NBN competition will rest almost solely on price


opinion Retail competition on the National Broadband Network will rest almost solely on price, in my opinion, as the importance of other differentiating factors between telcos like Telstra, Optus, TPG and iiNet will instantly diminish almost to zero as the network is deployed. And here’s why.

Over the past few weeks I’ve been doing a great deal of thinking about what retail competition will actually look like in Australia’s telecommunications landscape once the National Broadband Network has been rolled out across Australia, and the overwhelming majority of us have direct access to fibre connections to our premises.

I know what you’re thinking right now. It’s probably something like: “Get a job, you bum!” Well … no, I already quit my job and now I work for myself in an idyllic self-employed, no pants, pyjama lifestyle where I have enough time to sit calmly in my loungeroom, sipping tea and enjoying fresh yoghurt while contemplating the possible future dynamics of Australia’s telecommunications lifestyle. It’s good for the soul.

The conclusions I have come to through this mental debate might shock some of you; and they might also give pause to some of the assumptions which Australia’s telecommunications industry is currently making about what the future NBN retail environment will look like.

The first step in my mental process was to examine what retail competition currently rests on in Australia’s current broadband environment. Basically: Why do we buy the broadband services that we buy? And what are the available choices? The conclusion I came to when thinking about this issue is that much of the basis for retail broadband competition in Australia today is the principle of mitigating the limitations of existing technology.

If you consider Australia’s current retail ISPs, you’ll quickly realise that the services they provide are not fundamentally the same. If you talk to Internode customers, they will tell you that the company’s “network” is simply better than that of rivals like TPG, and so they get more reliable broadband services than they would elsewhere. If you talk to iiNet customers, they will usually tell you that the company’s broadband network is pretty reliable and they get great customer service. If you talk to Telstra customers, they will usually tell you that they’re with Telstra because they always have been.

There are several exceptions to this rule: Dodo, TPG, and Exetel. If you talk to customers of these fairly cut-rate ISPs, they will usually tell you that they get great prices and sometimes great download quotas (or even ‘unlimited’ quotas), that their ISP’s network performs acceptably “most of the time”, and that the rumours of awfully congested ADSL service at night are not true.

In all of these cases, a large part of the conversation revolves around the theme of network quality. ADSL is a rather shaky technology, built on a copper network which was never supposed to be used for that purpose. It’s a pain in the ass to get a new ADSL connection set up in the first place, usually requiring several weeks of work as your ISP goes back and forth with Telstra’s wholesale division, and most people are loathe to tinker with their connection in case they have problems.

ADSL broadband connections are notoriously unreliable when it rains; there can be problems at the telephone exchange layer which are tough to troubleshoot, and running IP telephony over ADSL can also be a tricky proposition. It’s for all these reasons that Australia’s current broadband customer base places great stock in having a “reliable” provider, usually with good “customer service”, and one that has a solid “network”. Those customers who do place more emphasis on price when buying ADSL will still admit they are making some potential compromises on network quality to do so, by going with an ISP like TPG.

Much of the reputation that companies like iiNet currently enjoy is based around this idea of mitigating technology. iiNet and Internode in particular have really based their customer service proposition over the past decade on the idea that they will go into bat for their customers against Telstra; helping to smooth over the inevitable connection and ADSL quality issues through a combination of their own great customer service and their understanding of the dynamics of network integration (especially with relation to customers’ interface at their local telephone exchange).

Other aspects of ISPs’ current service offerings are also based around the limitations of ADSL. All of the large ISPs have invested a great deal in network upgrades that will specifically allow them to deploy a very specialised form of IPTV over the shaky ADSL infrastructure, with content being piped in from overseas and mirrored everywhere locally. They have also set up extensive local gaming servers, partially in an attempt to minimise lag time to overseas platforms. And of course there is also the key historical differentiating factor between Australian ISPs — which have available DSLAM ports in your area and which do not.

However, the important thing which I’ve realised recently is that when the NBN’s fibre is rolled out, all of these factors will cease to exist as competitive factors.

By its very nature, fibre is an incredibly more reliable and flexible technology than ADSL. Once the NBN fibre infrastructure has been set up once at your premise, it will never need to be set up again. It will not require weeks of back and forth between an ISP like iiNet and Telstra in order to get your NBN connection working. It will be a matter of seconds to get it connected. It will similarly not be a matter of weeks to switch providers, should you decide your current ISP is not up to spec. It will be a matter of minutes, or at most, several hours.

The vastly reduced latency which fibre can provide, coupled with its extreme level of bandwidth (by the time the NBN is fully rolled out, 1Gbps speeds will be widely available) means that there will broadly be no need for Australian ISPs to deploy specialised infrastructure to stream content from local locations to Australians over the NBN. Instead, it will be a simple and efficient matter to stream that content from the overseas datacentres of companies like Apple, Amazon, Google, Hulu, Netflix and others to Australia. Or, if you prefer, it would also be a simple matter for Australian content providers such as Quickflix to do the same, independent of ISPs.

With a modernised NBN fibre network that never goes down, requires only the smallest amount of maintenance, provides uniformly excellent service and speeds with minimal congestion, you start to see that much of the current customer service, network quality, content availability and technical capability advantages that Australia’s current crop of mid to upper-tier ISPs enjoy will largely disappear.

What this leaves Australia’s ISP market with is the strength of the lower-tier providers: The ability to compete primarily on price.

Now, I am sure that the ISPs won’t see it this way. ISPs will try and entice customers to sign up for heavily bundled packages with value-added options such as the FetchTV and T-Box IPTV platforms; they will try to market additional access to quota-free content; they will try and tempt us with bundled voice and mobile options; they will add on free web hosting, anti-virus and anti-spam options, and they will keep trying to charge us through the neck for all of these services, to ensure their profit margins remain intact.

However, I am finding it increasingly hard to believe that such options will be tempting. Very few of my friends, family and professional colleagues buy much from Australian ISPs apart from basic telecommunications services. Everyone pays for a broadband connection, most also have a fixed telephone line, and we all also have a mobile phone.

But, when the NBN is rolled out, why would you sign up for an IPTV plan from your Australian ISP, when you could just download the shows you want for a price from a global provider such as Apple iTunes (or even, shock horror, BitTorrent?) Why would you need to pick a premium ISP with great customer service and ‘superior’ network quality, when the NBN fibre will never go down and every ISP will provide a more than acceptable level of service over this superior infrastructure?

In September last year I wrote an article highlighting the fact that many aspects of the current crop of NBN plans, such as tiny quotas, unrealistic shaping speeds and the lack of quota-free traffic between geographically close NBN nodes are all aspects which have been ported directly from the previous ADSL broadband world, yet have little or no place in the new NBN reality. But what I’ve started to realise is that it’s not just NBN plans which are stuck in the past; it’s the way that ISPs are thinking about all of the NBN opportunities.

The truth is that when a technology changes, the ecosystem around that technology also changes drastically. And when it comes to Australia’s current crop of ISPs, the entire way which they have set up their businesses and service offerings are shaped around the fact of ADSL’s unreliability and slow, asymmetric speeds.

But in an NBN world, where the underlying broadband infrastructure will be so much more reliable and so much faster and more responsive, there will be no need for customers to pay the premiums which ISPs like Telstra, Optus and iiNet currently enjoy, when basic broadband access is already so fantastic. And with the world of content suddenly at their virtual doorstep, there will be very little need for that content to be piped into Australia with a big fat extra profit margin slapped on top by local ISPs.

If I think about the world of the NBN, I think about signing up with TPG or Dodo for a 100Mbps plan that will probably offer me a terabyte of quota for something like $50 to $60 a month. Because when the fibre cable running to my premise will be so fantastic, there will be no more need to keep paying top-tier ISPs insurance money to ensure a steady service or for value-added products; that reliability will just come built-in, and the best value-added broadband products will come from global suppliers.

Now, I’m sure there will be a lot of people who will disagree with this commentary. So I want to challenge you: If you can think of why people will pay premium prices to access the NBN, tell me so. What could an iiNet or a Telstra possibly offer a NBN customer that a cut-rate provider like TPG can’t? I really want to know, because I’ve been thinking about this for several weeks, and I can’t come up with jack.

One further note: I am aware that I did not fully examine the implications of fixed and mobile bundling in this article. This does represent a significant advantage which a telco like Telstra can offer over a telco like TPG, although again, I feel there is a significant price aspect to this issue. FYI, I will examine this issue more fully in a follow-up commentary, as I feel it hasn’t been explored fully yet; in particular, with the potential entrance of Vodafone into the fixed NBN broadband market.

In the meantime, debate away: Because I am genuinely interested in your responses. Convince me that I shouldn’t dump iiNet and sign up with TPG when the NBN rolls into my areas. Because right now that’s going to be my default approach.


  1. To me, a deciding factor will be local service.. Thats why Im with IPrimus.. I can ring and they answer and they give me the support i want straight away. I can get the Case officers phone number and call back if needed. And from an Australian base..
    Optus is ok, but still long winded, Telstra is shocking..

    • I can see your point, but I only ring iiNet when there are problems with the ADSL — which there are every year or so, off and on, especially whenever I move house. But I would liken the NBN to being with Telstra for mobile. Its Next G network is so solid, you only have to call them when you change your plan or something, which is almost never. The rest of the time it just works and there is no need for customer service contact.

      • Then it should be quite cheap in the scheme of things to hire Australians.

        If you’re dropping the number of support calls markedly, then you’re also dropping the costs of those calls markedly. My experience from within the industry is that outsourced employees simply do not have the ability to action what they promise. There is no institutional trust. They are, effectively just monkeys.

        Amusingly the NBN itself may enable Australians to be hired to support Australians at a lower cost base than they are currently. Whether it’s telecommuting, or more regional based offices, the associated costs of running such a workforce may decline.

        iiNet I believe, use native English speakers in Sewf Efrica and Nuw Zulund during their daylight hours. Maybe not true blue, but certainly an interesting way to provide good service during traditional overtime hours.

        Aside from that, your main point that price will be the only deciding factor due to identical infrastructure overlooks the obvious contention ratio game providers play. There will always be the cut price over-contended providers and their customers will complain. Complaints cost time. Time is money. Either they will upgrade or they will see their customers leave.

        Other providers aim for better contention, and they often develop madly loyal customers. On that topic, it will be interesting to see the rate of churn from the Internode brand now it’s no longer distinct.

        People are fickle. Price is not the only factor. There will always be people who have been burned who will never ever consider a provider as a choice no matter how cheap they are. I certainly won’t ever fly Tiger again.

          • I can see only a few reasons to not go the cheapest route…

            1. video services, games, or some other form of value added service
            2. Service – while it’s true that the only times most people call their ISP is when there is a service interruption, or they want to upgrade/downgrade the service, or they are going on holidays and want to suspend the service, some of the interactions and greater time involved in sorting things out can cost much more than the small price differences between services.
            3. Bundling with a mobile service that has good coverage
            4. Contention ratios

            Horses for courses…

      • I think the point here actually demonstrates the whole point of the NBN. Different people have different reasons for selecting different ISPs/RSPs.

        For most people, price is a factor and a major consideration. For other people, it’s the features – (download limits, shaping policies, etc) – of the available plans. For some people, ideological concepts have an influence.

        In the current landscape, many of these factors – (and especially download limits, shaping policies, etc) – are governed by the underlying wholesale network. For instance, a number of ISPs have different plans available based on whether your connection operates on their own DSLAM infrastructure or on someone elses DSLAM network – (Telstra, Optus, NEC, AAPT, etc).

        The NBN ultimately removes all of that differentiation, and makes the services that the ISP provides you becomes the deciding factor.

        The ISPs who do it right will flourish. The ones that don’t will wither.

        ie: market forces

        The NBN removes the situation where one wholesale provider is available to service your connection, which artificially – (in many cases) – prevents you from choosing the ISP you want to choose.

  2. So to crib note that, you can look at whats on offer now, and extrapolate from there.

    Bigpond currently offers a 500 Gb/month service for $90 a month.

    TPG currently offers a 500 Gb/month service for $50 a month.

    Right now, it can be argued that the difference in costs is due to the investment made, and the quality of service provided.

    Under the NBN, because the technology pushing that service is universally identical, that argument no longer exists, so there shouldnt be a big difference in cost – both companies have the same underlying cost for the connection, so after that its merely profit margin.

    I agree. Under the NBN, both companies should be offering that same service for the same price, or near enough. A few dollars isnt a huge difference, and can be explained away as a difference in support provided, but a $40 difference cant.

    What I’m looking for is a plan that , for say $80 to $120, gives me my home phone, a fixed line connection, my mobile, and the ability to link my mobile broadband usage to that fixed line. If I have 500 Gb at home, why cant my mobile bandwidth be a part of that?

    I expect the first company to offer something like that will get a LOT of business. We all have mobiles, we all have laptops or tablets. Yet their minimal net usage are treated as a cash trough by the telco’s. $35 a month with TPG for 9 gig is cheap NOW, but in the post NBN world should be unacceptable.

    How the mobile net prices work when they are a secondary connection is something I’m looking out for. Its a deal maker/breaker for me.

    • I agree, I am going to be very interested in the mobile bundling. If Telstra’s fixed broadband prices come down to where iiNet’s are, and they provide a decent mobile bundle discount, I could switch to Telstra on the NBN.

      • Surely it will be Telstra’s big play, Bundling mobile+content? Also more offloading via customer premises to keep their 4G network nice and speedy? That network is going to be their big advantage, content deals the next one?

    • “If I have 500 Gb at home, why cant my mobile bandwidth be a part of that?”
      Because mobile bandwidth is much more expensive to provide?

      I agree otherwise, but that question struck me as odd.

      • Fair statement Karl, but what will the situation be for the 4% of population who are being given mobile broadband as they primary option? Or the other 3% using satellite for that matter.

        If one of those is their only option, how is it fair they get gouged for pricing along current lines? To compare, right now its $35 for 9 gig with TPG versus $50 for 500 gig with the same company.

        If they arent gouged, the ACCC will be pretty interested if the prices arent the same across the country. At which point, if plans are rougly similar despite the version, then hybrid models arent so far removed from reality.

        And with the NBN delivering a much more connected lifestyle, it will be something the telco’s need to consider.

        Wouldnt be hard to do, plenty currently discriminate between peak and off-peak, so why cant they just add a third category for mobile, and set a 10 or 20 gig portion as part of the overall cap?

        • But the 4% aren’t being given mobile broadband, they will be getting fixed wireless.

          very different kettle of fish…

  3. It will be interesting to see if your prediction is true. I thought the NBN would have the opposite effect, magnifying the difference between ISPs. Because any bad performance will not be governed (or blamed) on bad copper, congested exchanges, neighbours hot water heaters or compact flourecent lights (all excuses for evening slowdowns I have seen put forth by reps) it comes down the the performance of the ISPs nation, internatinal links, CVC provision, contention, etc.

    • I think a few years ago, it would have indeed had the opposite effect. But if you look at the very few ISPs which are left in Australia now, they all have pretty much awesome internal networks. iiNet, Optus, Telstra, and TPG (courtesy of PIPE Networks) … I can’t see many problems with contention ratios there.

      • “I can’t see many problems with contention ratios there.”


        Based on a current average resident to ISP downstream bandwidth of 5mb – I’ve seen numbers as low as 3mb and as high as 7 mbit in the last year – NBN resident to ISP bandwidth will be up to an order of magnitude times two better (i.e. 20 times as fast).

        The amount of overseas bandwidth will remain roughly the same. Download bandwidth will increase by a factor of up to 20, with future bandwidth expansion another 10 times faster again (1gbit). I see a problem here.

        Your average Joe may not give a crap because he doesn’t know any better, but the internet geeks braying for the NBN will quickly know who is the fastest ISP (and flock to them).

  4. I’d choose carefully if it means getting my NBN isp carefully if it meant getting into a 12/24 month contract. Those ISP’s like iinet who I’ve have used previously and have had superior customer service. The service providers reselling the NBN are likely to have there own internal teething issues and great customer service and technical support is everything. If its not to solve my issues, to admit that they have an issue and fix it fast is necessary.

    That said if its month to month, I’d consider the cheapest as well, knowing I can now fall back to tethering the data plan off my phone, and shop around other ISP’s.

    Just to add we (Australian bit torrent / p2p users) have iinet to thank to upholding the privacy rights of the consumers I doubt that the legal battle could have been in better hands.

  5. CVC provisioning (over or under) will be one of the fewer cost levers left for ISPs to pull. Cheaper ISPs will under provision CVC at each POI (or to the aggregator if they go down that path), more expensive ISPs will have sufficient CVC to handle peak demand, and will charge accordingly.
    Not going to name names, but take a look at whirlpool to see which ISPs have speed issues each night. That’s what I’m talking about :)

    • hey Douglas,

      actually the anecdotal Whirlpool evidence from Whirlpool which you cite doesn’t really match the current evidence on ADSL. Check out this study from ZDNet:


      What it showed was that there was very little actual evidence for big speed differences between the major ISPs. It’s only when you get down to the 10th biggest ISP (at that stage, AAPT), that you see a significant difference.


      • Hi Renai,
        That article/graph kinda proves my point, actually.
        Telstra average is around 6.5 looking at the graph. Dodo looks to be about 5.2Mbps. That’s a 25% difference right there big fella! :)

        • Yes but would that difference be magnified with higher speeds? I’m not really sure. In addition, I hasten to point out that TPG was actually second fastest on that list. And it has a strong international fibre footprint courtesy of its PIPE Networks acquisition.

          • I would expect TPG to be up pretty high. Along with PPC1 they also got the Pipe DF to all of the exchanges where they have their DSLAMs (which was a big reason for the purchase).
            If you translate this to an NBN world, it essentially means that they had unlimited CVC at no additional cost.
            Also in regards to that article you referenced, eyeball port 80 traffic (which this is testing) is always prioritised by any ISP worth their salt (and especially those skimping on AGVC/IP transit) It doesn’t show you what is happening to everything else. IE – why the heck are my torrents going so slowly??
            P.S.Having said all that, I am actually in furious agreement with you, it will all come down to cost at the end of the day when most end users are making their decision. Telstra will be reinforcing that when they are promoting their 4G network as well!

          • As I said elsewhere (Telstra take my money thread), I went looking at one (XL) at lunch and the local Telstra shop pushed me towards the Galaxy S II. Not even the S III…

          • Nice trolling :)

            Seriously though, as I said above, I want a package that combines my fixed line and wireless into one package. In this day and age we all have a mobile, most have a laptop or tablet (or both), and with the NBN promising such fast speeds (both fixed and wireless), then its a logical step to combine them.

            Mobile doesnt make NBN or fixed line redundant, it slots in exceptionally nicely as a complimentary technology.

            And given that it will be the only option for a statistically significant portion of the population (4%), the pricing for that option HAS to change, to the point its competitive with fixed line connections. TPG right now charges $35 for 9 gig for mobile broadband. Under NBN it should be closer to $50 for 500 Gig, like the fixed line plans.

            Follow on from that should be that fixed line customers arent gouged for mobile broadband prices.

          • Disclaimer: I was being playfully facetious…!

            I am the first to argue fixed and wireless are complementary and the stats prove it.

            But it was fun being on the other side for a change and not needing facts, common sense or anything to comment.

            It really is much easier being on the dark side…lol.

          • “I want a package that combines my fixed line and wireless into one package”

            Me too. I honestly don’t understand why the likes of Telstra and Optus don’t focus on this much more strongly. I suspect it has to do with the difficulty of integrating their billing systems; plus Optus just doesn’t really seem to care about the fixed broadband market.

    • Totally agree Douglas. The CVC provisioning I think is going to be one of the biggest differences.

      A cut-price RSP like Exetel may only buy 30 Mbps for 100Mbps plans, while iinet might buy 60 or 75, maybe even 100 for certain PIR tiers. So we will still see slow downs at peak periods, like these cut-price ISP’s do today, just for a different reason. Artificial peak periods if you like….

      There will also be the bundling too. ESPECIALLY with mobile data IMO.

        • Hmmm, interesting.

          I suppose really alot of the speed differences, particularly people such as myself would see, are as a result of short-term contention AND damage/water on the line.

          Pretty woeful speeds overall on that study though…..just highlights the need for SOMETHING to be done!

          • For example, the night of the DragonX docking, I was trying to stream it (Friday night about midnight I think) and ended up using my Telstra Desire, because my normal 6Mbps had been reduced to less than 1.5Mbps!

            I’d certainly agree on average the speeds wouldn’t fluctuate, as the study indeed indicates, but it’s the reliability of GETTING those speeds I think I mean more than anything else. You know, 98% of the time, 90% of the time, etc….

          • Personally, I think there definitely were a lot more widespread contention ratios in the past years, until the mega-ISP industry consolidation resulted in just four large ISPs, all with fat pipes. I would find it hard to believe any of the major players left would be cheating. I do experience problems from time to time on my iiNet connection (especially with GOMTV, a long-running issue for iiNet customers), but I attribute this more to the nature of ADSL itself, than any real problem with iiNet.

        • That ZDNet article doesn’t show the complete picture though. TPG is mostly reliant on its own dslams and hence it’s own backhaul, it is not at the mercy of Telstra and Optus currently congested backhaul in a lot of exchanges. You can see Internode’s forum to see the huge improvement in speeds users have gotten now that they are slowly being migrated off TW and Optus ports.

          The NBN will actually magnify the differences between the RSPs. Getting congestion during peak periods? You can compare up to 3 other RSPs at the same time to see the difference.

          It will all come down to a combination of the amount of CVC, international/national bandwidth plus peering arrangements domestically and around the globe. The last mile will no longer be an issue like it is in the DSL world.

        • I’m with Renai on this one. CVC under provisioning at the moment is hard to quantify because ADSL has so many issues. Under NBN ISP’s that under provision CVC will show up very clearly as speed customers get is not governed by the final leg technology. ISPS under provisioning CVC will be forced to provision better as it’ll show up much more. Added to that NBN co have strict guidelines regarding CVC provisioning.. an under provisioned link make the NBN look faulty.

      • Funny you should say that as i have just moved to exetel as iinet were having constant latency issues for months in WA
        I move to exetel and the problem is gone
        I know which one i would trust to have enough capacity so i wont be moving!

    • I’m with Douglas on this one.

      On page 103 of the NBNCo Corporate Plan (Dec 2010) there is this statement:
      “The Connectivity Virtual Circuit (CVC) in the product construct is an aggregation point where the Access Seekers can choose to contend their traffic to create differentiation.”

      I would infer from this statement that NBNCo are expecting congestion at the POI, because if there isn’t congestion then there isn’t differentiation.

      If we assume that there will congestion with the budget RSPs, then this leads me to wonder if there would be any point in buying faster speeds for use during peak times. It reminds me a little of the similar problem with Telstra RIMs with insufficient backhaul – little point in paying extra for the 8Mbps plan.

      The difference is that both budget and quality RSPs are limited by the quality of the copper between the premises and the exchange. With the NBN the major limitation will be RSP’s network (CVC interface, backbone and interconnects.

      NBNCo are currently providing 150Mbps of CVC free at each POI until they reach 30,000 customers which should happen in Tasmania next year. At that point I think we will start to see the gap between budget and premium RSPs widen and the number of directly connected RSPs shrink rapidly.

  6. tl;dr

    so i may have missed it if Renai covered this: but I imagine that at each POI, the RSP can distinguish itself by installing some type of multiplexer (or whatever the equivalent is for fibre) that can offer better multicasting; like, a cheap RSP will give you phone, broadband and pay tv on one colour frequency (pardon me if I’m mangling the physics here), which will lead to slower speeds when all services are being used.

    Premium RSPs will be able to separate your data stream and offer Pay TV and other high bitrate services on a separate light feed. So your 100Mbps is still 100Mbps when everyone else in the house is otherwise watching TV.

    Can someone help me put these thoughts into the correct terms?

    • You are a bit off…

      An ISP might sell you a 100megabit plan (with phone and TV), but actually buy a 250 megabit plan from NBNCo.

      They would then segregate their telephone and IPTV offerings into – say 150 megabits, and leave the other 100 megabits to your internet access. (they would use QoS to rate-limit your internet service to 100 megabits, and rate-limit your other stuff to 150 megabits – out of your total of 250)
      They would be over-provisioning your service.

      A cheaper ISP, might sell you a “100 megabits + TV + phone” service. And only buy 100 megabits from NBNCo. Therefore when watching TV, you can’t ever get your full 100 megabits from your internet access.
      In this instance they would over-contend your service.

      Seperately from what they actually purchase from you, there is an aggregate bandwidth that they provide to all of their customers on each Point of Interconnect (POI). Broadly speaking these are fairly large geographical regions.

      (warning: numbers are pulled from thin air)
      Your premium ISP might buy 10 megabits for every 100 megabits of bandwidth sold to customers.
      Your cut-rate ISP might buy 50 megabits for every 100 megabits of bandwidth sold to customers.
      (it is very unlikely an ISP will buy 1megabit for every 1 megabit, though vastly more expensive business plans might end up with this kind of ratio – these customers would likely be separated into their own 1-1 contention ratio pool).

      I am not sure I have exactly the right terminology for you, but “over-contended” (too many users) and “over-provisioned” (too much bandwidth) are the 2 ends of the scale you are looking for.

    • What Peter said Adam. Also, the NBN currently only uses one “colour” on the fibre. The splitting of speed is done through other means than colour right now, although moving up to 1Gbps and 10Gbps may see us use more colours.

    • @Adam

      Right now, most of the ISPs (with the exception of Telstra with its T-Box) have broadly failed to sell much in the way of IPTV. I don’t really see this situation changing in the NBN. I’ve posted some further thoughts on this here:


      It is possible that there could be a difference between ISPs who provision extra for IPTV and those that don’t, but given that current IPTV streams require something like 4Mbps, I don’t see that being a big deal, especially for a connection of 50Mbps or above. Even on 25Mbps fibre it should still be fine. This is what I am talking about with a ‘step change’ in technology.



      • i suspect content deals will be nailed down first, and then provisioning on the basis of that – how many channels, the screen res and data size (1mbit video, 5mbit video etc). because of the current landscape theres not much room for those who want to push iptv over, say, Telstra ports, and traction in that market correspondingly difficult to achieve. there will be a point where the viewing market will become large enough to sustain as well, now its a bit early with trial areas and first tranche and greenfield fibre being the only areas you can get a toehold in.

        Telstra have definitely stolen a march with t-box – even ive considered it. but thats due to the way things are now – once the content deals come and the landscape changes to a primarily NBN one rather than primarily DSL+cable, there will be more change. NBNIPTV is very much a nascent market at this point; its worth speculating but i think its also worth remembering a lot can change also. the provisioning (over or under) is something that will really show itself when there are volume customers on NBN so the effect becomes visible – right now its a lot like the Telstra LTE rollout, a small and relatively empty network. im personally leery of making any predictions as to how successful – or not – IPTV is in say 5 years time, given how much (i expect) conditions will change from now to then.

        as for me id be looking to the 50/20 tier straight up, so to a certain degree i would expect little trouble with any potential services – it will be about what *content* is available on whoevers offering that gets me to buy into them. for me the content is the horse pulling the cart – and talking about the cart is a tad premature.

        • It will be interesting to find out. For my own personal usage, I stream a LOT of HD Internet video and so does my wife (hers more in standard definition); often we’re streaming both at the same time. So far I haven’t seen any real problems on our 16Mbps ADSL2+ connection; we only run into headaches when upload at the same time.

          I suspect that with anything above 50Mbps, streaming in general will be completely a non-issue; whether the ISPs provision properly or not. As Malcolm Turnbull himself has pointed out, IPTV streaming actually only requires a few Mbps. But on ADSL networks, a few Mbps is often all you have ;)

    • Opticom claims to already offer this (disclaimer: I have no personal experience of their products) and they call it an RF overlay. I have no idea how they offer it up at the wholesale level, presumably pay-TV providers can buy a whole suburb at a time.

      NBN made the decision to provide pure Ethernet, no overlays, etc.

  7. Well Renai, in some ways yes there will be positives, better consistent faster service being the most paramount.. The ‘resellers’ will need to flatten costs and provide a much better customer experience, ahem Telstra..
    We will see bundling with Broadband, Telephone, Mobile, and perhaps some sort of content provision like T-Box. Price will be paramount.
    As for next steps along the NBN, I’m guessing that the ability to track BitTorrent users will become much easier and also the somewhat dormant idea of internet censoring from MR Conroy will rear its ugly head again. When the government owns, they can implement their policies and then build their policies in.
    So its a mixed bag really..

    • I agree with most of your comment, but not on filtering; filtering in general is going away, I think — except for in the realm of child pornography, where it will stay.

    • I’m not sold on tracking torrent files either. They can track that you’re downloading peer to peer, but they cant track WHAT you’re downloading. To the ISP’s, there is no difference between downloading a pirated copy of The Avengers vs downloading the latest World of Warcraft patch.

      Torrents are such a cost effective way for large products to be shared, its become quite a common practice to use the method. And that means there are a considerable number of legal uses, which means you cant just blanket block torrents.

      If you try to block the sites/IP’s, there are also numerous ways to get around that. It took The Pirate Bay mere minutes to get around British Telecom blocking their site. Minutes…

      On top of all that, the capabilities of VPN’s make for a simple way of masking who you are. Given that VPN’s are a perfectly viable online security method, you cant just make them illegal along the way either.

      So overall its not as simple as just tracking torrent downloads.

  8. The other issue you have missed is international speeds. Check the whirlpool forums- how many isps have long running threads about international speeds at night? Plenty! The good ones (internode) dont.

    If you use a tpg connection side by side with an internode one, browsing speed is hugely faster with internode. This has nothing to do with the quality of the adsl link.

    • Really? I know Internode co-locates a lot of traffic, but TPG has their own international pipe. In speedtest results, the two are pretty even…

    • This is what gets glossed over in the article. Just because all ISPs have the same last-mile connection speeds, doesn’t mean they all have the same capacity in their core network, backhaul and international links. Far from making all ISPs the same, the greatly improved speed of the NBN will only magnify the differences between their networks. I for one will continue to steer clear of the bargain-basement providers.

    • Having used internode,iinet,exetel ,aanet and a few others,there is basically no difference in browsing speed between them all
      With the connectios TPG now have they would probably be faster overall now

  9. If it’s not bundled services, maybe it could come down to customer service. Not the reliability of the NBN itself, but the customer service of dealing with a human on a phone over billing issues or accidentally borking your modem settings.

    TPG has appalling customer service. If you have a dispute with them they’ll refuse to budge until the TIO are involved. In a customer service orientated world, this behaviour won’t bode well for them.

    Why should I put up with a company using a dirt cheap off-shored, poorly trained, Filipino call centre that has little or no power to help people when I can switch to someone else offing the same product at the same price minus the blood pressure issues? Although considering the awful reputation Dodo has always held and they’re still in business (I actually overheard a bunch of teenagers demanding a Dodo product today) I may be wildly off base.

    • As I wrote above:

      “I can see your point, but I only ring iiNet when there are problems with the ADSL — which there are every year or so, off and on, especially whenever I move house. But I would liken the NBN to being with Telstra for mobile. Its Next G network is so solid, you only have to call them when you change your plan or something, which is almost never. The rest of the time it just works and there is no need for customer service contact.”

  10. Renai, I agree that an important factor in deciding which RSP I sign up with will be the pricing. I would expect that the basic pricing will be very similar because each RSP will be paying the same rate to NBN Co. I don’t think there will be much margin for great price differentials. If we look at the competition between Coles and Woolworths it seems to me that there is not a big price difference and their efforts are directed to appearing to give the customer greater benefits for shopping there. I think the RSPs will be the same.

    The first difference that I will be looking at will be customer service. If I want help from my RSP I want it now and I want it to be real. I don’t want to be foisted off to someone who can’t speak good Aussie english and is reading from a pre-prepared script. I want a good billing system that is going to track my usage acurately in real time and show what my account balance is in close to real time . I want to control how much and when I make account payments. If I want to make a payment every week I should be able to. There is no way I will trust anyone else but me to access my bank account so they can stuff direct debit where it belongs.

    The second difference that I will be looking at will be bundled services and the cost. I will be looking at what it will cost me for adding VOIP. I have to decide if I will bundle in my mobile or will I just go to one of the pre-paid providers who are cheaper than the main players at present. I will probable look at IPTV content offered but doubt if I will take it up. I am not big on TV apart from sport coverage and can’t see myself paying for a service which I am going to hardly use or doesn’t give me value for my money.

    As I am expecting to be on the NBN in the next 12 months I also have been thinking about what I will do and the summary is:-
    1) Pricing for my needs
    2) Customer service provision
    3) Add on services that I want and can use.

    • +1

      I am thinking broadly along the same lines as you; with the exception that I suspect customer service will become much less of a focus on the NBN than it is now.

  11. Life under the NBN umbrella is going to be totally different to today where ISP’s have at least some control over product differentiation, Annex M and Naked DSL are prime examples and the internal cost control that follows when they have their own DSLAM gear installed in Telstra exchanges.

    The NBN Co will supply wholesale pricing at the same rate to all ISP’s so the marketing emphasis will move to the bundle, Optus are already moving to downplaying fixed line BB, that is the emphasis is on phone and wireless with ‘free’ fixed line broadband quota thrown in.

    The support situation will not change much from how it is today with ADSL and Naked DSL, ISP’s take support calls and in many cases just pass it onto Telstra Wholesale for the fix, take out Telstra Wholesale and substitute NBN Co for the future, the difference of course being that the biggest ISP in Australia Telstra will also be passing on the support calls where NBN Co intervention is required.

    I suspect the support calls will be quite extensive relative to today as the ONT box with its own UPS which always remains the property of the NBN Co is quite a step up from a simple two wire copper telephone socket screwed onto the skirting board that requires zero power from the residence.

    The quality of the ISP call centre response is indeed important as many here have stated, but under the NBN I see the quality of the NBN Co call centre as being of much more primary importance.

    • Someone correct me if I’m wrong, but I don’t believe the customer will ever speak directly with NBNCo.?

      The ISP’s are the ones providing the customer service. They buy the service and assume all obligations, other than physical damage/upgrades, that come with it. Including troubleshooting.

      At least that’s what I’ve been lead to believe in the various documents about RSP’s I’ve read??

      • Cant see it making a huge difference to the end user. They have problems, they call their ISP, problem gets fixed. Who fixes the problem isnt an issue to the end user, as long as its fixed.

        It does, however, raise an interesting point. Do the ISP’s/NBNCo need techo’s that are within a reasonable distance of every fibre delivering exchange? For arguments sake, Broken Hill.

        They are large enough they will have fibre, but small enough they probably wont generate enough problems for a full time tech/presence. So if someone DOES have a problem, would someone be flown in from Canberra/Adelaide/Sydney/Melbourne to service the problem (or group of problems over a few days)?

        Plus, what would you expect the service standards to be? Given its going to be the key communications point for most people in the years to come, should any problems be fixed within a 24 hr period? 48 hr? 72?

        • I think NBN Co will be big enough by this point to solve these problems fairly easily — and if not, it will be contracting jobs out as Telstra does.

          • Yeah I agree, was more asking the question for the sake of debate. I expect it will work out very much like how things work now, I really cant see it being much different. All those telco technicians are going to need a job in the post-NBN (and fill those jobs created by fibre replacing copper) so at worst it might mean a new uniform and some retraining.

            But given the blanket approach the NBN provides, will that change our expectations of the service standards? At the moment, we have two sides to our comms – net and phone – and we treat the service of each separately, but in the future they will, for the most part, be one and the same.

            Will we be right if we expect something more?

          • That’s what happens with technology, there’s extra complexity.

            This is a good thing, it means those who want more can get more and those who want the basics can just use the basics.

            Bit like the mobile phone and the TV remote, which my dad was scared stiff of. He said how am I ever going to learn to use “all of this”? I said your not, here’s how you use the phone (just like the old landline) dial, speak, hang up. As for the remote the channels are now programmed, here’s on/off, volume and the numbers are the stations, don’t worry about the rest. And away he went.

            Then there are those who wish to squeeze every last bit of usability from them… horses for courses.

          • True – we pretty much have the structure in place now, just spread across telco’s and ISP’s. Its not a huge leap into the world after NBN.

            But with that combination of services, how will our personal expectations change? To the layman, its the same thingy attached to the wall, so there’s only one person fixing it, instead of two. You dont need to figure out whether its Telstra you need to call or your ISP, it will be one and the same, so a phone or net problem will effectively become the same thing.

            Will that change the average persons expectations? Are those changes in expectations justified? Will it really matter? Just questions to ponder over the next few years, nothing more.

            If the repair responsibility stays with NBNCo, its a moot point. If it moves to the provider, then companies like TPG need to adapt to that, and either increase their staffing footprint, or contract out which reduces their profits.

          • I think understanding about NBN Co will gradually grow over the years until people understand it as well as they currently do the electricity, water and garbage collection providers.

          • Good point. Without belittling anyone… more mature (note: I’m roughly in that group myself as a 40-something), its the younger (in general) more tech savvy age group that will see the best of the NBN. And they will understand it as a naturally as Gen Y understands video recorders…

            Other thing is that with a FTTH product, the number of day to day repairs that get done currently as a result of the copper lines being in bad condition will stop. So there will quite possibly be a glut of techo’s floating around anyway who currently fix what wont need to be fixed in the future.

        • @ GongGav,

          “Cant see it making a huge difference to the end user. They have problems, they call their ISP, problem gets fixed. Who fixes the problem isnt an issue to the end user, as long as its fixed.”

          Exactly +1.

          The end user will not care who repairs, as long as it’s repaired in a timely and professional manner. They/we will care even less…who owns the network. They/we will basically care about price/plan/service (value) and as mentioned repairs.

          IMO, anyone who suggests the repairer or particularly the network owner is important to the average Aussie, is either mistaken or on a mission.

          After all, when people signed up to iinet, People Telecom or whoever previously do you think they said, no I’m not signing, because that network belongs to Telstra? Or no I refuse to be “forced” onto copper? FFS.

          This is a generic comment… I am continually amazed at the ridiculous claims levelled at the NBN and only the NBN, which these naysayers refuse to translate across to other areas… which they would readily accept, if their logic was sound.

          Just shows how important the NBN is to those of us who want it for future prosperity and even more important to those who don’t want it, because it has already cost them two elections (the last one, readily admitted to/from the horses mouth) …

  12. If you are referring to my post I never said the customer calls the NBN Co direct, I even emphasised the NBN Co will take over the role Telstra Wholesale has today with ISP’s.

  13. It isn’t the last-mile (currently ADSL, soon to be NBN) connections that get congested on cheap ISPs. It’s the rest of the network, from the exchange (soon to be 121 NBN POIs) to the actual server you are trying to download from where the bottlenecks lie. This isn’t going to change with the NBN (if anything, it’s going to get more severe) – as other commenters point out, budget ISPs will underprovision their CVCs (connections to the NBN POIs), their international links and probably many other things too.

    Under the NBN you will (as always) get what you pay for.

  14. Surely still a lot of the same old. I agree about the potential for different kinds of bundles. Uptake of more advanced network services like VOIP and IPv6, plus more and more streaming content that could be sweetened somehow. Things to take advantage of the connection; apps, game servers and content servers will continue to exist.

    Congestion and backhaul management will continue to be real things; whether they matter is another question.

    I think at the end of the day, all this is forgetting one magical thing. Marketing. Demand for customer service isn’t the same as thinking its worth having. Businesses may see a relative downsize in service departments, while broadcasting more loudly how good their service is.

    • “Businesses may see a relative downsize in service departments, while broadcasting more loudly how good their service is.”

      I agree this will happen. *sigh*

  15. Customer service will always be a factor (which is why Telstra have finally picked up their game), however, in an NBN world how much “service” will be required once you are provisioned and underway. At the moment, you can need quite a lot of service, but this is mainly to address the copper line issues of ADSL.

    The big unknown for me is the backhaul – i.e. what happens once the NBN drops you off their POI and hands you back to your ISP. Being in Tas this is always an issue – both with the line to the Mainland (still only 3 of these, 2 owned by Telstra, none NBNCo), and from there to the rest of the world. How does this all work in an NBN world? How will this be differentiated from a customer perspective?

    • Are you saying that TPG, which owns PIPE Networks, is going to have slower international connections than iiNet, which has to lease capacity from other players?

      • This is something speed test sites should make available, very few do. No ISP is going to admit that they don’t provision enough for evening peak so that users aren’t slowed. There is no imperical data made available. The best you can do is have a good guess based on your personal experience and get a feel for it from post on the ISPs forum.

        From personal experience for speed I was ibtaining from international sites the international link quality would go for me, top to bottom:

        1 Internode
        2 Netspace
        3 iiNet

        • I just checked iiNet connecting into the USA, I get between 1400k and 1500k bits/second upload speed (that’s the TCP stream itself, so add a bit extra for overhead at the packet level if you feel like it) and I get about 9M bits/second download from the USA (or almost exactly 10M bits/second download from iiNet itself) again plus some protocol and packet overhead.

          That’s a single-threaded stream, ignoring the slow start TCP congestion avoidance when it first connects (i.e. small files average to a slower transfer rate).

          Anyhow, it’s not all that far from the ADSL line speed, and it’s right in the middle of the evening peak. Doesn’t look congested to me, but I guess I’m not going to sit repeating the same test every night.

          What’s the NBN minimum standard? 12M/1M or something. I’m not feeling too ripped off.

          • I wish I experienced that. With Internode I get full speed to the US, single stream at any time. During peak times Netspace would drop to 600KB, iiNet 150KB. I cannot say that the iiNet one was definitely their overseas link as local sites were a dog too. Apparently being switch to iiNet at Netspace prices, or so we were told, we were unprofitable and according to a Netspace rep we were better of just churning away because it was unlikely our evening speeds would be improved. iiNet reps just ignored the situation.

          • I think congestion issues were a factor when Australia had more and smaller ISPs, but now that there are only a handful of large ones, with quality networks, it’s hard to say that it will be.

      • The aggregate maximum bandwidth on ADSL copper tails across all premises far outweighs the sum total bandwidth of our international links due to bandwidth leveraging.

        Congestion is a function of both the size of the funnel and the quantity you’re trying to ram through it at any point in time. Assuming TPG “owns” more international bandwidth than iiNet leases, TPG users may still suffer greater congestion because TPG’s plans cater to heavy bandwidth leeches.

        The GPON NBN actually introduces a chokepoint in tail circuits via CVC (congestion) charges that doesn’t exist on home-run copper topology.

        • “The GPON NBN actually introduces a chokepoint in tail circuits via CVC (congestion) charges that doesn’t exist on home-run copper topology.”

          The equivalent would be exchange backhaul wouldn’t it?

          Whether it is ISP owned they probably wouldn’t have a problem but they have to pay for backhaul otherwise and could thus underprovision it. I have personally experienced this happening when one ISP supplied minimum Telstra backhaul to customers on plans they considered unprofitable. They said nothing publicly, in fact they ignored complaints on whirlpool. Luckily a tech from the aquired company investigated it and suggested we churn as it wasn’t likely to get better any time soon.

          • I believe that in the roll-out phase the NBN have decided to suspend all CVC charges until they can get some customers (someone correct me if that’s wrong). So right now there is no bottleneck.

            When the CVC charges kick in, the relative size of the bottleneck will depend on how much the RSP wants to pay, and they would probably be kind of dumb to provision backhaul bandwidth at a significantly different size to their CVC provisioning (but you never know).

            By the way, the present-day ADSL network is a lot more complex than you describe. The fundamental unit of transfer for an ADSL modem is ATM cells, but on top of this are (usually) Ethernet packets, then on top of that PPP (to make PPPoE) and on top of that TCP/IP. Some providers (from memory only Telstra) use PPPoA and some use straight Ethernet over ATM (the best option).

            So when you send a PPP stream, it doesn’t get unpacked at the DSLAM, it travels as ATM until it gets to a PPP concentrator and then gets unpacked down to TCP/IP.

            The exchange backhaul is ATM and ATM happens to support the idea of a “network contract”, and also explicit flow control. These parameters should be visible in the modem somewhere (because flow control is useless unless it is visible at the endpoint). I’ve never been bothered digging out how to read the details, I’m sure someone has done the exercise. Thus, exchange backhaul as far as the PPP concentrator should be directly measurable by the end user.

            This makes me wonder, will NBN implement ethernet flow control? It’s kind of a pig to get it working in a fully end-to-end manner on a large network, and most providers don’t bother. Could make a difference for people who want tight management of their queues (probably most people just don’t care, welcome to “why bother?” land).

          • I thought that the first 150Mb was free for now? The problem was if you had only a few people using the CVC you had to have at least 100Mb even for 1 user or they could never achieve full speed.

          • I am not sure what the NBN use. One interesting thing I found when “the exchange” was congested with Telstra (well, not really congested, only iiNet people shifted from Netspace experienced this congestion, on many exchanges) there was major packet loss between the DSLAM and the first hop connection to iiNet. As the link got slower and slower, more and more packet loss, like 20%. Never experienced this on Optus DSLAMs. Wasn’t the modem, not packets lost between it and the DSLAM.

  16. I’d be interested in some sort of bundling with wireless services & Foxtel with Telstra. If I can get a 100/40 NBN service via Telstra, as a bundle, for the same price as if I had Foxtel & 2 mobiles with Telstra, but NBN elsewhere, I’d get it with Telstra. Even if it was $10-$15/m more to keep all my services with Telstra, I’d do that. One point of service, one bill, etc.

    • I think this is going to become a bigger and bigger goal/desired product as the NBN is rolled out. The playing field will be levelled, so to differentiate yourself the ISP’s will need to do things like this.

      Its called competition… :)

      Otherwise, as was said by others above, why wouldnt you just pick the cheapest ISP? The reality for most is that the NBN will overdeliver for a number of years, so even if those ISP’s underbuy their CVC, it wont be an issue for quite a while. By which time it will be pretty clear what ISP’s have failed to deliver.

  17. I would argue that it already makes sense to ditch iiNet for tpg.
    What isp invests in improving its network instead of advertising and buying customers?
    What isp owns an international link and a backhaul network within australia?
    What isp is still the only one to deliver voip in a way that requires no extra hardware?
    What isp offers iptv for free (its crap but not much worse than fetch)?

    Tbh i wonder if tpg priced itself 3 (maybe 4 times) as expensive and marketed itself as a premium isp it would get as many fanboys as iinet/internode.

    Tpg over by far the cheapest pricing, they have the best network in real terms (owning an international link is a whole level above having heaps of Pop’s, instead of offering unmetered content they “unmeter” everything. I think tpg is the isp that best understands the “fat dumb pipe” concept that could be the future of australian isp’s.

    • Yes, I largely agree (and I have heard a lot of good things recently from TPG customers), but because of the unreliability of ADSL, customer service is still an issue; which is why I pay my monthly insurance excess to iiNet, to be able to call them when something goes wrong.

      It’s also why I must always have a backup 3G/4G mobile broadband connection handy, so I can still get to Delimiter when the ADSL goes down. I am literally on the Internet 16 hours a day most days; and often automated stuff is happening when I am sleeping.

  18. An interesting thought, Optus AKA Singtel. Vodaphone etc
    we have students from Singapore, the possibility arises for a “port” of service within Singtel to a “temporary service” on the NBN in Aust retaining email addresses and possibly even VOIP services and numbers, maybe temp change or additional Aust No. Vodaphone is a major global player, many business accounts, the same for them

  19. Things will change, but there are always options for competition. Undoubtedly some providers will fail in the revolution (just look at Fairfax).

    Service WILL still be important. Service is important not just for support but at sign-up, for billing, plan changing etc.

    Economics and number of POIs serviced. There may be space for local providers delivering on only a few POIs (like the local IGA competing with Coles).

    Marketing and product design. I’m sure some will come up with plans that are more appealing mixes of speed and quota

    Network from the POIs to places want to go. NBN isn’t doing this. Telstra, iiNet and TPG have got to be in better positions than some others.

    Efficiency – this is where Telstra may really suffer as they are still a huge bureaucracy. Can they control their costs?

    Video (that you pay for). Telstra has an advantage here as 50% owner of Foxtel. For all new installations, why would I want to pay for a satellite dish and a fixed connection point in the wall as well as the NBN connection and viewing anywhere in the house?

    Cloud Services – e.g. if an ISP peers with a large provider (like Amazon) and others can’t that gives a better option.

    Bundling: all comms from one provider will be attractive to some.

    Bundling Value Added Services: I’m some will be dreamt up once the bandwidth and consumer base is there.

    The NBN gives more scope for competition. It just won’t be over the same things as today. Just think about the electricity market – there are multiple providers offering different pricing and packages. Some do better than others.

  20. I tend to stay with a company that I have had good experience with, so I will be staying with Internode/iiNet.

  21. “If you can think of why people will pay premium prices to access the NBN, tell me so.”

    Certainly. Glad to oblige :-)

    I can think of two things off the top of my head.

    1. Bundling and services based on corporate level arrangements. Ie Content provider A signs deal with Service provider B to provide content at a discount rate. This allows for price differentiation for content that may be available elsewhere but at a higher price.
    2. It’s called ‘Customer Service’. That’s CUSTOMER service. Not the content, not access or terms of the deal, but the actual customer services. And this is a big one. Would I go with a provider that puts me on hold for 45 mins, connects me to a o/s help desk with bad English who doesn’t know what they are talking about or with a provider with shorter wait times and better support? And what would I be prepared to pay for it.
    I think this is a big plus for consumers. They will be able to judge providers on a level playing field on issues that matter to them. Quality of access is guaranteed, after that it’s what is behind it that will become more critical.

    • I kinda addressed both of these in the article ;)

      I don’t buy any content now, apart from StarCraft II streams online, downloaded video through iTunes, DVD rentals and so on. Why would I need to buy it directly from my ISP in future? What will iiNet be able to offer, for example, that an independent provider like Quickflix won’t?

  22. Funny you should post this as I was “running some numbers” earlier in the week WRT residential vs business and the contention ratios typically associated with them.

    At present, our business connection is running at 4M, but bursting up to something like 20M on ADSL2+ (from Telstra), but it costs a bomb. The thing is though tat we need it as the business relies on connectivity.

    My home connection (Exetel) fluctuates between 4-8M, usually syncing at 6M, but I can tell you now that downloading anything is faster at the office even though the advertised speed is lower.

    I reckon the same will be true of NBN plans /eventually/. There will be plans offered to the market with much higher CVC provisioning, or at least guaranteed bandwidth, with the numbers actually advertised as opposed to this “up to 24M” crap. e.g.
    XYZ Corporation NBN plans :
    100/40M maximum, 12M minimum, 500GB data, $300/m
    100/40M maximum, 4M minimum, 500GB data, $160/m
    100/40M maximum, 1M minimum, 500GB data, $80/m

    I know which one I’d buy for work, and which one I’d buy for home, and both are less than what we currently pay!

  23. I don’t agree with Renai that it will come down to just price. I think ISP’s might morph more into the realm of providing their own services that sit on top if the NBN, such as IPTV, VoIP, gaming and cutting edge cloud based services and bundling the access to those with their connection for a streamlined setup. That way you only have to make one call your ISP to setup all these services, rather than individually subscribing. It’s probably the big 3 ISP’s that will be focusing on that.
    There will also be those that simply want Internet access without any of those extra services and that is where the TPG’s and Exetel’s fit in.

    • IPTV: Why would I buy this from an ISP, instead of an independent player like Quickflix, iTunes, etc, as I currently do?
      VoIP: Basically a commodity product on the NBN
      Gaming and cutting edge cloud based services: When have ISPs been able to develop these things? I already pay for this kind of thing (eg Xbox Live Gold membership, extra space on Dropbox etc), but I don’t pay my ISP for them, I pay an independent provider.

      I guess what I’m getting at here is: Why pay your ISP for an Internet service which can likely better be provided by someone else? Someone not tied to your telco infrastructure?

      • “IPTV: Why would I buy this from an ISP, instead of an independent player like Quickflix, iTunes, etc, as I currently do?”
        I’d say quota. ISP IPTV in all the cases I know of doesn’t count toward your quota. It would cost the ISP for on demand CVC but fixed time broadcasts could benefit from multicasting. Correct me if I am wrong here but someone external to your ISP cannot multicast? Who would pay for the multicast CVC?
        This next bit is theoretical. I saw someone asking about having multiple ISP connections on the one ONT and apparently it is possible.
        So maybe the video traffic could be provided that way. it would probably mean signing up for a monthly service and I doubt it would be doable at an exceptable price if they need to pay full price AVC for that service.

      • Renai as you mentioned sure there will be independent players you can pick and chose from to provide you with the services I mentioned (IPTV,VoIP, Cloud). However like I said it’s consolidation and bundling of these services that might offer the customer value in the future. ISP’s could form partnerships with these independents that will bundle their services with the ISP as well as the ISP being able to deliver more efficient back end networking to these services that will reduce the ISP’s operating costs by perhaps better routing, using less bandwidwith etc. This lower operating cost then might in turn enable them to deliver these servers cheaper to the customer.

        What I’m saying is for the ISP’s to create more value might have to shift its business model away from just being the big dumb pipe and provide a different sort of service than they have traditionally, and working with these new services that will take advantage of the NBN.

  24. I can’t see how the situation will be much different to now.

    True, the range of options will drop slightly. Now we have ADSL, Cable, and 3G. With the NBN it will be Fibre and 3G. The ISP’s ability to purchase their own DSLAM’s will go, but that at most effects costs. Apart from that – not much changes. We still have everyone using the same local loop technology, provided by a 3rd party monopoly at a uniform price.

    That doesn’t mean there won’t be competition. ISP currently compete in all sorts of ways, and I don’t see this changing. Examples off the top of my head are:

    – Advertising.
    – Unload and download charging.
    – Download quota’s and peak and off-peak.
    – Whether they charge for excess data or slow you down – and by how much.
    – Providing VOIP, including doing QoS on the backhaul to VOIP works reliably.
    – IPTV.
    – Mirrors.
    – Gaming servers.
    – Customer service levels.
    – Bundling of services (like a 3G SIM, which TPG offers).
    – Email hosting. Web hosting. Spam filtering. Provision of DNS names, DNS hosting.
    – Whether they put in a transparent HTTP proxy.
    – Provision of IPv6 like Internode does.
    – Content Filtering (like WebShield).
    – Additional services like Bob2 and the T-Box.
    – Port blocking.
    – Provision IP fixed IP addresses, reverse DNS lookups.
    – Monitoring the traffic for viruses.

    You are probably going to say some of these don’t matter much, and you’re right they don’t. I run a number of ADSL links which I spread around the various ISP’s. Even when there are multiple ISP’s going into the one location during day to day operation they are near impossible to distinguish. It’s a bit like various brands of beer or cola really. People swear by one brand and the ads make them seem worlds apart – but in blind tests the differences are hardly noticeable. Yet no one would say beer or cola compete only on price.

    So I recon you are 1/2 right. Yes, in the end they will all be offering much the same thing. But that is no different to now, and more importantly it doesn’t mean they won’t invent lots of artificial things to compete on.

    @alain – ONT box with its own UPS which always remains the property of the NBN Co is quite a step up from a simple two wire copper telephone socket screwed onto the skirting board that requires zero power from the residence.

    I’m not sure what magic technology you use alain, but every I know with a broadband connection needs a modem of some sort, and that modem requires power from the house. If you want to keep it running during a power outage it requires a UPS too. If you don’t maintain the batteries in that UPS you lose your power protection. Which it identical to the NBN ONT, because it is your responsibility to maintain the batteries in it too.

    And I am not sure whether it is true to say the ONT is a “step up” from an ADSL modem. The ONT is just a glorified ethernet switch, whereas an ADSL modem implements the entire stack – from ATM through to IP with NAT and tunnelling through to a web server. There must be 100’s if of settable options on your typical ADSL router. That can make them a bitch to set up. Which is why every ISP offers to sell you one pre-configured, I guess – just like NBNCo gives you a preconfigured ONT.

    Even if it the ONT is technologically more complex than an ADSL or Cable modem, it is not true to say it will be harder to use. Your ADSL Modem is indisputably far more complex than the old POTS modems, yet I came to view diagnosing misbehaving POTS modems with fear and loathing. Besides, in comparison to your ADSL modem with it’s line filters and 100’s of options, configuration of the ONT will be a just case of choosing the correct ethernet port to plug your router into.

    So if I was an ISP, I would be looking forward to lower support costs with the NBN, not higher as you seem to be claiming. Of course that’s only true for Labor’s NBN. The Lib’s current proposal will result in a hodge-podge of technologies out there and the ISP’s will have to support all of them. It will have the ISPs’ yearning for the days when all they had to ask was “what brand of ADSL Modem to you have”?

    • @Russell Stuart

      ‘ but every I know with a broadband connection needs a modem of some sort, and that modem requires power from the house. If you want to keep it running during a power outage it requires a UPS too. If you don’t maintain the batteries in that UPS you lose your power protection. Which it identical to the NBN ONT, because it is your responsibility to maintain the batteries in it too.’

      I understand all of that, I was comparing the ONT box with a two telephone socket on the wall powered from a Telstra exchange.

      ‘And I am not sure whether it is true to say the ONT is a “step up” from an ADSL modem.’

      I didn’t say it was step up from a ADSL modem, I said it was a step up from a two wire Telstra socket and my Telstra socket doesn’t come with a 20 page manual on how to use it either, hence my point that NBN support has a extra layer of complexity that the Telstra Wholesale client service does not currently have.

      ‘Even if it the ONT is technologically more complex than an ADSL or Cable modem, it is not true to say it will be harder to use.’

      Once again, I wasn’t comparing it with a ADSL or a cable modem, I also made the point about residences only using the NBN ONT’s voice port for their existing PSTN handset, so the jump for them in terms of equipment in the house and the maintenance thereof is quite a jump.

      • Hopefully it won’t be much of an issue. It’s way less complicated than the change from analogue to digital TV. Set top box options, how to plug it in, cables of various types to TV, how to set TV to video in, aerial to the set top box, need to chain aerial to VCR? Maybe need input switch box for DVD, program the channels, how does this remote work? Why do I set volume on TV and change channels on this other remote.
        I’d prefer the ONT, plug your phone in here. Ethernet in here. The rest that would have already have had to do to setup for ADSL2 anyway.

      • “I didn’t say it was step up from a ADSL modem, I said it was a step up from a two wire Telstra socket and my Telstra socket doesn’t come with a 20 page manual on how to use it either, hence my point that NBN support has a extra layer of complexity that the Telstra Wholesale client service does not currently have.”

        While I would certainly not dispute that the ONT is more complicated than a pair of copper wires, that’s only because it moves the circuitry from conversion from copper to optical into your home. That used to all be done at the exchange and SOMEONE had the (more likely) 2000 pg manual to read to fix things.

        I guess what I’m saying is, the average consumer plugs in their cable and never bothers with the ONT otherwise. If something goes wrong, rather than the bloke going out to the exchange, he remote checks the circuitry on the ONT and if there’s a problem, he comes out with a new one/new part. It’ll be just a different type of tech support than now, rather than necessarily more complex.

        You’re certainly correct though in the fact that the RSP will now have responsibility of the more complex matters, whereas it was all Telstra before. So I think that confirms it’ll be more important than ever for customer service and a differentiator for RSP’s.

  25. Good article renai but your ignoring backhaul and transit links, these will all contribute to the quality of ISPs services under the NBN.

    • That is why it is a actually a bad article. If Renai doesn’t understand what parts of an ISPs network the NBN will and will not be replacing, then he doesn’t have the knowledge to be able to make informed commentary on it.

      • You’re right, I don’t understand jack, I should just give up.

        Wait … yes, I do understand these things. I used to be a sysadmin for an ISP and I’ve covered the telco space in Australia for the best part of a decade ;)

        @Djos I’m not ignoring backhaul and transit links; can you explain to me why these things will be a differentiating factor for any of the four major ISPs (Telstra, Optus, iiNet and TPG) which currently overwhelmingly dominate Australia’s broadband market? I agree these things used to be a problem with smaller ISPs, but with today’s mega-ISPs I don’t believe they are.

        • Hi Renai for the big 4 that is a fair enuf comment, but there are still plenty of smaller region specialist ISP’s like my former employer Adam Internet, Skymesh etc.

          I know Adam has a damn good network with well dimensioned transit and backhaul links but there are other smaller ISP’s that dont have such well designed networks. Adam Internet has done a very good job of positioning itself for the future with it’s own Tier3+ Data Center which hosts half the SA Gov Server fleet and plenty of large well respected Corporate server fleets plus Adam has invested in its own Cloud Computing infrastructure in partnership with DiData/BluFire and so forth. (I sometimes wonder why I left, but that’s another story)

          Many of the smaller ISP’s also have their own Locally based Support Desks and advertise that as a benefit to their customers and point of difference from cut rate ISP’s like Dodo and even big players like Telstra etc.

          So although your article is quite correct in very broad terms, I think there is definitely still room for ISP’s to provide differentiated services in an NBN world. :-)

        • Sysadmins usually don’t know much about ISP networking, and nor are they likely to be exposed or involved in ISP network architecture, capacity planning decisions, traffic management decisions and compromises, and the costs involved in operating the network.

          The ISPs with good reputations, somewhat beyond their size, (e.g. the ones without significant amounts of performance complaints in their whirlpool forums) are managing well their ADSL backhaul, peering and Internet transit networks and links. This is indirectly also reflected in their customer satisfaction figures, as less performance problems directly results in less helpdesk calls, and consequently happier and satisfied customers.

          The NBN is only a replacement for the ADSL portion of the network. ISPs who currently get significant numbers of complaints about national and International performance will continue to get them after the NBN has been deployed unless they change their network management and performance practices. Given the current NBN CVC and AVC pricing model, the CVC will be another place where the cheaper ISPs will choose to accept rather than avoid congestion.

          If you’re going to write articles critical of politicians’ level of understanding of the NBN and the telecommunications industry then you’re also putting yourself out there for the same level of scrutiny.

          • Mark,

            you do raise some points worthy of debate, but I must warn you, the next time you attack me or my knowledge personally, I will ban you from Delimiter for a period of two weeks. Please see out commenting policy here:


            One of my points, which I have made above, but which you appear to have ignored, is that the market is currently overwhelmingly dominated by four major ISPs (Telstra, Optus, iiNet, TPG), which have strong networks; thus the performance difference between them due to network provisioning factors is likely to be minimal, as it is today.

            Can you name some ISPs which have the sorts of capacity problems you mention?



          • Hi Renai, when I worked for Adam I spoke to plenty of Network engineers in the industry who would cringe whenever TPG was brought up – the general consensus was that TPG skimp on everything they can and their network is designed for average loads with almost no headroom for peak loads (hence the regular congestion complaints on WP).

          • I think it’s a dangerous road to tread saying all those ISPs have good provisioning and are equal. Speed test can be made fast on even the slowest ISPs, especially threaded ones. Some ISPs always ask the customer to do tests on speedtest.net and the user is surprised they are way faster than the slow downloads they have been experiencing. Of all these “equal” ISPs only one has a continual stream of people posting nightly about how slow their connection is and getting lame suggestions from RSPs. RSPs who don’t seem to know the word congestion exists and would think it more likely that the neighbouts hot water heater would be the cause. I’d suggest you do some reading in those ISPs forums and identify the one that has huge evening congestion problems.

          • I have read those forum complaints ;) However, averaged speed tests do on average show TPG to be one of the fastest ISPs around; as you would expect, given its acquisition of PIPE Networks. In addition, I have received many comments from Delimiter readers stating that they find TPG’s speeds absolutely fine.

            I understand the issue of contention; but there continues to be less evidence in an NBN world than people seem to be suggesting — or even in today’s ADSL world. The evidence that does exist appears mainly to be anecdotal rather than statistical or technical.

          • Here’s some more anecdotal for you Renai ;)

            I’ve moved 5 times in 10 years. The last 3, we’ve had broadband. Iinet, every one (after the usual Telstra “we’ll give your broadband, if you go with us” routine at the first). First one, got 1.5Mbps plan. About 2.3km from exchange (by attenuation) REGULARLY got less than 750Kbps after 5pm. Any night. Friday, Saturday and Sunday, almost un-useable (2003-2007).

            Next one, bumped up to “ADSL1 Up to 8Mbps”. About 2km from exchange (same estate, different part). Never saw above 4Mbps. Almost un-useable on Friday, Saturday and Sunday nights. (2008)

            Next one, same town, different area, about 1.5km from exchange, on “up to” 8Mbps. Managed 6.5Mbps sometimes. Again, reduced to less than 3Mbps on Friday, Saturday and Sunday nights. (2009-2011)

            Current one, completely different town (and exchange), a little further from exchange, at about 2.8Km. Once iinet merged with Internode, was able to get “turbo boosting”. Gives me (I’ve seen as high as) 9Mbps. Averages 6.5Mbps. Last night? 1.7Mbps. The night DragonX docked, it was under 1Mbps (both were Friday nights, after midnight).

            Contention is EVERYWHERE in my district. The exchanges are chocked and people have horror stories of weeks to even get lines fixed, let alone complaints of slow speeds.

            I agree contention will be of much lesser impact in an NBN world, but I still believe it will be a large differentiator in price. After all, there will ALWAYS be those that are simply handed a plan, as recommended by a company, family or friend because of price, yet they don’t use even half the 100Mbps they have. That’s the people these budget RSP’s will be targeting and they can get away with high contention (low CVC that is) to shave off expenses.

          • That’s odd you had those speed issues with iiNet multiple times. I had the same issue.
            A run through my ISP history. I was a long time Netspace customer, since the dialup days.
            When ADSL first came out I went to it, then ADSL2, etc. For years I had no speed issues. Then Netspace started offering the cheapest and largest plans. Evening slowdown started, down to 6Mb on a 12Mb link, at one time it got lower, like 4Mb and Netspace said that they needed to upgrade their links and they did and it improved (one thing I liked about Netspace they were honest about their own bottlenecks). TPG cheap unlimited came along, slowly Netspace leeches moved to them and Netspace got faster, up to 8Mb of an evening.
            iiNet bought out Netspace and moved us to the iiNet Network. Evening speeds, 0.5Mb, non peak speeds, 2Mb, still with 12Mb sync. After 4 months of them not fixing this for Netspace customers most just churned to Internode.
            Internode, no evening slowdown, always get 12Mb.
            I would be interested to try TPG personally, but they don’t offer plans without contracts. Stuffed if I want to get stuck on a potentially bad connection with a 12 month contract that is very expensive to break. Another factor is I found someone with TPG on my exchange, on an TPG DSLAM. They said TPG was excellent, always got full speed. When asked about it he could always max out his torrents at a full 10Mb. Fine, I asked him to turn off the torrents for a bit and run some speed tests. Result, speedtest.net was “really fast” according to him 4Mb (I can get faster on my mobile). Ozbroadband (a single thread speed test) 0.4Mb. So really fast for one person can be a dog for another. I don’t run hundred thread bittorrents, I stream video, I download files, for me it would be classed as unusable if I was getting 0.4Mb.

          • Mmmm, TPG isn’t really any more of an option than iinet down here. We ONLY have Telstra DSLAM’s. I get a Sync rate of 8Mbps and usually don’t get more than 6.5Mbps and, as I said, less than 2 on busy nights.

            We’ve stayed with iinet, cause over the years they’ve been good to us. But frankly, if I KNEW I could get better, I’d pay and go with them. I doubt though that even Telstra would be better. The ONLY thing that’ll change the status quo down here is the NBN. Telstra have ZERO incentive to improve any lines and we’d certainly never get FTTN regardless. We only have 70 000 people total in an area the size of greater Sydney.

            I’ve spoken to iinet and they can’t really do anything. The contention is a fact of life down here.

          • Renai,

            Shouldn’t you warn Bob.H in the same way you warned Mark?
            He attacks my integrity and knowledge just as much as Mark did yours.
            And Mark at least made valid points (as you acknowledged).

            Note that I do not attack you personal, I question your opinion in your article.
            You have as editor the right to have this opinion and the fact that you allow
            comments assumes you allow critical responses to your opinion.

            Feel free to e-mail me if you do not agree with me. :)

          • Renai, congestion initially shows itself as latency, and gamers and people who use VoIP are the first to notice.

            Compare these two searches for Whirlpool threads titled ‘games’ in the TPG and Internode forums, which should pick up people having latency problems –

            TPG search

            Internode search

            This is further supported by the “How do gamers rate the connection quality?” question in the Whirlpool 2011 survey (ignore the incorrect year in URL)

            Clearly TPG aren’t running their network well compared to Internode – Internode are using the same Telstra copper to subscribers homes as TPG are, so the elements to address these complaints are in TGP and Internode’s hands. If the only cause of the gaming problems was the copper/ADSL network, then there’d be proportionally equal numbers of complaints from gamers in both the TPG and Internode forums.

  26. Renai,
    Your proposed price point and quota seems unrealistic with current NBN Co pricing.

    Taking the upper end $60 figure, subtracting $6 for the GST and $38 for a 100/40 AVC you have $16 left for CVC which will get you 0.8Mbps of CVC per user or a 125:1 contention ratio (and this is ignoring the NNI cost and all RSP costs).

    0.8Mbps is 0.1MB/s which when multiplied by the 60*60*24*31seconds per month gives you 268GB/month/user.

    Your proposed 1TB quota plan would only have enough CVC to allow the average user to download 268GB/month and that is only if the CVC is completely saturated 24/7 (i.e. the network is constantly congested).

    Please correct me if I have stuffed up the calculations o but if my figures above are right your suggested $60 price point seems to be way too optimistic. On the figures above I’d think a bare bones absolute minimum price would be $80/month and $100/month would be more realistic.

    Can you provide a breakdown of the NBN Co products that would be used to provide your proposed $60/month 100Mbps service?

  27. Correct me if I’m wrong Renai, but you appear to make two arguments under the NBN:

    1) Network quality will be equal.

    2) There is no advantage in local hosting.

    Both of those seem to be flawed from what I’ve seen.

    1) NBNco only provides the ‘last mile’ to the PoI – and even then you have CVC to differentiate providers. Routing, backhaul etc. all differentiate the product.

    2) There are a whole raft of services under of the banner of ‘cloud computing’ that may benefit from reduced latency, as well as the old culprits such as gaming. I have a better ping with someone 2000Km from me than I do with someone across the road.

    3) There are many possible avenues for ‘bundling’ with other products to help differentiation, as well as customer service, branding so on and so forth.

  28. With ADSL the speed you get is determined by the copper from the POI (exchange) to you. But you get that speed all day, every day (backhaul and faults permitting). With the NBN the speed you get is determined by the CVC capacity purchased by your ISP. It will vary from ISP to ISP and it will vary by time of day. A 12Mbps service from a premium ISP might be 12Mbps all day, every day. A 12Mbps service from a cheaper ISP might only be 4-6Mbps at peak times. That might be a significant difference if you’re planning to stream video at the same time as using your connection for other purposes.

    Streaming video services offered independent of ISPs is a bit confusing. If the content provider is planning to deliver the stream over an internet service then they are dependant on the ISP having sufficient CVC capacity for this to work. If they are planning to multicast then they will need a ‘data’ AVC as well as the multicast AVC which makes the connection a bit expensive ($5pm + $24pm + GST). Perhaps you could do an article on how video services (VoD, ipTV) will be delivered with the NBNCo’s product set.

    The way the NBNCo has bundled the wholesale access services the end customer is effectively forced to get their voice, internet and ipTV services from their ISP. The alternative is to buy (via their RSPs) multiple AVCs at $24pm + GST. The way ISPs can offer these bundles is going to be a differentiating factor.

    Then there’s price.

  29. This debate sounds like the state school vs independent school debate. Unless you experience both, you don’t know if you are getting value for money by going for “quality”.

  30. yoghurt? You’ve put me off telework for life, Renai! :)

    Seriously, it’s probably going to boil down to price for most people. Which scares me a little because you can see what Coles and Woolies are going to do.

    However, what I think this means is that the higher end ISPs are going to have to actually start educating people about the quality of the connection. Yes, there isn’t a huge difference in provisioning among ADSL providers – unless you end up with the likes of Dodo, but that’s DSL. It’s self limiting to some extent.

    Once you have a lot of people out there using 50Mbps+, the differences in provisioning of backhaul are going to become much more obvious. I think there will become more advertising around this issue, rather than just raw speed or even quota.

    If the argument were entirely true that its all much of a muchness, then you have an argument for the NBN being Layer 3 and RSPs simply being marketing and support companies. Instead we have a network that allows differentiation not just on service but actual performance and I suspect that for some this will matter.

  31. I don’t think the premise about local servers not being necessary in relation to latency makes much sense. If I can ping a local server in 32ms and an overseas one 332ms currently, there is not going to be a substantial improvement from NBN fibre. It will still likely take 300ms more to ping that same overseas server going over the same international link (and the speed of electromagnetic waves is not likely improve much :-) ).

  32. I think AFL/NRL will play a big part in which ISP you pick. Telstra will attempt to pay big money for the footy rights and then you must go with them. At least the AFL are looking at running there own channel through over NBN.

    • If content owners are smart they do not sell any broadcast rights to ISP’s as a commercial model in which the content owner can charge the viewer directly will deliver more income for the content owner. Why sell your rights to be viewed by the end-users of one ISP, when you can have access to all end-users from all providers? Plus there is the added benefit of keeping control over their own content.
      For the end user it will become important if his ISP has fast connectivity to the content provider. This is one item on which ISP’s will differentiate significantly. Not the total bandwidth of the ISP’s backbone will be important, the “how many” will be more signifcant.

      • I want to know why the morons in the movie & Tv content industries cant make all their content available to Netflix / quickflix in the same way that the Music industry can provide MOG /Spotify with 16 million songs for their catalogue and spotify claim to add 10,000 new track a day!!

        Seriously MPAA and AFACT, pull your finger out and stop crying poor over piracy, you idiots are the cause of your own problem!!!

        • Tend to agree there with you. I believe that the record and movie industry never really understood the concept of creating income out of the possibilities Internet gives them. It is from this aspect very interesting to see what is going to happen on the NBN. Are they going to take own control? Or do they maintain on the road of using ISP’s. An even more interesting question will be what the artists will do. Spotify has been already in the news for creating hardly any income for artists. Why using record labels, when you got the largest publish tool available in your own home/office? Will be an interesting time to see what is going to happen on content offerings.

  33. My last ISP was TPG no way in hell I would join up with them again even for the NBN…

    • I’m on TPG here, you insensitive clod! Been with them essentially since about 1998 starting with Niterider (1am-8am free dialup).

  34. How will the “low price” ISPs be able to offer lower prices when they have to sell the same product that they have purchased at the same price from the same wholesaler? Price differentiation won’t become what people buy on, it will cease to exist. The low-cost ISPs will simply be driven out of business by having lost the one thing they have to compete against the biggies with. Oh, sure, in the initial phases of the NBN they’ll offer low prices, by living off their their ADSL profits and losing money. But once they lose those two, once they run out of money, there’ll be a market rationalisation that largely only leaves us with the big operators who currently charge the highest prices.

    • If you have a read of the comments Gordan, there are numerous ways those cut-price RSP’s can save money and offer lower prices. Purchasing less CVC and allowing throttling that way to limit usage, while keeping prices down is one way.

      Cheap RSP’s are not going to disappear. They will just offer different types of services than now, for a lower price, under the NBN.

      • …there are numerous ways those cut-price RSP’s can save money and offer lower prices. Purchasing less CVC and allowing throttling that way to limit usage, while keeping prices down is one way.

        Does the ACCC have a position on the service being advertised and practices that prevent a customer from getting what they pay for? I do recall an ACCC rumbling but my searches have been fruitless.

        i.e. X RSP advertises their NBN offering at 100mbps. I sign up and the 100mbps service is provisioned. In theory I have a 100mbps but the reality is the best I can hope for is half that because they skimped on the CVC and are throttling.

        I don’t envision this being an issue for larger ISP’s/RSP’s but the NBN is supposed to set us free of household internet contention issues that ADSL services impose on us. For this issue to then resurface at the RSP level would be a negative that the peanut gallery could latch on to for the purpose of discrediting the NBN irrespective of the fact that it has nothing to do with the NBN itself.

        • It will be the NBN equivalent of “up to” the advertised speed. With ADSL2 the “up to” component is limited by distance, under NBN the “up to” component will be limited by traffic. You’ll get 100Mps at 3 in the morning when there are 2 other people online, but at 7pm when every man, their dog, and the dogs fleas are online, you’re going to have a traffic jam.

          Think of the NBN speed as a speed limit. Under normal circumstances, you can drive along at that speed limit, and get the best of the road. When there are too many other cars on the road, the average speed slows down, even though everyone can potentially drive at a faster speed.

  35. They won’t be the same and here is why. NBN is an upgrade to the access network, not the methods which get it there.

    We have 2 primary issues at the moment. The first is the international links which has already been pointed out.

    The second is that consumer broadband works on a contention basis – and contention is already occurring quite often with ADSL. The NBN will solve the access speed side of the equation – however you are looking initially at 100Mb speeds – approximately 10x higher than the current average to any residence (actually probably even more than that).

    So the end story is, that the backhaul networks to the points of interconnect and the core networks will make a difference – we already see that and that is with ADSL. With exception to a few RIMs and the like, they already have fibre to the nodes there.

    You have made the argument in all your posts that the limiting factor is the copper network – while true in many cases that is not an absolute – and ignores the architectural issues behind designing a contention based network.

    The backhaul networks won’t scale the same amount as the access networks so you are likely to see contention ratios significantly rise (and even more when 1Gbps access plans are available).

    Does it make NBN a bad idea or irrelevant? Not at all – however ftth in Australia isn’t the panacea that is claimed. So often you see it compared to what happens in Asia. You have to remember that with the population density in asia, that allows them to mesh their networks with multiple routes to each node to handle the varying loads. In Australia, we quite often have 1 and maybe 2 paths to each node so rerouting around congested nodes just isn’t possible in the same way it is there.

    If anything, under NBN the core networks which the ISPs have invested in will become even more of a differentiator, not less. Also, while NBN will fix some of the current copper issues we face, I don’t think it is as simple or virtuous as you have made out. There are still a lot of practical issues that need to be worked through. While ftth in Aus doesn’t have the problems of copper, it has its own unique issues.

    As I said, that doesn’t make NBN a bad idea – the concept is still good and having a top rate access network is a great idea (although the price is an issue – but that is another argument).

    I tend to agree with you about content being irrelevant though.

  36. Theoretically, it makes sense to say that content and bundling of services, video, entertainment etc. will not work on NBN. However, current practice suggests otherwise: Telstra manages to sell phone/Foxtel/T-Box bundles *with* their “superfast” cable internet plans. Whatever the reasons are for people getting those bundles at a premium price, NBN won’t change that.

  37. As a director of a small local ISP providing NBN plans I wonder where “we” as small ISP’s fit into the story. In the majority of the IT (online) magazines “we” are not even mentioned when actual some of them offer better terms and conditions, low amount of TIO complaints in comparison to their customer database plus deliver better value for money. Net neutrality becomes quick fast a hot topic in America and Europe. Which of the ISP’s is shaping their traffic without disclosure to the end-user by packet discover? Becomes pretty important when cruising at 100Mbps. So dear writer where do we fit in? My experience is currently that large ISP’s currently struggling with delivering quality local customer support. And that this is actual what end-users start to expect when buying top speed plans on the “perfect” NBN. This article is extremely shortsighted both in market as product analyses. Please do better plan research on whirlpool and try again. Seriously. We build in the meanwhile a huge leads database of people interested in our services. Problem is the poor address data supplied by the NBNCo plus an extremely slow roll-out that looks more patched than a patchwork blanket. From the 50 enquiries each day, we have heaps of problems locating them and effectively we only have 1 sale a week. Did somebody actual think for some brief moment with common sense why the uptake is so low? The NBN will be remembered as a “perfect” network, but extremely poor marketed. 50 of those enquiries will not be connected in the next two year. Showing ad’s on the TV has this effect, but question is, how effective is this. And ISP’s, big and small, pay the price each day by waisting valuable staff hours on telling them they cannot get it. At the moment it is completely unclear what is marketed, where and when to whom. And to me that sounds like poor marketing …. was better the NBN started to fix up their databases, pushed their rollouts according to planning date and stop marketing people for political reasons in areas that will not see NBN for several years. My two cents.

    • Yes I would say it was bloody poor marketing to not even choose to give a web site when you posted your comments.

      In fact I am so astounded I am wondering if you really are the director of a small ISP.. Incidentally I think that the term fro retail providers for the NBN is RSP. which just compounds my suspicions.

      • Believe can be found in Church … unfortunately your post says more about your state of mind then about my comment. If you followed my suggestion and took the effort to go to Whirlpool, you will find many many NBN-plans from other ISP’s (including mine), which are just as much worth mentioning in the article. However if the writer would have done this, it would undermined the preset opinion of the writer as it would have shown differentiation. And for this reason it is left out. Your negative comment does not contribute in any manner to the discussion at hand. Also it is not poor marketing on our part, not to provide a link. We prefer to be committed to end users who are committed to us, rather than having the risk that we pickup a end user which starts to question our services on the basis of the use of a abbreviation.

  38. Renai, if you’re right then why has Telstra paid a small fortune for the online AFL rights? I think sports will be a major differentiation between RSPs in the NBN world.

    Australia is a melting pot of people from all over the world, so I think there’s a place to make it a lot easier to gain access to content from overseas. My housemates are thai and I had been thinking of moving over to TPG to try out their IPTV service. As with everything these days, there’s an AP for that and they both watch TV shows via iPad / Apple TV or Android tablet. Still, it seems the range is limited. An RSP who can provide a broad range of content from certain countries would probably have a competitive advantage, especially if it’s exclusive.

    • Currently Jeffrey, Telstra have not indicated at ALL that they will be moving to an NBN distribution platform.

      They will keep their HFC (and satellite) Foxtel customers once the NBN has removed their internet HFC customers and as yet, we’ve yet to see whether they will offer anything in the way of NBN multicasting. The current AFL agreement is for 4 years, for the disgustingly high sum of $1 Billion, but that won’t really be affected by the NBN, which, even if it is continued at pace, will not be even half done by then.

      IPTV, IMO, will go somewhat the way of broadcast TV- extinct. When we have VOD for subscription amounts each month, people won’t watch broadcast TV/multicast IPTV, because it’s a waste of time. Yes, it’ll be 20 or 25 years before we see it, but TV, as we know it now, both broadcast and IPTV, I believe, will go the way of betamax….

    • @Jeffrey

      I see Telstra’s licensing of the AFL rights online as being fairly pointless — wishful thinking, in short, that they can make substantial revenue from online streaming. I don’t really see the point in this, however; when the AFL is generally broadcast over free to air TV, which has more channels available for this than ever.


      • > I see Telstra’s licensing of the AFL rights online as being fairly pointless — wishful thinking, in short, that they can make substantial revenue from online streaming. I don’t really see the point in this, however; when the AFL is generally broadcast over free to air TV, which has more channels available for this than ever.

        I think it is a protective play for FoxTel. Each week there are 9 AFL games played. Only 4 are broadcast over free to air. If you are an AFL fan, the only way to watch the majority of the games live is with a FoxTel subscription to the Fox Sports AFL channel. For many sport is the only compelling reason for a cable TV subscription.

Comments are closed.