’t resulted in any meaningful gains for business.

            also, look at the chart on pg. 7:

            pushing fibre everywhere and subsidising 100Mbit hasn’t led to an explosion of ICT capital investment. in fact, the proportion of private capital investment devoted to ICT is just over a third of US and significantly below even Australia !!

            indiscriminately pushing fibre to every premise doesn’t achieve anything. it’s like giving a fishing rod to every household and expecting the nation to become the biggest fisheries exporter.

            also, look at the chart on pg 6. despite pervasive fibre, avg monthly IP traffic in Japan is only a measly 11GB. more money wasted. no wonder the Japanese Government is in so much debt. big, expensive projects that don’t achieve anything just like Labor’s NBN pushing fibre to Whoop Whoop.

          • Discus sucks at nested comments, so this comment is addressed to Tosh.

            I’m from the US, and we spent $200 billion in 1996 to pay telecoms to lay out fiber all over the US. They basically broke the rules and pocketed the money, but nonetheless the money was spent in the form of grants and tax breaks.

            So you’re quite wrong about “no one” investing as much money as Australia. Most Asian countries are also investing in fiber, since they have no lines yet and copper is a dead end.

  44. Speed is not the only positive that I see in the NBN, however when discussing fibre vs the coalition plans it would be good to consider the upload speeds more, and to (start to?) take into account that many households do have a growing need to run multiple applications at the same time. Bandwidth requirements are really only going to be going in one direction in future.

    My own recent example shows how bad things are currently. Just trying to connect with a relative, with an ADSL2+ connection on my end (5mbps down/1mbps up) and an ADSL1 connection on their end (8mbps down/384kbps up), to use a webcam. Had to opt for VGA only (min bandwidth required was 256kbps) due to their low upload bandwidth. I’ve also used remote desktop over this connection, which is very laggy even on low settings. It would be great to be able to have an audio channel going at the same time via webcam, but that just slows the remote desktop down even more (someone speaks and you see the lag grow in the refresh of their desktop). I was thinking of using VoIP on my end to phone them over PSTN, but that adds additional costs/services to my end which shouldn’t really be needed.

    A HD webcam requires min upload bandwidth of 1mbps, which essentially makes it a single use application on my ADSL2+ connection, by tying up all my upload bandwidth in one go. No wonder collaboration over video is not happening as you would be limited to whatever the lowest upload bandwidth is of every participant involved. Even with VGA resolution, 384kpbs is just enough to send video to one other person.

    On a 25Mbps/5Mbps fibre connection, a lot of these constraints start to disappear, and on a 100Mbps/40Mbps connection I can see even large households being able to run multiple applications per user without impacting on each other. It is not until this connectivity is the norm that you will start to see more people taking advantage of the technology, both at home and in the workplace.

    I *really do not want* to be constrained to using Bigpond HFC with (at most) 2Mbps upload speeds (going on Melbourne’s Ultimate cable service) and being told by the Coalition, hey, you have 100Mbps download speeds, that’s more than we promised. If you think that is adequate for the future, Renai, then you’re not such the techo extreme that you think you are.

  45. Very nice article!

    Only time I saturate my adsl2+ which syncs at 18.6mbps is when I am downloading a tone of torrents.

    I do not believe that the whole country should pay for my torrenting though =)

    At work I configured a HD video conferencing link between to Tandberg HD video conf boxes from Sydney City to out West. We used our own atm network and built it as a 5mbps pvc knowing we only needed 2mbps.

    Who needs this NBN? I can not think of many or any besides torrent users, who by the way will be pin holed when they try to leave Australia.

    • So you assume every other Australian uses the internet for the same purposes as you, which would have you imply that the only time anyone would saturate an ADSL2 connection is when downloading torrents among other luxury activities?

      Video conferencing is a bad use case for the NBN, so you’re partially correct in that bandwidth requirements for video conferencing are minimal, though the NBN will still allow for much higher quality video streams which are more important to some sectors than others.

      Joes, can I ask how important your digital data is to you? Do you currently back it up? if so, how? If a fire burnt your house down, a flood came through, or you had a large electrical fault that fried all your computer equipment, would your data be safe? Online off-site backup is clearly what I’m getting at, and unless you’re only backing up tiny files, online backup is too impractical on anything other than a fibre connection. That’s just one use case for the NBN which will demand ever more bandwidth as time goes on.

      Personally, I’d love to be able to access all of my data at home while I’m at work, and vice versa, but at the moment, I’m limited to accessing only text documents and other small files. What if I want to install a copy of Windows 7 at home from a 4GB disc image at work? It would probably take a day to copy over current ADSL2 connections. There’s another basic use case that’s going to become increasingly common.

      In 10 years time, no one’s going to be saying “Hmm, it seems we didn’t need 100mbit after all”. In fact once you get use to and get depend more on the use cases that make use of the speed, you’ll wonder how you ever got by on DSL.

      • “In 10 years time, no one’s going to be saying “Hmm, it seems we didn’t need 100mbit after all”. In fact once you get use to and get depend more on the use cases that make use of the speed, you’ll wonder how you ever got by on DSL.”

        Exactly… No one will ever say that it was “enough” back then. They would only ever think of how fast they could have done things back then. I still recall the days it would take leaving the dial-up modem on overnight to download a 30mb file from a website, and hoping to god the connection didn’t cut off. That was around 10 years ago… I don’t know how I didn’t lose my mind there and then!

      • “In 10 years time, no one’s going to be saying “Hmm, it seems we didn’t need 100mbit after all”. In fact once you get use to and get depend more on the use cases that make use of the speed, you’ll wonder how you ever got by on DSL.”

        Exactly… No one will ever say that it was “enough” back then. They would only ever think of how fast they could have done things back then. I still recall the days it would take leaving the dial-up modem on overnight to download a 30mb file from a website, and hoping to god the connection didn’t cut off. That was around 10 years ago… I don’t know how I didn’t lose my mind there and then!

      • “In 10 years time, no one’s going to be saying “Hmm, it seems we didn’t need 100mbit after all”. In fact once you get use to and get depend more on the use cases that make use of the speed, you’ll wonder how you ever got by on DSL.”

        Exactly… No one will ever say that it was “enough” back then. They would only ever think of how fast they could have done things back then. I still recall the days it would take leaving the dial-up modem on overnight to download a 30mb file from a website, and hoping to god the connection didn’t cut off. That was around 10 years ago… I don’t know how I didn’t lose my mind there and then!

      • Great post Tom, totally agree, the anti-NBN crusaders will take that speed and technology for granted just as they do now, problem is they assume because they can do everything they want with the speed they get now then that is all they need and all everyone else will ever need.

  46. 50%~ of voters have spoken and want the NBN. Stop talking about it, stop standing in the way of progress and get the hell out of my way. You’re all embarrassing yourselves, considering we are an embarrassment as is, with out pitiful connections that are ‘enough’ according to Joe Bloggs and some idiots who only use the internet to ‘talk’ on Facebook.

    We are ranked somewhere around the 70-odd mark in comparison to our download speeds with countries in Africa (Yes, Africa. Some countries there can’t afford to feed their own people) beating us over the head with downstream speeds, and let’s not even bring up upstream.
    Upstream is EQUALLY important as Downstream people. You can only download as fast as the file/data can be uploaded. Logic 101.

    Now this is the first govt. project, I as a 22-year old IT Professional can actually see some major benefit not just to my industry, but to everyone’s industries including families, pensioners and even those who seek to take money from Centrelink. Jobs will be created by the NBN, industries will be advanced and Australia will no longer be an IT-backwater country. I voted for it, and by god I want it to happen.

    This article is an embarrassment. Whoever says “This is enough”, in regards to performance and space, obviously doesn’t know human nature, enough is never enough… and why should it be? If we can have something better, more efficient and cost-effective in the long run, why the hell should we compromise? Malcolm Turnbull is out of touch with the industry and should be taken out back and beaten with 56k modems, he’s quite obviously stuck in the past and I am quite frankly happy to leave him there.

    We need more bandwidth, we need more speed. There are many things in the future that will demand this.

    Cloud Computing and Cloud Storage/Backups are a prime example. No one wants to send 50GB Backup every month at an upload rate of 256kbps. By the time you finish uploading it, you’ve already begun uploading the third month’s backup. Start thinking what we are limited in now, and you will see why the NBN is a good idea and a must. Sure, we may not have a Miner operating a mine 500kms away, just think about how you would use that bandwidth, how you wouldn’t have to buffer every 5 seconds to watch a video your niece sent you, how you wouldn’t have shit quality Skype conversations with your Grandmother overseas. How your life actually becomes better knowing everything is near-instantaneous with little to no waiting to get the content and quality you need and deserve as an Australian.

    Think for yourselves for one second and stop buying into the Opposition’s bulls###. They are there only to oppose. They always do it, even Labor does it. If it’s not there idea, it’s a shit idea and frankly, I am sick of it because lots of “good” ideas that require a non-political spin to get up and running, always end up being flown into the ground just because Politicians are running it or opposing it.

    Sincerely,
    An IT Professional

    • The upstream bandwidth really is the killer feature of the NBN. It’s going to enable many applications which are not possible or practical today. Being able to send files and video streams directly to another computer without a server in-between is going to be truly liberating, and so much more convenient than the way we currently share data. It’ll be like having a local network over the internet. I could make my NAS accessible over the net, and have my dad and brother access all my files from their own lounge rooms. This is something that would be possible right now – I could set this up right now, but it’s going to be way too slow to be of any use. There are many use cases for the NBN that don’t depend on new software or service developments being able to ‘leverage’ it.

        • Sorry, Why does your opinion count any more than the person you are replying to…. Ohh that’s right it doesn’t.

          They have just as much right or just as much say on how tax dollars should be spent as you do.

          Difference…. Their comments don’t make me think… Tosser while reading them.

          • and you know what i think when i read your comment? another ignorant, uneducated loser who’s only aspiration in life is to stream video at 100Mbit.

    • *Stop talking about it, stop standing in the way of progress and get the hell out of my way.*

      get the hell out of your way? you want GPON fibre at home? go talk to one of the many telco providers currently offering fibre and pay for it yourself. no-one’s stopping you. just don’t expect to bludge off the taxpayer.

      *some idiots who only use the internet to ‘talk’ on Facebook.*

      you have a problem that not everyone is a tech geek? some people are actually very sporty and have active outdoors lifestyles and don’t spend much time sitting behind a PC monitor and have ZERO need for fibre.

      *Upstream is EQUALLY important as Downstream people. You can only download as fast as the file/data can be uploaded. Logic 101.*

      really? last time i checked most people “download” from commercial servers. the most active web servers around the world already have superfast “upload” speeds. if you are one of the small minority (5% of users) who torrent heavily and account for over 40% of network traffic, i suggest you get yourself a fibre connection and pay for your luxuries YOURSELF.

      *Jobs will be created by the NBN, industries will be advanced and Australia will no longer be an IT-backwater country*

      if you actually bothered to do some basic research:

      http://www.ustelecom.org/uploadedFiles/Learn/ReportsAndStudies/052711-Research-Brief.pdf

      pg 7 – note how Korea and Japan which have massive fibre roll-outs sure hasn’t stimulated investment in the ICT sector. you’re looking at 13% (among the lowest) vs 20% for Australia.

      *This article is an embarrassment.*

      an embarrassment to Labor and NBNco. agreed. stripped of all that rubbish Alcatel fibre marketing spin, the truth embarrasses.

      *Cloud Computing and Cloud Storage/Backups are a prime example.*

      prime example of niche applications that the average Australian couldn’t give a toss about, and definitely not worth spending $50bln of taxpayers’ money on.

      *No one wants to send 50GB Backup every month at an upload rate of 256kbps.*

      how about a 70Mb/s USB 3.0 connection to a cheap 500GB WD external HDD then?

      • “really? last time i checked most people “download” from commercial servers. the most active web servers around the world already have superfast “upload” speeds. if you are one of the small minority (5% of users) who torrent heavily and account for over 40% of network traffic, i suggest you get yourself a fibre connection and pay for your luxuries YOURSELF.”

        Last time i checked… 70% of all internet traffic was p2p.

        Get off you high horse. Stop judging people based on what “they” want the NBN for…

        Guess what…. If someone says they want the NBN because it will allow better entertainment. They are perfectly legitimate in saying so…. It is just as much THEIR TAX dollars as it is anyone else.

        “prime example of niche applications that the average Australian couldn’t give a toss about, and definitely not worth spending $50bln of taxpayers’ money on.”

        Again off that high horse buddy. Not sure what industry you work in. But there is HUGE demand already for cloud services within Australia which i constantly have to advise against due to our crap communications infrastructure.

        • *Last time i checked… 70% of all internet traffic was p2p.*

          and that 70% of P2P traffic is generated by what proportion of users?

          *Again off that high horse buddy.*

          move to a marginal electorate, son.

          *But there is HUGE demand already for cloud services within Australia which i constantly have to advise against due to our crap communications infrastructure.*

          there’s plenty of fibre around. what you mean is your clients can only afford crap.

  47. Unless I am wrong 50% of voters have not spoken for the NBN. I believe the that the Libs won on primary votes. Labor won over all by green preferences?

    I do think for myself too. I am a network design engineer (in response to your IT professional signature)

    Call an election today and this terrible government is gone

    • I did put “around” (~) because Labor didn’t win by primary vote, I watched the same election you did mate. I do know some of my friends who wanted the NBN, voted Liberal because they were discontent with them on a few things.

      I also know some who voted Labor (when they would ONLY vote Liberal before) solely for the NBN. It was that much of a big deal for them, It was safe to say allegiances changed solely on this policy alone.

      I would honestly say 60% of the country, maybe even more. Would welcome the NBN with open arms. It just seems to me, the rest are really vocal hardcore liberals (with Internet connections) and if not that, are up there with “It’s enough for me, so it’s enough for the entire country” agenda. Which is completely unfair.

      Also it’s funny is it not? They (Liberals) talk about how they don’t want to waste money and how Labor blows money away like it was growing on trees, but do you (Joes) have any idea how much it costs to run an election? You do know that the parties get tax-payer money for each vote they attain right? Also where do you think the money comes from to pay for the people counting the votes? the people who marshal the schools? etc.

      It honestly boggles the mind each time I hear “This government is wasteful! Call an election”. It’s right up there with “We hate change! and you ain’t done a damn thing!”

      Get over it, the government is in power because of the Independents who signed on with Labor for various reasons, most of them because of the NBN. Wait til next election to have your say again, quite obviously you weren’t the deciding vote now, were you?

      • Well next election, Labor will most likely lose, almost every political commentator is stating that its nearing impossible for Labor to win the next election unless some miracle happens

        If you want to get into a political debate, then sure, but Labor is not very high on those grounds…

    • “Call an election today and this terrible government is gone”

      And that would have nothing to do with the NBN. Stop conflating poor popularity of the existing government with high popularity for the opposition. Both sides stink, that’s why we have a hung parliament. The sooner they both realise this the better.

  48. Unless I am wrong 50% of voters have not spoken for the NBN. I believe the that the Libs won on primary votes. Labor won over all by green preferences?

    I do think for myself too. I am a network design engineer (in response to your IT professional signature)

    Call an election today and this terrible government is gone

  49. Unless I am wrong 50% of voters have not spoken for the NBN. I believe the that the Libs won on primary votes. Labor won over all by green preferences?

    I do think for myself too. I am a network design engineer (in response to your IT professional signature)

    Call an election today and this terrible government is gone

    • Apparently the “foreseeable future” only goes up to 2012… oh wait no it doesn’t according to NBNco (if you bothered reading the graph below) the “foreseeable future” goes up to 2028 (the Mayans are wrong!) and according to them 36% will be on 12/1mbps and the other 64% (most) will be on speeds 25mbps or higher. Not only that but of that group the 25 & 50mbps plans are the two smallest with the bulk opting for 100 or 250mbps, even 500mbps outweighs both of them, as I’ve said before wouldn’t surprise me if NBNco eventually phased out these two speeds.

      52% of residential End-Users expected to be on the 12/1Mbps service in FY2012.

      Note the FY2012, were you expecting every residence in the 93% fibre footprint to be wired up by that time?

      • ~40% connecting at 12/1Mbps in 2028 is still a huge number and the decline has virtually stopped. These people cannot justify an extra $5/month to double their speed. I doubt anyone posting on this website will be connecting at 12/1Mbps, whereas clearly NBNCo are predicting that 12/1Mbps will continue to be the most popular plan.

        If the NBNCo’s predictions are correct then the NBN looks likely to increase social inequality rather than decrease it.

        • ~40% connecting at 12/1Mbps in 2028 is still a huge number and the decline has virtually stopped.

          I can say something like that too: ~60% connecting at 25mbps or greater in 2028 is a huge number and the uptake of higher than 100mbps plans continues. Sorry but facts are facts, most people won’t be on 12mbps according to NBNco. You’ve obviously tried to represent the numbers in a way that favours your argument against the NBN. I’ve read the very page and the graph in your link and it simply contradicts what you said in your initial comment.

          whereas clearly NBNCo are predicting that 12/1Mbps will continue to be the most popular plan.

          There are seven plans according to the graph so you can basically say the 12mbps plan is always the most popular right down to something like 14% even though the rest (86%) totally outweigh it and that really doesn’t mean much does it.

          If the NBNCo’s predictions are correct then the NBN looks likely to increase social inequality rather than decrease it.

          How will it? According to their predictions most people WILL NOT be on 12mbps.

  50. e-health, e-health.. All I hear is e-health with benefit. I don’t want to bay 50 billion dollars so a doctor can look at me via a computer. Sounds like we are trying to justify why the NBN is good, rather than have justification FOR the NBN.

    • If that’s the only justification you’ve heard for the NBN you haven’t been paying attention.

    • 10-30Mbps per user with 150Mbps to 300Mbps per cell with current deployments, compared to up to 1.4Gbps per user with 3.75Gbps per OLT. Wasted investment? Quite.

      Why don’t you look into the technologies you are talking about before sprouting non-sense.

    • And that joo, is why people like you are thankfully not in a position of significance on this matter. Such a strong opinion, yet so ill-informed, much like the vast majority of Australian’s who believe they’re experts on everything such as climate change, mining taxes, refugees, the NBN, etc. NBN is my thing and I have very technically sound reasons to support it, I don’t however dare comment on things that I haven’t researched extensively like climate change, the mining tax and refugees. Though I may have sway a particular way on those topics, I understand my knowledge of those topics is very limited and thus I’m completely open to rationale reasoning either in support or opposition of any of those. I don’t quite understand though how someone can comment with such confidence on topics which they clearly have very little knowledge of.

      I do agree however that video conferencing and e-health are probably two of the poorest example benefits of the NBN. It’s unfortunate the government’s pushing these examples so hard, as they rarely are not helping. I don’t agree however that anything other than fibre is the future of Australia’s broadband network.

  51. I believe that there is potential for growth in bandwidth usage in education at home.

    Probably the biggest area for potential bandwidth usage is in using video for learning support. As a teacher, I can explain/demonstrate/teach a concept or method to my students. But not all students will get it the first time. Traditionally, students who have poorer short term memories will need to have help from the teacher once they are expected to do their work after instruction. This often involves repeating the what has just been said so the student can remember it. Sometimes it is explaining it in a slightly different way. Sometimes, writing things down helps, but is useless for students who have reading difficulties. Video can help support in this area. Students can watch video of a concept being explained/demonstrated/taught as many times as they want, and can come back and refer to it later. While pulling video off the web at a school that is fibered up would be fine, doing it for homework while others are using the net is going to demand bandwidth at home.

    An Australian based online maths program called Mathletics is gaining popularity fast, and is becoming a hit internationally too. Probably, part of the success is the fact that it uses low bandwidth flash animation, and so works on just about everyone’s connection (though I doubt it would be great on dial up). My class uses this a little at school, and they can also use it at home if they have a computer/internet connection. However, I can see potential for using much more video and audio, and higher quality video and audio, to increase learning. Mathletics currently has learning support in flash animation, but requires reading skills – more difficult for students with reading difficulties (ie. 5-10% of all students that have come into my class in my experience). Video/audio would be much better for these students. So, this is an example of how current bandwidth usage can, and would benefit from, being increased for an application.

    Mathletics is only for one subject. And there are so many other subjects with so much potential for web applications. And not only with video, but other web content too. Increasing the number of subjects with applications on the web would put more demand on bandwidth, noticeable if accessed from a limited home connection.

    Lastly, children trying to do difficult work at home will not persist if they have to wait. Think three kids trying to do homework and accessing video at the same time for support while the rest of the family entertains themselves online. Might work for a while on constrained bandwidth, but I think it would become unworkable if any QoS had to kick in.

    So while I have not mentioned any current applications for the home that would put pressure on bandwidth, I can envisage that in the future, these will slowly appear.

  52. I believe that there is potential for growth in bandwidth usage in education at home.

    Probably the biggest area for potential bandwidth usage is in using video for learning support. As a teacher, I can explain/demonstrate/teach a concept or method to my students. But not all students will get it the first time. Traditionally, students who have poorer short term memories will need to have help from the teacher once they are expected to do their work after instruction. This often involves repeating the what has just been said so the student can remember it. Sometimes it is explaining it in a slightly different way. Sometimes, writing things down helps, but is useless for students who have reading difficulties. Video can help support in this area. Students can watch video of a concept being explained/demonstrated/taught as many times as they want, and can come back and refer to it later. While pulling video off the web at a school that is fibered up would be fine, doing it for homework while others are using the net is going to demand bandwidth at home.

    An Australian based online maths program called Mathletics is gaining popularity fast, and is becoming a hit internationally too. Probably, part of the success is the fact that it uses low bandwidth flash animation, and so works on just about everyone’s connection (though I doubt it would be great on dial up). My class uses this a little at school, and they can also use it at home if they have a computer/internet connection. However, I can see potential for using much more video and audio, and higher quality video and audio, to increase learning. Mathletics currently has learning support in flash animation, but requires reading skills – more difficult for students with reading difficulties (ie. 5-10% of all students that have come into my class in my experience). Video/audio would be much better for these students. So, this is an example of how current bandwidth usage can, and would benefit from, being increased for an application.

    Mathletics is only for one subject. And there are so many other subjects with so much potential for web applications. And not only with video, but other web content too. Increasing the number of subjects with applications on the web would put more demand on bandwidth, noticeable if accessed from a limited home connection.

    Lastly, children trying to do difficult work at home will not persist if they have to wait. Think three kids trying to do homework and accessing video at the same time for support while the rest of the family entertains themselves online. Might work for a while on constrained bandwidth, but I think it would become unworkable if any QoS had to kick in.

    So while I have not mentioned any current applications for the home that would put pressure on bandwidth, I can envisage that in the future, these will slowly appear.

  53. BTW, my school are one of the ‘lucky’ people to already have a form of FTTN – let me explain. There is a RIM about 300m up the road that is connected (I assume) by fibre back to the exchange several km away. Why on earth our ADSL line is synced at only 5.5 Mbps and not 8 Mbps is beyond me. Perhaps it has something to do with the fact that when I walk out the front gate, I see a (permanent) temporary orange Telstra barrier fencing off an open pit with Telstra copper coming up and into two little black boxes and then disappearing back into the pit. While I realise that these visible lines are probably used only for voice, and our ADSL line probably stays in the pit, it still gives an indication of the state of the current copper network in the area. I also regularly see Telstra vans that have driven out from the exchange and parked next to the RIM doing “stuff” inside the cabinet (I won’t pretend to know what they’re doing).

    I’ll let you draw your own conclusions and parallels to the proposed (partly) FTTN upgrade from my partly tongue-in-cheek comment.

  54. Yes we are still discussing the NBN, S.Korea already has it. My friends there are getting 1Gbps connections at home. I’m lucky to get 1.5Mbps. It is a struggle to do any video on that. My father in a UK village gets 8Mbps as the minimum. BTW I live in Metro Sydney!!

    Video will drive the bandwidth, but more importantly so will latency, wireless can’t do this. Gaming networks will require high bandwidth, low latency connections just so people can compete. 12Mbps wireless connections will not deliver that.

  55. Fibre is the only connection that can give me speeds greater than 5Mbps at the moment. Wireless is rubbish and so is ADSL2+. So while you say that 100Mbps isn’t necessary now. Who is to say in 10 years when the roll-out is complete? It is the only thing that can give me the speeds I need now (25Mbps would do me fine).

    • He is basically contradicting his earlier viewpoints. Either he’s been bribed or he’s just gone off the deep end.

  56. only a dinosaur with its head firmly cemented in a iron box could see no value for something far superior. enough of the dribble already it gets tiring.

  57. I’d start by saying I work for an ISP, so have some idea on these numbers. VC may start to take off but 1080p VC is mostly used @ around 2Mbps. Often goes up to 8Mbps but you don’t get any quality improvements just refresh rate. As for multi-party conferencing a simple search will show you that every man and his dog is releasing cloud based conferencing/media mixing, including google+. This means you only need the one stream to the endpoint and a lot less BW then stated by many.

    A question never asked is will our money well spent by NBN Co – on GPON? It is not p2p fibre and is shared between users. Active ethernet and p2p fibre might be the network everybody thinks NBN is/should be but never got a look in. I must also mention, turnbull has as well, that copper alternatives exist that can do 100/100 non contended compared to GPON 100/40 contended. It can also do this at half the installation price (no trenching of fibre into the house from the curb) and via a power method much greener then GPON.

    So is there a need for faster broadband? Maybe. Does it HAVE to be FTTH at all cost and is GPON the most appropriate technology. Only a brave, or ill informed person can surely make that case? Politicians seem to have made both a technical and commercial decision on this. Seems backwards to me. So is closing down Optus/Telstra DOCSIS 3.0 networks that pass 3 million homes and can already provide 100Mbps speeds (albeit currently more contended then GPON).

    Seems to me none of this is mentioned though as pollies turned it into a Fibre vs copper debate and completely glossed over the detail or just made up figures, as this article points out.

    Others have mention this but my last point is that all the banter about about e-health ignores the fact that remote locations will likely be delivered by wireless or (probably) satellite NOT FTTH. It would be nice if everyone saw some real numbers/stats instead of people in power deliberately misleading the country. I must say I am looking fwd to having affordable high speed p2p fibre for business, pity this won’t be available to every business.

    • I’d start by saying I work for an ISP, so have some idea on these numbers.

      I don’t, but I consume media and know about encoding and network techniques, as well as a bit about network engineering, so so do I.

      VC may start to take off but 1080p VC is mostly used @ around 2Mbps. Often goes up to 8Mbps but you don’t get any quality improvements just refresh rate. As for multi-party conferencing a simple search will show you that every man and his dog is releasing cloud based conferencing/media mixing, including google+. This means you only need the one stream to the endpoint and a lot less BW then stated by many.

      Okay, so apparently you know the figures and then you quote 2Mbps as the requirement for a 1080p video conference? Jeez. I’d hate to be receiving that call. What are we sending, 5 frames a second?

      A question never asked is will our money well spent by NBN Co – on GPON? It is not p2p fibre and is shared between users. Active ethernet and p2p fibre might be the network everybody thinks NBN is/should be but never got a look in. I must also mention, turnbull has as well, that copper alternatives exist that can do 100/100 non contended compared to GPON 100/40 contended. It can also do this at half the installation price (no trenching of fibre into the house from the curb) and via a power method much greener then GPON.

      Had you known all that much about GPON you’d know that it is ever so slightly more power efficent than FTTN, in that it doesn’t need cooling due to it mostly being passive, but not by much mind, as there has to be extra equipment run at the end user premises that is more expensive. So the idea that VDSL2 will be “greener” is laughable.

      The fact that P2P fibre is slightly better and more flexible however is a good point, but considering this is to home users, the shared nature of GPON, and the associated reduced complexity of the rollout, mean that it is a slightly sounder choice. Granted upgrading to the direct fibre may not cost all that extra.

      You would further know that those copper alternatives that exist are heavily dependent on what the distance is you have to the node. Capable of does not mean you will get, if the node density is too low you may only get 12Mbps.

      Also you mention DOCSIS 3. Well, you do know that if you were to offer plans up to 1Gbps as you can on the NBN, that it would be contended about 125:1, and that’s assuming you do a 1 to 4 node split for Telstra and Optus networks. This is compared to about 13:1 for GPON. That assumes that you use almost all of the bandwidth you can on HFC for the purposes of Broadband.

      Seems to me none of this is mentioned though as pollies turned it into a Fibre vs copper debate and completely glossed over the detail or just made up figures, as this article points out

      Although I acknowledge there are (cheaper) alternatives that can in some ways (powered nodes) be more flexible. However, I think you need to realise that all of the technologies you referred to have been covered in great depth by those commenting on the Broadband Issue.

      It’s important that you get your facts straight. I’m curious, where is it in this Telco you work, because I shouldn’t have to repeat any of this too you if you’re an engineer?

  58. Renai, the premise of your article overlooks the intrinsic superiority of ubiquitous service that is CAPABLE of faster speeds, the very same 3-pea trick perpetrated this week by the shifting sands of Mr Turnbull’s “alternatives”. NBNCo’s own revenue forecasts will be met even if in 2020 there are 50% of users still only running 12/2 Mbps services. The other 50% will use a range of speeds from a competitive market according to their needs or the depth of their pockets.

    Delivering an architecture at taxpayer expense that is incapable of being upgraded from copper speeds is madness. We have already seen the folly of leaving it to the market, which cherry-picks well-off niches and ignores the less-profitable third of the population.

    Consider a typical residential street with thirty or forty houses. A third probably have school-age children. All of them probably watch television. Five or more are probably running a business. They will choose the speed they need on fibre, but be condemned to copper limits by Mr Turnbull.

    You know that fully 15 of every 16 premises within a 4km radius of an exchange are further than 1km from that exchange and cannot ever get the speed you enjoy from copper, which is also susceptible to flood, electro-magnetic interference, and would need fan-cooled, power-hungry and flood-susceptible switches in large cabinets in every residential block to deliver anything faster.

    Of course he knows better. He is playing politics to slow the NBN and poison opinion against the government (which it has pretty much done to itself anyway). In years to come he will laugh at his waster of parliamentary weeks and months in frivolous debate and say it was all politics.

    But today we are delivering national infrastructure that will stand the test of time and technological progress for at least fifty years and probably longer. It is entirely self-funding and should remain publicly owned once built. It will cut everyone’s phone bills to nothing (NBN-to-NBN calls are just a trickle of data with no connection costs), make all world television channels available to anyone who wants them, give home businesses a window to the world market, make offsite backups of home computers more reliable and the backup restoration swift.

    Only ubiquitous infrastructure that is CAPABLE of well-above-average speeds will allow households and business the choice of an entry-level or a high-end service. Copper worked fine in the 1950s, but optical fibre is what we should be laying now. And you and Malcolm Turnbull both know it.

    The coalition should adopt Labor’s NBN as its own and deprive it of its only popular policy differentiator.

  59. Can I just say this argument doesn’t even take into consideration NBN CVC charges and contention.

    I believe NBN recommends a 1:250 contention ratio to achieve price outcomes similar to those of today’s DSL? So that will give a 100Mbps subscriber 0.4Mbps guaranteed throughput? Of course this is no different to contention on the existing residential networks.

    Out of all the examples listed in the article only the streaming video the daughter is doing could be using multicast. This means that if everyone in your service area was doing tasks remotely similar to this house at the same time over TC4 the whole thing would come crashing to a halt.

    Our existing model works because people spread their downloads over non peak periods via bit torrent and do things later. Data on demand requires a lot more beef that NBN is designing and is already available today, it’s called a business grade uncontended connection.

  60. I’ve got Telstra cable, it’s supposed to be “up to” 30MBit/s, in actual fact it’s 10.5 average and can burst up to 27, for about 0.25 seconds every few minutes. The installer said it’s often as low as 7 average speed. Aiming for the barely adequate would be a monumental a mistake.

    That 10.5MBit/s is not sufficient to watch two HD streams, and no Netflix, Hulu, Pandora, ABC et al won’t buffer the full stream for copyright limitations.

    BTW on ADSL2+ I was getting 2.5Mbit/s, and a lot of dropouts due to the poor state of the copper where the SNR would vary constantly from 0-12dB.

  61. Seriously Abbott and Turnbull should be both put into jail for halting our countries progress and wasting our countries resources for their own selfish political gain. stop trying to send us back into the dark ages. in years to come these clowns will look like complete baboons. that is a fact.

  62. I personally designed an implemented a remote view for ore stockyards for Rio Tinto comprised of a full realtime 3D machine posture and various other pieces of data fed from the anticollision system. This was designed to bring data from a remote site, and display it on a desktop. This system is in use today – and consumes bandwidth in the “bugger all” region of modem speeds.

    Line of business applications do not require high resolution video, and 99% of them can be delivered over remote desktop technologies. While I would like fibre to my home, there is just about no business case for it, short of entertainment – which has dubious economic benefit.

  63. I am no expert but do work in the technology sector. I forsee the NBN predominantly being used for real time or on-demand streaming of private/commerical video and free-to-air TV(2D and3D), Desktop/Server Virtualisation, online backups and even online data storage access, computer gaming (please stop being in denial but none of you can stop the influence of computer gaming….plus I won’t even mention porn).
    We can have some of the work force work from home. A company may have the majority of it’s employees based in different geographical locations and interact via different streaming technologies and virtualisation. VoIP will be the predominant use of voice communication nationally.
    I’d like to be able to pull up my digital copy of the daily newspaper with all the full bells and whistles while being able to move from page to page easily with little to no latency.
    Wireless will always have it’s use to provide connectivity to extend the reach of the NBN. That is the way I see it. Not as a dominant last mile access technology. Definately a requirement for mobile professionals, mobile estates/homes, land/water transport etc… Maybe as a backup network in the event of a major outage due to natural disasters or damage by a third party.
    However I don’t feel convinced that ISP’s nor the government will be able to provide a reasonable QoS without obviously using QoS policies to priorities traffic simply because their servers would not be able to handle the load. But then again all the above will progressively manifest over 10 years at least.
    I also believe the copper network should remain and somehow be maintained.

Comments are closed.