Rethinking the NBN: Hackett’s just getting started

140

simon-hackett

news Internode founder Simon Hackett has declared that the innovative ideas he has put forward over the past several years with respect to reforming the NBN are “just the tip of the iceberg” in terms of potential improvements and cost reductions to the project, as debate continues about its future under the new Coalition Government.

Up until last week, the Coalition’s NBN policy, put together in Opposition, focused on using a predominantly fibre to the node rollout style for the project, as opposed to the more ambitious and technically superior fibre to the premises approach preferred by the previous Labor Government. However, last week new Communications Minister Malcolm Turnbull appeared to drastically modify the policy, declaring the Coalition was not wedded to its FTTN model and was “thoroughly open-minded” about the technology to be used in the network.

Many of the Coalition’s choices about the future of the NBN, Turnbull intimated in a press conference, would be guided by a strategic review which NBN Co was to conduct into its operations within 60 days after its new board its appointed. The review is to set to estimate the cost and time to complete the NBN under its current model, as well as evaluating how other models could potentially reduce that cost and time to complete the rollout.

Telecommunications analysts such as Paul Budde have immediately interpreted Turnbull’s policy backdown as the Minister having given NBN Co “the opportunity to save the current NBN”.

In the wake of Turnbull’s comments, a number of commentators, including the writer of this article, have highlighted the fact that Internode founder Simon Hackett has constantly publicly over the past several years canvassed a number of improvements which could be made to the NBN rollout to make it both more cost-efficient, as well as improving the competitive outcomes to be stimulated by the rollout.

Hackett accurately predicted that the large number of points of interconnect model mandated by the Australian Competition and Consumer Commission for the NBN would lead to a rapid consolidation of the ISP industry; he predicted moves by smaller ISPs such as TPG to deploy their own fibre if the Coalition won power and modified the NBN rollout; and he predicted the need for ownership of Telstra’s copper cable to be transferred into the NBN Co; an issue currently being discussed by the two telcos.

And in July this year, Hackett outlined a wide-ranging series of changes which NBN Co could make to its network rollout, including using a drastically simplified network termination device and reviewing the design of its network rollout, that would deliver, as the Internode founder said in a speech at the time, “fibre outcomes on a copper budget”.

Because of Hackett’s ability to predict the dynamics of the NBN’s impact on Australia’s telecommunications ability, his extensive experience in the sector and his innovative ideas regarding the project, Delimiter has proposed the executive as a candidate to be appointed to NBN Co’s board (subscriber content).

In a series of new comments posted on broadband forum Whirlpool last week, Hackett said that his comments regarding the NBN so far represented only a start.

“Its also worth appreciating that while I put a lot into around 20 minutes in that talk,” Hackett said of his July speech, “that talk is just the tip of the iceberg in terms of things that can be done. I could have talked all day – but I only had a 20 minute slot … that talk is the tip of the iceberg.”

Hackett said one of the benefits of the Coalition’s “reset” of the NBN, “can (and should) be to lay out every single source cost in building the NBN, and every single timeframe element required in the build process, and to aim to apply lateral thinking to removing as many of them as can be removed while still delivering a working network, and optimising the hell out of the rest.”

The Internode founder said that in particular, he envisaged NBN Co’s fibre rollout process should be massively simplified to be “an exercise in purely rolling out fibre”, with much of the rest of the complexity of the rollout, such as the network termination units sitting in customers’ premises (and associated battery backup equipment), being shifted away from NBN Co and to retail service providers. The NTU cost is believed to make up a substantial chunk of the approximately $2,400 it is costing NBN Co on average to deploy fibre to customers’ premises.

“It’s about maximising competition and hence competition driven consumer benefit outside of the one part that needs NBN Co – the replacement of copper with fibre in the streets,” wrote Hackett on Whirlpool.

“It’s much more about collapsing the NBN Co fibre installation to a single pass, single truck roll exercise – and the logistics and labour saving of that, versus a process that typically requires NBNCo to undertake multiple truck rolls today. Its about enhancing end user site competition (for the benefit of consumers), not reducing it.”

“Its about taking a fresh look at the FTTH rollout design and being lateral about maximising every opportunity to simplify the rollout and focus on the fundamental task – getting the glass in the ground, cost-effectively and time-efficiently.”

Hackett’s comments on the issue reflect the fact that in the current broadband environment, retail ISPs such as iiNet (which owns the company Hackett founded, Internode) have taken over responsibility for virtually all areas of network management when it comes to providing broadband over Telstra’s copper cables, apart from management of the copper cables themselves.

From the customers’ connections, to equipment in telephone exchanges, to backhaul infrastructure and more, retail ISPs such as iiNet have taken over management from Telstra over the past decade when it comes to their own customers, with Telstra largely getting involved in the situation when the underlying copper cables suffer faults. It appears as though Hackett envisages a similar approach from NBN Co, rather than the current situation, where NBN Co is managing many aspects of customers’ connections, including the equipment on customers’ premises and other factors such as standard speed tiers.

opinion/analysis
As I wrote on Delimiter 2.0 last week, the Government and NBN Co need to start listening to Simon Hackett’s suggestions when it comes to the NBN. The Internode founder has proven himself to be better at predicting the big ‘macro’ outcomes from the NBN’s specific rollout approach than either NBN Co or the Government. Why is he able to do this? Because this is a broadband industry which Hackett, with the help of others such as iiNet’s Michael Malone, Netspace’s Stuart Marburg, TPG’s David Teoh and others, virtually created, in the past two decades. Is it any surprise Hackett is an expert?

Image credit: Internode

140 COMMENTS

  1. The trouble with what he’s said was that the NTD shouldn’t cost as much as he’s inferring, and really it’s one of the best things about the current rollout – the ability to have multiple connections and not be locked to one RSP. If you have an option to upgrade or intially choose to have the NTD still, I’d be more for Simons idea, but removing it actually removes competition and decreases the benefits of the network.

    I believe 7T did a fairly thorough analysis on this which would be good to see here again.

    • From what I have been told, the NTD cost is part of the FTTP customer connect cost/demand drop cost, detailed here as currently averaging about $1,100:

      http://delimiter.com.au/2013/09/27/leak-shows-coalitions-nbn-costings-wrong/

      If it costs $100, then it’s not a big deal. But personally, I suspect Hackett’s right and the things are costing a great deal more. Alcatel-Lucent isn’t Billion or Netgear, and Hackett’s right, there’s a lot of redundancy built in here. I wouldn’t be surprised if the NTUs were costing $500 or $600 — way more than any retail customer would themselves want to pay for something that should just be a gateway to the fibre, as normal ADSL models are to the copper.

      • @Renai

        I don’t see how it’s possible for the NTD to be costing that much. Which is why my argument against removing it makes sense. If it does, then it definitely has merit. But as far as I understand, Alcatel’s contract with NBNCo. amounts to about $1.5 billion for GPON equipment. One has to assume that includes the NTD. In that case, there’s no way the NTDs could cost any more than $100-150 each.

        It also makes sense. The NTD is not a complicated piece of kit. It is a fibre converter, a VLAN tag reader and a POTS emulator. You can buy them for $100 retail in the US. And bulk from AL they’d be a lot cheaper.

        I don’t know where people get this idea they’re enterprise level switches just because they’re from AL?….it’s CPE with a lifespan increase.

        • http://delimiter.com.au/2010/06/24/alcatel-lucent-wins-up-1-5-billion-nbn-contract/

          “Other GPON equipment suppliers will also be engaged once NBN Co achieves “full rollout scale”, according to the executive. “The contract with Alcatel-Lucent allows NBN Co to purchase what it needs when it needs it, and allows NBN Co to engage other suppliers at a later date of our choosing,” he said.”

          I’m sorry, but you can’t extrapolate per-NTU numbers from the Alcatel-Lucent contract when NBN Co acknowledged that the contract is the not full extent of the network equipment it will need. In addition, the actual break-up of the different network components from that contract was never published.

          “One has to assume that includes the NTD. In that case, there’s no way the NTDs could cost any more than $100-150 each. It also makes sense. The NTD is not a complicated piece of kit. It is a fibre converter, a VLAN tag reader and a POTS emulator. You can buy them for $100 retail in the US. And bulk from AL they’d be a lot cheaper.”

          Hackett’s point is that the NTU is a lot more complex than that and a lot more complex than it needs to be, and I agree. In addition, I have to say, mate, that there is just no way Alcatel-Lucent is charging NBN Co $150 for a device this complex. I’m sorry, but that is just not possible. That’s the same price you’d pay for a Billion ADSL router. The NBN Co NTU is more complex than that, and it comes from Alcatel-Lucent, who are known for paying through the nose.

          • Well, there is the NTD, a wall plate for the fibre cable (my installation doesn’t have this) and the power supply/battery unit. Replace the battery with a wall wart and there is not much else left to remove.

          • The the “how do i use my phone when the power goes out” brigade and the “I need a POTs phone” brigade had a bit to do with that.

            I do partially agree with Hackett on the this the NTU is overly complicated for most installs. It should really be no different to what we have now with POTS connections, where you can pay to have extra lines installed, the difference being you’re only paying for a NTU upgrade, a significant saving.

            I think another contributing factor to the choice of NTU would have been Telstra given how many millions of connection would have an existing POTS service with Telstra along side an ADSL service with another provider. Users faced with prospect of having to move either their the phone service or internet service to a new provider on a single connection NTU their would be large number who would just drop the phone services costing Telstra billions of dollars of revenue every year.

          • Renai,

            I work for an FTTH equipment vendor (not alcatel).

            With it being near impossible to find a bloody datasheet on this ONT, I cant say for certain what the costs /should/ be if these guys were running in a competitive environment, however what I can tell you is that for a 4 port, 100mbps GPON NTD, with 2 pots ports, for a rollout this large you are looking at around USD150/ea.

            This doesnt include UPS costs.

            If they are charging 600 per, thats just crazy, noone in their right mind would buy something worth that much.

          • Thanks for that info Charlie, that’s very useful :-)

            Can I ask what that (or a similar) device would cost with gigabit ethernet ports?

          • What are the costs exactly?

            Is it the cost charged to NBN Co’ to receive each NTU in a warehouse somewhere?

            Is it the cost to receive the NTU, store it until it is needed, transport it to the required location, obtain permission to install it and install the device?

            At which point are these costs being calculated up to as they will be very different figures.

          • Why would they be storing them in bulk in a centrally located warehouse? They would use just-in-time delivery as much as possible throughout the project. I think you’re trying to fabricate add-on costs that don’t exist – deployment is covered by labour costs and other transportation costs are factored into end user connection costs. There’s no way you can manipulate those figures to somehow inflate the cost of the NTDs.

          • You are missing the point.

            The price difference between an NTU delivered to NBN Co’ and an NTU installed will be different.

            They will be different costs.

            Which one are people quoting and comparing?

            There seems to be a lot of confusion. They are looking at it as a proportion of the total installation cost and comparing it to how much each one costs to purchase. Apples to oranges.

          • The price difference between an NTU delivered to NBN Co’ and an NTU installed will be different.

            The price difference between a “delivered” NTU and an “installed” NTU is irrelevant, the installation cost will be the same regardless of the NTU used, the important part is does the NTU offer the best value for the given role…

          • To consider whether it is worth reducing the cost of the rollout by outsourcing it to RSP’s by having them install NTU’s.

            You claim it is only worth looking at whether they get the job done? Not the cost of an NTU or how much it will cost to install an NTU per premise on average?

            I’m lost.

          • No, I’m saying the cost of installation is the same, regardless of what the NTU costs. I thought it was obvious (as thats what I said), guess not…

            I’m lost.

            I know…

          • I understood that point tinman,

            I just have no idea as to its relevance to my original question.

            Why are people comparing vendor and installed costs for NTU’s?

            I’m talking about averages so your divergance into types of NTU’s is a complete sidetracks which makes no difference.

            Thats why I said I was lost, I assumed you meant it to be relevant but judging by your reply it was not…

          • I’ll try rewording it.

            The install cost for either hardware X or Y, will be the same, so from a mathematical point of view, you can remove the install cost from the equation.

            When you say
            The price difference between an NTU delivered to NBN Co’ and an NTU installed will be different.

            They will be different costs.

            it’s a true statement, but worded in a way that makes it sound like the installation cost is a factor between different makes of NTU, which it isn’t (no matter the model, they will both incur the same “bolt to the wall” installation charge).

          • You cannot compare the retail, single unit cost of one device with the wholesale volume purchase (in the millions of units) cost of another. That’s ridiculous.

            I don’t disagree that the NTD is more complex than a single port modem, but so is a TP-Link 802.11AC wifi router with five gigabit ports which also costs around $150 (from memory). The fact is we don’t know how much it costs, so Seven_tech’s speculation is just as valid as yours at this point.

          • @Renai

            The AL contract is for ‘up to’ $1.5 billion. The NTDs are AL. Therefore, the NTDs must cost less than $1.5 billion total. Unless I’m misssing something? I know other GPON gear is being used. But we know the NTDs are AL.

            Secondly, on the rest of the gear, that’s now not true. It is optional. The reason they were doing it that way was because the government mandated it. It’s no longer a mandate now, so they’re not doing it. That’s a direct and major increase in efficiency right there.

            I don’t really understand the argument about it being more complicated than a Billion router. It’s not. It’s a managed switch with POTS emulation and fibre conversion. You can buy 16 channel enterprise versions for $300. (without the POTS emulation, granted). Ans that’s retail.

            There’s anecdotal evidence Telstra’s fibre NTD is only about $150 for customer replacement.

            Look, I agree we should know about the true cost. There’s no argument from me there. But loosing the NTD is a considerable support cost increase and a standardisation issue. How much money & time is wasted on DSL because so many routers are different.

            As Trevor has said and I’ve argued before- the direct monetary cost is not the only function at play here.

          • > Look, I agree we should know about the true cost. There’s no argument from me there. But loosing the NTD is a considerable support cost increase and a standardisation issue. How much money & time is wasted on DSL because so many routers are different.

            This is only a minor issue because (almost) every home is going to need to supply a router in addition to the NTD. Sure a support person will be able to ask the customer to connect a computer directly to the NTD but they may not even have a computer with a network port.

          • @Matthew

            Nope. That’s not what I’m talking about. I’m talking about the actual fibre. An NTD has diagnostics built in to allow NBNCo. to test the integrity of the path, its’ identity, its’ route etc. A router cannot do any of those things.

          • > Nope. That’s not what I’m talking about. I’m talking about the actual fibre. An NTD has diagnostics built in to allow NBNCo. to test the integrity of the path, its’ identity, its’ route etc. A router cannot do any of those things.

            Are you 100% certain of that? What is to prevent a router with fibre connection performing the same diagnostics? My ADSL modem can give some diagnostics about the copper line.

          • With the cost of increasing the complexity of the install. The advantage of having an independent NTU is that you can avoid problems in user installations, thus reduce support costs and complexity, and make the broadband service to the RSP a “black box” problem.

            How many line issues on your ADSL2+ line have you had to diagnose where the technician could not reliably determine if the problem was with your modem, the internal wiring of the house, or somewhere else in the network?

          • “My ADSL modem can give some diagnostics about the copper line.”

            But your ADSL modem doesn’t have 31 other connections on that line…if it stuffs up, it’s only your connection that is affected. The same is not true with an NTD.
            What is important is that with the NTD, NBNCo can reduce maintenance and diagnostics to only themselves instead of having to figure out if it’s their problem or not. The savings on support might actually pay for the NTD outright…

          • I am not sure what makes you think that the NTU is a lot more complex than an ADSL router. The NTU isn’t doing anything more than a decent router is doing, and compared to one of the handful of GPON VoIP routers around (eg Billion 9300VNX), it’s doing a lot less. GPON routers aren’t widely available, either, so they won’t be super-competitively priced.

            I would be suprised if an NTU + regular Wireless router cost a lot more than a GPON router from the likes of Billion / TP-Link / etc. Hell, a router isn’t even required if a single device is to access the NBN.

            Also, people might complain about the number of boxes on their wall. Fibre outlet: can’t avoid it. UPS: perhaps it should be optional. NTU: well, it’s either that and a router – or a GPON router. Big deal.

            I’d like to see evidence of the NTUs’ exhorbitant pricing. How about you put out your journalistic feelers to TransACT/iiNet or Opticomm and see what they’re paying for theirs.

            Lastly, a common demarcation device on a homogenous network is a very good idea for support. If it saves an average of, say, $100 in support (easily done), then add that to the extra price difference of a GPON router and let’s see how that stacks up.

          • I was buying xDSL modems wholesale 10 years ago for $70. While these NTU’s have to be far more robust than the cheap gear designed to last 2 years we were buying at the time, if NBN is paying over $120 in todays money for something that can built with a couple of $3 asics and a power supply then they are paying far too much.

        • The NTD being used by the NBN does cost that much: typical pricing for a commercial grade optical termination with residential facilities such as data and voice ports is in the order of $500-700.

          This is the same difference in price between, for example, a consumer-grade 3G/4G gateway with wifi and Ethernet over the commercial grade device with exactly the same capability. If all you want is an off-the-shelf solution that is good enough for today’s usage in your home, you can buy the $100-ballpark solution from Virgin Mobile or Telstra. If you want a high grade solution that ties into your gigabit Ethernet and offers better antennas and minimal latency over the transition from 4G to Ethernet, you’ll pay for the $1000-ballpark rackmount solution which still uses only one provider.

          The Alcatel-Lucent box being supplied as part of the NBNCo installation is a commercial grade option. It’s intended to be in service for decades and has a UPS attached.

          That you can’t see why the box would cost that much indicates some bias in your thinking.

          • For sakes God! My in NBN Co asks me why this discussion persist.
            a. The Chinese Huawei bid did for NBN Gpon. It was winner second supplier. Alcaluc premier one. Who knows what governments did.
            b. NBN get world best price. Alcaluc margin fraction.
            c. You think vendor prefer complex?
            d. NTD just one particle of Alcaluc supply contracting.

          • @Grail

            The NTD has a design life of 7 years max. Not decades. NBNCo. have stated that on multiple occasions. The NTD is designed to be replaced a few years after it is installed when XGPON is needed.

            The NTD is not an enterprise solution. It is CPE.

          • “The NTD has a design life of 7 years max. Not decades. NBNCo. have stated that on multiple occasions. The NTD is designed to be replaced a few years after it is installed when XGPON is needed. The NTD is not an enterprise solution. It is CPE.”

            hey seven_tech,

            can you please provide references for your seven years’ claim?

            Cheers,

            Renai

          • Sure Renai

            Page 110 Draft CP. ‘typical design life between 6-8 year’s.

            It was also a question asked during a Senate Estimates or Committee hearing but I can’t remember which one so I will attempt to find it.

          • Interesting. I would make 2 points in relation to this.

            1. Consumer-grade ADSL routers typically give up the ghost substantially before this point. With the exception of some premium brands such as Fritz!Box, usually your average Netgear or Billion router will die after 3-4 years of operation. This indicates, once again, that the Alcatel-Lucent gear is being designed for the long-term and is a substantially different calibre of equipment than the consumer-grade CPE people are comparing it to on this thread. This is just not the same calibre of cheap-ass equipment — consistent with Alcatel-Lucent’s nature as a high-end networking tech vendor which usually supplies to telcos and enterprises and not consumers.

            2. Once a piece of tech gear lasts 7-8 years, it will typically last forever, or until it’s made obsolete. I don’t expect the majority of NBN Co-supplied NTUs to be replaced until at least a decade after they’re deployed. Most people won’t need 1Gbps in that timeframe.

            This gear is being built and put in to last for a decade — it’s just not the same as a $150 Billion router from Harvey Norman. I’ve personally handled the NTUs … they are just nothing like consumer-grade routers.

          • @Renai

            I’d make a couple of comments back:

            1- I agree consumer gear is designed to be replaced in a shorter time period than 7-8 years- primarily it’s planned obsolescence to generate return sales & to limit support requirements. It has little to do with actual lifetime of the components. Saying that, I’ve only recently replaced our original iinet Belkin ADSL/VOIP router after 6 years of use. A colleague is still using it 7 years after it was made. I also have 2 other routers that have been going more than 5 years. One is a TP-Link. One is a Netgear.

            2- My company builds IT & automation equipment. Our industrial grade systems are designed with a 7 year replacement life. Some have lasted for almost 20 years. They are primarily made of Chinese off the shelf components + a few purpose built Circuits. Our power supplies have lasted 25 years in some cases and the equipment they run still works. They cost us dollars to put together. The R&D is what costs the money.

            3- I have seen the inside of the original (before integrated power supply) NTD. It has 2 CBs, about 8 transceivers, a primary & secondary CPU, a bunch of logic circuits and an optical converter. We have gear we make that has more on it and costs about $75 to make. I’m not saying therefore it must cost this. I’m saying from my experience, the only expensive part is the programming in the chips and CPUs. The rest is mostly off the shelf stuff that has a surprising lifetime.

            4- The Customer Connect costs are $1100 according to NBNCo. If we assume a (primary) contractor price of $60-80 an hour, 2 men for 2 hours for the drop and 1 man 4 hours for the internal install, that’s between $480 and $640 just for labour. That leaves $500 for all the other expenses, including PCD, cable, power supply, NTD & any analogous testing time added. I don’t see how it’s possible the NTD could be $500-600 and again, the AL contract is for the entirety of the rollout and only up to $1.5 billion including, presumably, OLT cards as well. I cannot see how the NTD could cost anymore than a maximum of $200 from the evidence we have.

          • Dude,

            some of your other points are valid. But let me say this ONCE AGAIN.

            NBN Co explicitly said when it signed the AL contract that it may also sign OTHER VENDORS:

            http://delimiter.com.au/2010/06/24/alcatel-lucent-wins-up-1-5-billion-nbn-contract/

            “The contract with Alcatel-Lucent allows NBN Co to purchase what it needs when it needs it, and allows NBN Co to engage other suppliers at a later date of our choosing,” [said Kevin Brown]

            You simply cannot extrapolate *anything* from that $1.5 billion contract.

            Secondly, I’m not basing my estimates on thin air. Both Hackett and my own sources consider the NTU to be a ‘substantial’ portion of the customer connect cost. It’s just not going to be a $100 piece of equipment, or else nobody would be focused on the issue. Is it $500 to $600? I think, given what I have been told, that this is not an unreasonable estimate. I would expect, say, a min of $200 and a max of $600. It’s not a cheap piece of gear.

            A lot of people on Delimiter on this article thread are obsessed with comparing your garden variety piece of shit Billion/Netgear router, manufactured in China, with this Alcatel-Lucent gear. I just can’t understand this. This is not the same calibre of equipment, and that’s a fact. Alcatel-Lucent does not sell equipment priced at $100 or $150, even at wholesale prices. They just don’t.

            Sometimes this is like hitting your head against a brick wall. I can tell you guys, but you just don’t listen. You keep on making ridiculous comparison between enterprise-grade gear and consumer CPE, like it has some relevancy.

          • Renai,

            The point about the AL contract is that AL make the NTD. No one else. The NTD is a custom piece of hardware, hence why we’re talking about its’ cost. I don’t understand why you believe NBNCo. can just change NTD whenever they want? Hackett even uses the opposite of that argument when saying getting rid of the NTD can save money because of vendor lock in. The OLTs, the splitters and other GPON equipment can be provided by other suppliers. It’s just industry standard. The NTD can’t. That’s Hackett’s point. So one has to assume then that $1.5 billion covers the cost of the NTDs if nothing else because of the custom nature of them.

            Could Huawei or another company be commissioned by NBNCo. to make an NBNCo. compatible NTD from their current products? Sure. But I thought that’s what a contract was for??

            Secondly, my point about CPE was not in direct comparison to Enterprise. It was just a explanation about why CPE last less time- it’s not isolated to just because of components.

            Finally, I was simply giving my experience based on my knowledge of my company building industrial systems. If yourself & Hackett are correct, fine. Then we have a reason to discuss the total cost of NTD deployment. I’m simply saying, from my, admittedly limited, experience, systems like an NTD do not have to be expensive to be robust & long life. I could be entirely wrong on the NTD. I just can see evidence for not being, with the cost breakdown I did above.

          • hey mate,

            I’m sorry, but I’m banning you from Delimiter for a week based on the following aspects of the comments policy:

            “Comments which inject demonstrably false information into the debate”

            You are actually arguing, in the comment above, that NBN Co can’t source its NTDs from other suppliers than Alcatel-Lucent. This is just not true. I’m sorry, but they can. There’s no problem with NBN Co going to Huawei or anyone else and saying “make us an NTD that’ll fit our network”.

            Sure, there’s some specific smarts in the Alcatel-Lucent NTDs. But to argue that other companies can’t supply either directly compatible devices, or devices that offer similar functionality, is a nonsense.

            You know I respect you. But you’re extrapolating way too far in this situation.

            Renai

          • After our Netgear & Billion ones quickly failed we then purchased a couple of D-Link G64T routers assuming a spare would be handy. But the first one’s still going strong at around 9 years constantly online. Better quality or just lucky?

          • Renai,

            On the one hand you say people can’t compare consumer devices to enterprise grade equipment (which may or may not be true, depending on what equipment you’re talking about) but youalso keep talking about retail prices which are fundamentally irrelevant in a circumstance such as this. ‘Enterprise’ switches that retail for $1k have sometimes two or three hundred dollars markup. And that’s for single unit wholesale purchases. The actual distributor may be buying them in for less than $500 each. Big players like Walmart may be buying for $100 less again. When you’re buying a million (or 10 million) units at a time you will have a cost plus maybe 20% markup from the manufacturer. Pricing of everything you have previous experience with is simply not applicable here.

            Now, if you have a source who can confirm NBN Co’s pricing per unit for the NTDs that is one thing, but the only thing we really know at this point is that we don’t actually know anything. It would be nice for you to admit you have as much hard evidence as the rest of us, because right now, from my perspective, it looks like you’re suggesting that your guesswork is somehow more valid than everyone else’s guesswork.

          • “Now, if you have a source who can confirm NBN Co’s pricing per unit for the NTDs that is one thing, but the only thing we really know at this point is that we don’t actually know anything. It would be nice for you to admit you have as much hard evidence as the rest of us, because right now, from my perspective, it looks like you’re suggesting that your guesswork is somehow more valid than everyone else’s guesswork.”

            If you’re not used to the fact that I have inside information on most topics yet, then you’re not reading Delimiter closely enough.

            “On the one hand you say people can’t compare consumer devices to enterprise grade equipment (which may or may not be true, depending on what equipment you’re talking about) but youalso keep talking about retail prices which are fundamentally irrelevant in a circumstance such as this. ‘Enterprise’ switches that retail for $1k have sometimes two or three hundred dollars markup. And that’s for single unit wholesale purchases. The actual distributor may be buying them in for less than $500 each. Big players like Walmart may be buying for $100 less again. When you’re buying a million (or 10 million) units at a time you will have a cost plus maybe 20% markup from the manufacturer. Pricing of everything you have previous experience with is simply not applicable here.”

            Stop treating me like a child, mate … this isn’t my first spin around the wheel.

      • Contrast your estimate against cost savings for simplified diagnostics of the network and efficient problem elimination, something that’s only possible if you build diagnostics into the NTD (whatever form that happens to take). While I think your estimate is actually rather high, that is immaterial if it gives you features that are not replaceable with a cheaper device that give you cost savings elsewhere.

          • You’re misquoting there Renai – if you pay more for a device that provides you with a feature that saves money you would have to spend elsewhere that results in savings greater that the cost of the device over the course of a year, then the cost difference between that and any other alternative that cannot provide that feature is immaterial because that device is the only one that can pay for itself within a year.

      • “From what I have been told, the NTD cost is part of the FTTP customer connect cost/demand drop cost, detailed here as currently averaging about $1,100…I wouldn’t be surprised if the NTUs were costing $500 or $600”

        So you think that the NTU could cost as much as all the rest of the labour and equipment used in a drop combined? I find that very difficult to believe, especially as the cost of similar NTUs are so very low.
        With the price of labour for the install being so well negotiated, and since Quigley was so connected with Alcatel, I find it somewhat implausible that NBNCo got so heavily shafted on the cost for those NTUs…

    • I agree 110%….

      The ability to have more than one connection and try various ISPs at any one time is what I like.
      A RSP like Skymesh who let you trial them for free could be run along side whoever. At least you have the option to compare for a month. The problem is the likes of iiNet..Node..Telstra for example want to lock you in for 2 years and with one port you not only have an option to try another ISP but you have to stick with said ISP for 1-2 years. I think this is why Skymesh is gaining immense popularity is that they do offer a trial and from what I can ascertain there are no lock in contracts and excellent quota limits to boot.
      Having a single option to use only one ISP takes away the customers option and choice IMHO.

      • If the idea of being locked in to a contract period is unattractive (which it is to many) and other RSPs aren’t insisting, won’t the lock in product wither on the vine unless a useful discount is offered?

        Surely the best way to exploit the advantages of a universal wholesaler is to take all the delay and complication out of the churn process. That would really deliver on the benefit of the government’s emphasis (obsession?) on the merits of competition.

    • I think the key is choice. If someone wants the option to run the current NTU and have 6 service providers connected, more power to them. If however if someone wants to go to iiTPExetelNode and grab a service as they do now for naked ADSL including a router/VoIP service etc. then why shouldn’t that be an option.

        • If you remove the NTD from the rollout you reduce customer choice. Don’t pretend this gives any better outcome for users, that is spurious and deceptive. This gives better outcomes for RSPs who get to make extra money selling ‘modem’ devices.

          • > If you remove the NTD from the rollout you reduce customer choice.

            Simon has said repeatedly that a multi-provider NTD should be offered by NBNCo. The NTD should be covered in just wholesale AVC costs with 12 months, so there is no reason why it shouldn’t be provided free or at a significant discount.

          • @Matthew

            Offering multiple types of NTD reduces savings of bulk equipment prices. What % taking a single port NTD would provide enough savings to offset the customers who then, at a later date, want a multi-port NTD?

            And you cannot simply say the AVC covers the NTD cost in a year. The NBN costs more than the NTD to build and AVC + CVC has to cover that.

          • > Offering multiple types of NTD reduces savings of bulk equipment prices. What % taking a single port NTD would provide enough savings to offset the customers who then, at a later date, want a multi-port NTD?

            How many customers currently have more than one service connected. I’d suggest it would have to be a very small number simply due to the costs. How much can a customer save by buying a single device rather than an NTD plus a router?

            > And you cannot simply say the AVC covers the NTD cost in a year. The NBN costs more than the NTD to build and AVC + CVC has to cover that.

            But in the hypothetical case that you are discussing where the customer has two services, the second service is effectively a ‘bonus’ since the only significant extra cost to NBNCo is to upgrade the NTD.

          • How many customers currently have more than one service connected.

            If we assume the ballpark 70% broadband penetration rate for broadband remains consistent within the Foxtel footprint, approximately 1.75 million.

            How much can a customer save by buying a single device rather than an NTD plus a router?

            Nothing, the NTD is provided free of charge by NBNCo.

            For more information see the thread where myself and tinman replied.

          • > Nothing, the NTD is provided free of charge by NBNCo.

            There is no such thing as free lunch. The cost of the NTD is built into the network build costs which need to be paid for.

        • I’ve read the presentation. It’s pretty cool, but comes from a commercial perspective. And that colours it slightly.

          Why have multiple ports, if you have just one provider, who provides value add with TV, Radio and umpteen other services? That’s a pretty valid question – if your provider also happens to be into tripple-play and or bundled services.

          There are some pretty obvious cost saving exercises, and from a commercial perspective again that’s a logical, expedient way to speed up deployment.

          As it happens, FTTN will result in exactly this; a single provider, with no simultaneous service option.

          The whole point behind NBN, the choice to use multi-port NTDs and so forth was that it made it an entirely ubiquitous, open service, with the ability to allow for a number of services to be supplied.

          Want TV from one mob, Internet from another? Done. e-Health or other services? not a problem.

          I have a deep respect for Simon; what he has achieved and his clarity of thought makes sense to me more times than I care to count. A pretty smart guy.

          But, being that pretty smart guy, he’s looking to solve this from a purely commercial, single supplier standpoint. And that’s an answer to an entirely different question. :)

          The issue for NBNco has always been getting premises completed done. That’s really what anyone cares about.

          Changing the NTD might help a bit, might reduce costs a bit; but it creates the same basic situation we have now & relies on RSPs (almost purely) innovating alone.

          It’s the right answer to the wrong question, imho.

          • > The whole point behind NBN, the choice to use multi-port NTDs and so forth was that it made it an entirely ubiquitous, open service, with the ability to allow for a number of services to be supplied.

            Have you looked at the costs of a second provider? The minimum wholesale AVC charge is $24, so you would have very little change from $50 for the most basic service. Spending an extra $50/month with a single provider is likely to bring significant benefits.

            I agree with Simon that there will be very few cases where more than one service provider is used in a household. Meanwhile our monthly bills go up because of Labor’s gold plating.

          • @Metthew

            SkyMesh, Exetel etc. disagree. 12Mbps BB is available for as little as $30. Retail.

            And the ARPU is not per port it’s per user ie. premises. Which can have multiple ports.

          • > This is like saying “one port should be enough for anyone”.

            No. Simon clearly states that a multiport NTD should be available for installation where the customer requests it.

            > Are you stating there’s never going to be a reason? Simon is responding from a commercial standpoint, logic dictates that his position by default is that a single RSP is the answer.

            No, but the costs of $50+ for a separate service make it unlikely. How many people today have:
            – traditional phone line and a naked ADSL service?
            – multiple ADSL services?

            > You are also making a number of assumptions over what might be offered in parallel. So is Simon.

            Unless you can find a common usage scenario then it doesn’t matter.

            PS Are you missing the herring button?

          • PS Are you missing the herring button?

            Mathew, you were put on the pre moderate list by Renai. Do you really think testing your luck with such an emotive comment is a good idea?

            See mine and tinman’s reply to another post where you peddled the same ideas and reply there.

          • This is like saying “one port should be enough for anyone”.

            Are you stating there’s never going to be a reason? Simon is responding from a commercial standpoint, logic dictates that his position by default is that a single RSP is the answer.

            You are also making a number of assumptions over what might be offered in parallel. So is Simon.

        • @Renai

          Whilst I’m all for choice of modem/router whatever, the problem with giving users an ability to choose their own NTU will result in a freaking support nightmare. Just ask any ISP if they’d prefer to support a single brand of modem/router or if they’re happy to support every combination of modem/router that currently exists.

          Dollars to donuts, the answer will always be support of a single brand/vendor is preferable.

          It also ensures that full compatibility testing can be done to make sure that whatever is chosen operates to the requirements of the network.

          This is where the concept of Turnbull’s “$50 VDSL modem from the corner shop” will fail.

          • This, to me, is the single biggest benefit to a sole supply of NTU’s – the LACK of choice in this regard is a good option, for support reasons alone.

            The NTU isnt the router though, only the equivalent of the phone jack you connect the modem/router to. So consumers are still going to have that freedom of choice when it comes to the router.

            All this NTU does is provide multiple phone jacks, and have you ever seen someone conciously make a choice over which phone plug to use? Of course not.

            50 years ago, you only really needed 1 or 2 electric sockets in a room, none in some. Today you need a dozen in some. Our needs change, and in future years, as we get more and more connected, its going to seem strange if we limited our options to just one NTU connection. So which is better, dealing with it now, or having to worry about multiple connections in 20 years?

    • Please don’t anyone touch the NTD, its one awesome competition enabler. Sure, find a cheaper option if possible, but make sure it can at least support two services. The battery option is already now an option, and make the PSTN optional as well.

    • The current NTD does indeed promote competition, especially when even today we can see how it’s possible to deliver 3 or 4 completely separate services over the one fibre connection. It’s fair to argue that NBNCo shouldn’t be encouraging a “triple play” where customers are locked to one provider for all of them.

      The question though is how much money is it worth spending on this? I have no doubt that a minority of NBN users will be using multiple ports on their NTD and this will continue in future – but the key word is still “minority”. For institutional and other “special” uses it may be popular, but for the average home user the normal deals on offer mean they’ll probably be more than happy to just plug the fibre in and go. Same thing with battery backup, some people need it but it’s fair to say a majority won’t care. Is it worth spending four figures for every premise on a box a fraction might use (and which just draws more power)?

      Perhaps a fair compromise is that the multi-port NTD and battery backup can be installed free of charge to those who really need – or want – those items, but not installed “by default” as part of the normal rollout. I really don’t think we’d see a huge number of people opting in, so it saves a huge amount of money but service obligations can still be met to those who need them – therefore politically safe.

      • @Michael

        There’s an argument to be made for an optional “less complex” NTD. Perhaps with only a single ethernet port. But doing away with the NTD altogether brings about support issues that RSP hardware would bring with it. Not to mention, what happens when someone else moves in or buys the premises with only a single ethernet port NTD but wants multiples? The customer likely would have to pay. Again, there’s an argument to be made for that, but I still don’t see it as an argument for the removal of a standard NTD for support.

        Secondly, the battery backup is already optional:

        http://www.smh.com.au/it-pro/government-it/nbn-co-to-increase-options-for-battery-backup-20130812-hv1cf.html

        NBNCo. are already implementing these changes. And at the Draft CP we can see the dividend ($700 million in “efficiencies”). But I do not see removing the NTD altogether as an efficiency. Quite the opposite for ongoing support.

        • Can you even get single port ones?

          Every one I saw in my research on it had a minimum of 2 (and it being the odd one out), the rest all had 4 Ethernet and 1-2 voice ports. Most of them are between $35-$120 (though thats mostly Chinese suppliers).

          Is this really an issue?

      • How many premises are in our situation with a granny flat or rental apartment that requires it’s own service?
        We have 2 independent lines in use. (comes in handy at times when one is down).

        • So run Cat 5 under the lawn. Honestly, do you expect NBN Co to resolve every fucking problem for you? They just got fibre to your premises — can’t you take care of the rest??

    • Hey Renai,

      How’s your reality check coming along !

      Similar state of denial to Conroy it seems.

      But just as he’s fallen of a cliff, his delusion can still persist if he closes his eyes and mind for a little bit longer . . . . and then perhaps even to infinity and beyond.

      Govt spending over $ 90 Billion, for the most part giving people access to 100 mb/s which is
      A/ Already available to millions and is ignored.
      B/ Not close to required by the mass’s
      C/ Not be affordable to subscribe to by half the population in the foreseeable future.
      D/ Copper, and Wireless and Commercial business interests are moving forward fast, servicing the need of the people.

      • Nice work really funny stuff…

        Oh… you were serious :(

        Love the moniker too *rolls eyes*

      • Govt spending over $ 90 Billion, for the most part giving people access to 100 mb/s which is

        [citation needed]

        A/ Already available to millions and is ignored.

        As per the latest (leaked) business plan 26% of NBN users opted for 100Mbps. I cannot find stats for those on HFC cable or existing FTTH estates, however that statistic alone disproves your assertion.

        B/ Not close to required by the mass’s

        [citation needed]

        C/ Not be affordable to subscribe to by half the population in the foreseeable future.

        You, like Mathew, need to learn the difference between a conservative estimatio and malicious intent. 100Mbps services are more than affordable, the problem becomes demand for them. Demand for them already exceeds expectations, what makes you think that this trend is not going to continue in the future?
        D/ Copper, and Wireless and Commercial business interests are moving forward fast, servicing the need of the people. There is no fault, or conspiracy here, NBNCo is merely responding to market demand. This is exactly the behaviour we would expect of any commercial organisation.

        You yourself said it’s not required by the masses, so why would were force the masses to use it? We wouldn’t.

        D/ Copper, and Wireless and Commercial business interests are moving forward fast, servicing the need of the people.

        Copper: bullshit. That’s the whole reason the NBN and its predecessor (OPEL) was proposed in the first place, or haven’t you been paying attention to the past decade of Federal politics in this area?

        Wireless: bullshit. Over 90% of broadband traffic is over fixed-line connections, because a universal wireless solution is not scable.

        Commercial business interests: bullshit. Apart from Greenfield FTTH estates, ULL DSL services, and commercial grade fibre back-haul, there has been absolutely no movement in majority of Australian fixed-line telecommunications to improve the state of Broadband since Telstra and Optus engaged in the HFC wars in the early 90s, and that was actually built with PayTV in mind, not Broadband.

        The majority of the Australian population have only seen minor incremental improvements to services thanks to competition in the DSL market. These improvements have been enough to tide us over, but in the past 5 years we have reached the practical limit to this technology and require radical investment, which now both governments have committed to doing. One in the form of a complete overhaul of our base network, another in an incremental upgrade to reduce line lengths and improve back-haul.

        It seems you are the one who require’s a Reality Check, considering your comments seem not to be based on reality at all.

        • Oh so that’swhere point D went when I copied and passed it. Please ignore the un-blockquoted version in the middle of that paragraph.

          • Just ignore him, there is no interest in reality, what happens with broadband, anything. Pure Lib troll.

        • Hey NightKhaos,

          Your leaked report details i wouldnt rely on if i was you.
          But either way just a few more weeks and the ugly truth will come out.

          And there will be plenty of embarrassment to go around for those who have even a slight grip on reality ;-)

          True premises connected numbers which we all know have been revised down and revised down and revised down. And the even more minute and much more important actual subscriber numbers to the NBN.
          26% of NBN users subscribe to 100 mb/s you say. We will see.

          So your hanging your hat on something like –
          163,000 premises “connected” to the NBN.
          Which in fact shows something like 55,000 cant actually connect.
          So the true figure is approx. 108,000.
          Of which something like 40,000 people actually subscribe to anything.
          And your trumpeting 26% of that ridiculously small sample.

          Meanwhile cable pass’s 2.8 million premises. And has done for years.
          Approx. 0.0001 % of subscribers use 100 mb/s today !

          Are you attempting to deceive or are you just a simple fool ;-)

  2. I would love to know the actual costs.

    Sometimes it makes sense to go the better option, because when you take into account economy of scale, the end result of cost is not that different.

    I would be surprised if they were paying $5-600 even for a relatively high end device. The sheer volume that they would be purchasing should bring the cost down significantly.

    Also what is the redundancy and the “warranty” on these.

    If they are paying more for a better device because they get a longer life span on it, and the coverage to support it, then it would makes sense. They don’t want these devices failing and having to be constantly replaced. They would want devices that can be put in a box and left pretty much alone.

    We just don’t have enough info… Anyone on the inside able to drop Renai an anonymous tip ;-)

  3. I hope that Turnbull’s strategic review is made public in all its details. Then we can say for sure whether Simon’s inferences and assumptions hold true and whether his ideas have merit. But I’m cynical – not about Simon’s ideas, but about the transparency of the review. Abbott’s put a gag order on all his MPs, and I think it’s fair to say that the new Liberal government has got “a public information policy that would put the Kremlin to shame”. I doubt Turnbull’s gonna go rogue on this one. He’s been toeing the party line a very long time now.

  4. Hackett, with the help of others such as iiNet’s Michael Malone, Netspace’s Stuart Marburg,

    Nope

    what turnbull should do is to get the accc looking at iinet , buying out its competitors

    • Accc should get off thier backsides and make iinet show what companies they are have taken over

      Not everyone is aware internode and etc are not sperated from iinet ownership

      There is not much competiiton left thanks to accc allowing iinet to buyout the competitors

      • At the end of the day, ACCC have shown that they are happy with just a handful of big national competitors in many sectors.

      • Mate ACCC will only care if there are genuine competition concerns. Given that even after a bunch of acquisitions, Telstra is still a lot bigger than iiNet, it’s arguing the wrong point.

        If there was a genuine concern, it’d have been ruled on.

        iiNet and Internode also have wholesale services available. So, apart from getting upset that the Australian market is consolidating because it’s the only way it can remain viable against a recalcitrant monopoly, do you have an actual point?

        • the point is

          Not everyone wants to iinet, and they think westnet, netscape, internose etc are competitors which they are not
          they are iinet.

          If the ACCC is for competion and consumers , they would make iinet tell which companies they brought out

          • my apology

            Not everyone wants to be with iinet, and they think westnet, netscape, internode etc are competitors which they are not
            they belong to iinet.

            If the ACCC is for competion and consumers , they would make iinet tell which companies they brought out

          • And Jetstar is owned by Qantas.

            Whilst I agree with your sentiment, why does it apply any more here than it does to the various other companies that own “competitors”

            The ownership is of public record.

          • The way Renai trys to make out iinet micheal Malone and Netspace’s Stuart Marburg are different companies which they are not

            They are all under one banner iinet

    • It would be difficult for the ACCC to review this, since it is a direct result of the ACCC determination to force NBNCo to move from 14 POIs to 121 POIs meaning that only 4-5 National RSPs could survive, plus some resellers.

  5. Hackett is no independent industry expert, he’s the largest shareholder of iiNet.

    He will be saying what is in the interests of his company, that is selling termination units, allowing limited flexibility over the network and lumping fibre maintenance with the government at the lowest possible rental charges.

    For all the railing against Telstra he and his company have done over the past decade, they will only want to see a recreation of Telstra Wholesale. Someone else to blame for problems, while they skim the most profitable part of being a telco – backbone infrastructure and retail plans.

    iiNet should be putting its money where its mouth is and rolling out fibre, as part of Turnbull’s co-funding provisions.

  6. Alcatel Lucent, would be making millions NTD units for NBNCo the cost would be well under $100,
    as mentioned its a optic line terminal, VLan switch & POTS ATA..not really some $$$ magic box!

  7. I agree with everyone else, in that Hackett is only interested in locking customers into a contract as the sole provider, and that’s why the ntd should not be changed.

    one way to both speed up the rollout, cut costs, and quickly get users on the network is to use the “pull-through” method. Make things so that customers are connected instantly and seamlessly, instead of waiting 18 months for the contracted Telstra cutover. Then once a fsam is complete, do a letterbox drop informing customers that it’s done, and they can choose their providers.

    • > I agree with everyone else, in that Hackett is only interested in locking customers into a contract as the sole provider, and that’s why the ntd should not be changed.

      Then I suggest you read what Simon has written because he directly contradicts your statement. Simon would be confident that iiNet could successfully compete based on a combination of price and quality.

      The window of opportunity to identify the faults with Labor’s NBN plan and propose solutions is rapidly closing. If you want FTTP, then I suggest that the best strategy is to rip Labor’s failure apart and explain to Malcolm Turnbull how he can fix the mess and claim credit for it. That is a story that Liberal media minders can sell.

      • No, no it doesn’t. Suggesting they replace the NTD with a single port device implies that Simon is assuming that customers will only want one service per household. Thus locking providers into only one RSP.

        A multi-port device removes this assumption, and allows service providers to compete where they hold a specific niche product. At the moment niching isn’t really possible without engaging in infrastructure competition. This is why IPTV providers partner with specific ISPs rather than trying to deliver the service to customers directly.

        • > No, no it doesn’t. Suggesting they replace the NTD with a single port device implies that Simon is assuming that customers will only want one service per household. Thus locking providers into only one RSP.

          If an multi-port NTD is available on demand then lock in doesn’t exist.
          The simple economics of $50+ for a second service mean that the vast majority of households will require only one service.

          > A multi-port device removes this assumption, and allows service providers to compete where they hold a specific niche product. At the moment niching isn’t really possible without engaging in infrastructure competition. This is why IPTV providers partner with specific ISPs rather than trying to deliver the service to customers directly.

          So an IPTV service can either partner with specific ISPs, offer the service directly to customers (e.g. ABC iView, netflix, etc. model) or add a minimum of $24/month to their service costs and add additional support staff.

          Secondly you need to think about the complexities of this. A household now has two separate internet connections meaning that the Smart TV can either access the IPTV service or other online content. The consumer could setup a fancy router with multiple gateways, but that is non-trivial.

          About the only sensible option I’ve seen suggested so far is for teleworkers where the company supplies a separate connection just for work. For this to happen the company would need to partner with a service provider to provide a secure virtual lan and I just don’t see it being that realistic when a simple VPN client can suffice.

          • Secondly you need to think about the complexities of this. A household now has two separate internet connections meaning that the Smart TV can either access the IPTV service or other online content. The consumer could setup a fancy router with multiple gateways, but that is non-trivial.

            Actually, what your suggesting would require exactly what your complaining about. A multi-port setup would just require “Plug that one in the TV, and that one into the computer”, which would be a lot easier for an average consumer (using the “your mum” test).

          • > Actually, what your suggesting would require exactly what your complaining about. A multi-port setup would just require “Plug that one in the TV, and that one into the computer”, which would be a lot easier for an average consumer (using the “your mum” test).

            Except then your smart TV is limited to only accessing FoxTel or you need to have multiple set top boxes.
            It ignores the fact that in many households, wireless is the principal means of access. What happens in dwellings where there are multiple FoxTel boxes?

            It still doesn’t ignore the fact that FoxTel’s service will be $40-50 a month more expensive than competitors can offer a similar service and require FoxTel to run backhaul to every POI or enter into agreements with another RSP.

            Yes you can create a straw man argument, but like suggesting that the majority of the 47% of 12Mbps fibre connections are purely for voice, it doesn’t survive Occam’s razor.

          • If an multi-port NTD is available on demand then lock in doesn’t exist.

            Doesn’t take advantage of economies of scale at all. It will likely work out more expensive to provide multiple tiers of devices for different use cases than to provide one “bare bones” device like the NTD.

            The simple economics of $50+ for a second service mean that the vast majority of households will require only one service.

            There are 2.5 million Foxtel subscribers. With a sub $30 wholesale per customer cost for standard definition service delivery and approximately $50 wholesale per customer cost for high definition service, taking into account content deals and of course cross-subsidisation from the premium and basic packages, I have just presented one service that could easily leverage off the NBN network, and for user ease of use sake, would justify the use of a second port on the NTU, quite easily bringing justification to the previously stated economies of scale argument.

            Your “simple economics” are flawed. You don’t even know how much the product you’re so biased against actually costs.

      • Mathew,

        “Then I suggest you read what Simon has written because he directly contradicts your statement.”

        Does it? Simon has suggested a single port, because it’s better for a single RSP. It’s a commercial response to a problem. It shaves a few dollars off, which adds up over the entire deployment.

        Again, not knocking Simon; a number of his suggestions are logical. However to ignore the fact that it’s a commercial response, ignores the context with which it is proffered.

  8. As Simon says in the presentation – NBN charge a wholesale price PER PORT so you’d have to be insane as a customer to pay for your TV service to be delivered separately to your internet as any benefit from competition would be offset by the NBNCo port wholesale rate. It would make far more sense for IPTV providers to deliver their service over your ISP connection – there can still be competition here.

    If you are a business and want redundancy (two ISPs) you could still buy a multi port NTD that does the same thing as the current NBNCo NTD

    • By doing that you cannot deliver using differentiated priority service levels. There are four levels, Internet delivery is 4, the lowest priority. I believe video conferencing is meant to get a specific service level that is still yet to be determined. But it’s useful to understand – by being able to offer a service over a dedicated port at a higher service level you can guarantee (within a margin of error) the service quality and reliability all customers can expect. If you had to deliver your service over a shared Internet tunnel you could be affected by factors beyond your ability to control.

      This is one of the fundamental features of the NBN, and it is lost without a multi port NTD.

  9. Substituting a 4-port NTD with a 1-port NTD for a $20 saving would be a retrograde step for the nation and for customers, but a nice little earner for ISPs like iiNet and Internode.

    Universal urban availability of a spare fibre port means that occupants who have a basic or phone-only broadband service can always be delivered a plug-and-play device by a hospital, university or government department without any involvement of their main RSP.

    Churning to a better provider can be done by activating the new service, moving the cable to the new port, then deactivating the old service, with no down time.

    Eliminating the backup battery and the phone ports could be made an option for the occupant, because they will eventually be decommissioned. In the first rollout, however, we are starting from a world where the vast majority still use a fixed-line telephone, and it would be disruptive and unpopular not to provide a phone port option. A four-data and zero-phone unit with no battery is a good design, in my view. Those requiring backup power can buy a UPS or solar solution supplied by a third party market.

    Yes, there are some good ideas in Simon’s 20-minute elevator pitch, but the real savings are to be found in labour hours, not this simple yet advanced and effective piece of kit.

  10. Sorry guys, but anyone who thinks Alcatel-Lucent is selling NBN Co NTUs for $100 or even $150 … doesn’t know Alcatel-Lucent ;)

    I could believe Huawei could go that low, but I don’t believe Alcatel-Lucent would. There’s a reason Hackett is harping on about how expensive the NTU is, and from what I’ve heard from inside the mothership as well, the NTU is a decent portion of the demand drop cost. It’s not $100. I don’t know how much it is — but it’s not $100 or it would be a non-issue.

  11. Hackett has it mostly right, other than the simplification of the NTD device. Reducing this to a single channel removes a lot of options, and seriously, how much extra would it cost at manufacturing time – bugger all.

    I currently have cable modems from Telstra and Optus at home, how am I going to have 2 ISP’s with only one channel on the NTD. What about future capacity. What about the “government services” promised over one of the dedicated channels.

    Leave the NTD alone Hackett! Its part of the NEXT generation network we need.

  12. Renai, its a little bit amusing that you haven’t mentioned just how much Simon’s ideas are being demolished on that same whirlpool thread that you referenced.

    Simon has a lot of cred in the industry (and respect from myself), but he has left himself exposed on this one.

    • Whilst I (probably like you) love whirlpool, I don’t agree that the arguments against Simon’s views have been proven, only that they have been forcefully and repetitiously made. With pertinent details shrouded behind commercial-in-confidence, it can’t be said if Simon or the “demolishers” are more correct.

      • Possibly only because he refuses to actually respond to the criticisms that are raised in the thread. Take a look. His main defense (used multiple times) is that the criticisms are from people who havent listened to his speech, when it is patently obvious that they have (I should know, I listened to it myself).

        Its disheartening to be treated like that by someone who is supposed to be one of the good guys.

        • It’s also disheartening when you propose innovative ideas for the NBN rollout and get derision from Internet forum posters ;) From what I have seen, Simon has been responding to the critics.

          • I have said a few times that I feel that Simon isn’t really trying to innovate that hard, that he wants a solution that makes it easy for him.

            I know this is hard on him, afterall he’s only human, and if I were in his position I’d do the exact same thing. Afterall, if I can get a solution that benefits my interests and reduces risks to myself, all the better.

            But just because we empathise with how hard it must be to have your ideas come under such heavy scruity does not mean that we should “go easy on him”. As a jounralist, Renai, I think you can understand this.

            I do however wish that, and yes, I know on occasion I suffer from this too, the people discussing this could be a little more diplomatic and level headed. Unfortunately we’re talking about a group of people that have been having this debate for a little over 4 years now, and have been constantly ignored by people with vested interests, threatening to undermine this thing they so passionately believe in.

            And it is my experience that people who are disenfranchised or feel that they haven’t been heard are prone to lashing out, even against people who are legitimately trying to help them. I think this is what we’re seeing with Whirlpoolers.

    • They are being “demolished” by a bunch of nimrods who haven’t yet recognised that they are arguing about the quality of the damn deckchairs while Simon is trying to save the boat from sinking…

      This comments section is exactly the same.

      Hackett is a goddamned saint when it comes to promoting and deploying faster internet in Aus. His predictions have been often spot on. He has been pro-FTTP since before it was announced. His credentials have been beyond reproach for a long time.

      And now Turnbull has left the door wide open for someone to come along and save FTTP. I honestly thought everyone would be jumping for joy since the petition kinda petered out to little effect.

      And all most of you (and the WP chuckleheads) can do is attack him for suggesting slimming the build down on the taxpayers dollar might be a way to save the ship. Oh woe is you, you might have to shell out some cash for the big expensive unit rather than a consumer one to get the service you (and likely very few others) want, what a bloody tragedy… /eyeroll

      Here’s a great idea, keep on attacking Simon for going in to bat FOR YOU, and perhaps you’ll end up with FTTN. You won’t have to argue about what fibre unit you’ll have installed. Would that make you people happy?

      • @Asmodai

        I think that’s unnecessarily harsh.

        I don’t think anyone here believes that Simon does not have a considerable wealth of knowledge, experience & innovative ideas for the NBN.

        And I don’t think anyone here would disagree a smaller cost of network would be a win for FTTP possibilities.

        But there is a very very poignant argument that saving money for the sake of it or saving money by removing a huge amount of potential or innovation of the network is a huge mistake.

        What would be the point of a nationwide FTTP system if it operated at the same level of frustration & cost for support and maintenance because of lack of planning on that facet of design? Sure, we’d end up with faster internet….but that’s not the point of the NBN.

        IMO, the NTD is an enabler of cheap, reliable & ubiquitous government, education, health & aged care services from the get go. Could 3rd parties come up with OTT solutions instead? Of course. But dozens of OTT services across a network designed for giving the same experience to all is a recipe for support frustration, high cost and loss of innovation.

        • Well it’s time someone gave you a short, sharp, harsh dose of reality…

          The ALP lost. FTTP is on life support and Turnbull will be able to hit the switch once the senate changes. That’s the situation.

          Let’s put it another way, you can’t pay your mortgage and are about to lose your house. Someone comes along and suggests ways to slim your budget down to keep it, trimming the luxuries. Do you insult him and refuse, or thank him and implement?

          You can go on about the ideology of build the best now and save later on, it’s irrelevant because the Libs aren’t going to see reason when their primary concern in this matter is cost. The door is open but you continue to try and run through a brick wall.

          I don’t have a vested interest one way or another here, just pointing out the absolute insanity of refusing the offer of help when your drowning and you’ve lost the last straw…

          • Do you insult him and refuse, or thank him and implement it?

            This is called a false dichotomy, it is when you present two extremes as the only possible solutions.

            We are arguing against one solution he has proposed. ONE. Where are the arguments (in Delimiter anyway) against the other points he brought up in that 20 minute presentation?

            Seven_tech might be exploiting a little to much, he may be a little too passionate, but he doesn’t need a “hard dose of reality”, he’s well aware of the situation. As am I, and everyone here who disagrees with Hackett’s stance on NTDs.

            That’s what it is, a disagreement. We don’t think it is in the project’s best interests to implement this one particular change proposed by Hackett. That in no way implies we think that his solution isn’t viable (the closet we’ve come is seven_tech saying he thinks the costs of the NTD have been overestimated), nor does it say anything about what we think about his other proposals.

      • So attacking is now your version of a discussion? A particularly one sided discussion, since Simon refuses to defend his ideas.
        I am not seeing anything but a discussion trying to be constructive but getting frustrated from the lack of respect from one of the participants.

  13. A story based on potential cost savings when the exact cost break down is unknown, I am all for debate but seriously this is just confusing the public even more so.

    Secondly, Simon has a vested interest in a system that works to his advantage, it’s commonly refereed to as a “conflict of interest” which I find most interesting that a media outlet is promoting, well I’ve come to the conclusion that it’s more of a personal blog.

    Without knowing the specifics of the NTD, having been a part of many large infrastructure projects, I can say that repetition is the key. A standardised delivery model although not always the cheapest will save costs in many other ways, the most simplistic being staff training and support.

    It’s the ongoing support costs that tend to add up over the lifespan of the service, minimising those costs is important, offsetting them to somewhere else (ISP) is simply shifting the cost.

    AL products have been used extensively by Telstra in the past, if there weren’t “good deals” to be had in bulk purchases I can’t see Telstra having used them. food for thought.

    • “Secondly, Simon has a vested interest in a system that works to his advantage, it’s commonly refereed to as a “conflict of interest” which I find most interesting that a media outlet is promoting, well I’ve come to the conclusion that it’s more of a personal blog.”

      And the people calling for FTTP with all the trimmings no matter what don’t have a vested interest?

      You guys have got to rid yourself of the delusion that faster internet is a number one priority for most people, it’s not. The election proved that. Now, FTTP is hanging by a very slender thread, Turnbull’s pre-election statement that if the costs bore out, he would use it where affordable, and the post election invite to demonstrate how savings could be made. Simon is trying to reinforce that thread and is copping nothing but flak over it.

      I can’t stop /boggling at the sheer inanity of being in a situation where FTTN is almost certain and you are biting the hand that is trying to help you get FTTP…

      • Slow down, tiger.

        I’d like to have my jugular left intact. An emotive dummy spit that the world will end because people don’t all simultaneously agree that a single port NTD will save the NBN, is a bit overly dramatic.

        A single port NTD, won’t save the NBN any more than going to a two port, would. Ascribing a technical change to one component out of many, in isolation, as damning us all to FTTN, basically means you really, really haven’t been paying attention to what is actually happening.

        Frankly, changing the NTD isn’t so much shifting the deck chairs around, as removing them entirely. The vast percentage of cost is going to be the actual deployment of fibre, and the ongoing support of it and associated systems; how many chairs you have, is but a single part of a far bigger beast.

        That’s where Hackett is coming from; rip out the need for anything more than a single RSP providing value add. Like I said earlier, that’s commercially and deployment expedient in getting the job done faster. It’s cheaper and likely simpler to manage.

        NBNco’s issue isn’t so much the design; it’s not perfect by any means, but it’s main sin is simply being slow to deploy, compounded by delays from Telstra for various reasons.

        Said delays would have existed regardless of how many deck chairs you have to arrange.

        • “I’d like to have my jugular left intact. An emotive dummy spit that the world will end because people don’t all simultaneously agree that a single port NTD will save the NBN, is a bit overly dramatic.”

          Srsly, that’s your best shot. Simon and I are suggesting that changing the NTD will make the thing affordable..? /facepalm

          Simon suggested many measures, this just happens to be the one that people have latched on to… I’m not suggesting anything other than “You may have to give up the frills to get the core product”…

          I’m not saying the world will end as I don’t really care if it’s FTTP or FTTN… My harsh reaction is generated by sheer disbelief that a drowning person would take the time out of their demise to spit on the person trying to save them…

          You want to continue not only shooting the messengers but the bloody relief force, go nuts.

          • Drowning people now? Of course I don’t consider the NTD to be the only factor. Really?

            Sorry – but that’s all a bit silly. The whole situation is retarded as there’s been functional fibre network being built. It’s only sin has to been to be deployed slower (initially) than anyone could predict, and started outside in, which has lead to the “are we there yet?” bitching.

            If every project was canned because it ran into some kind of issue at some point, we’d have no roads, no infrastructure, no cities and pretty much still be hanging out in trees, to be honest.

            Probably a simpler life, I grant you – but nothing is ever going to be simple about this build.

            Simon’s NBN, Rudds or Turnbulls will all face challenges, delays and revisions. FTTH or FTTN. Doesn’t matter how much you do, or do not simplify.

            FTTB, or a single port NTD might answer a few issues; but it doesn’t solve asbestos, delays from Telstra, low-balled quotes, and so on. Those are the things even the best design won’t magic away.

            I’ve read Simon’s recommendations. I don’t agree with all of the suggestions. That’s ok. I’m allowed to. Even Simon suggests it’s not a cure-all. It’s a creative way to speed up the build, at the cost of functionality.

            The thing is, it’s all now entirely a moot point. Turnbull isn’t Simon (more’s the pity) – this is now, purely, a political decision and it matters very little what the constituents may want, as Turnbull has stated more than once the people have voted and so he shall be the arbiter of that vote.

            So, whilst I hope logic and reason will form the eventual outcome, that’s presuming Turnbull will see fit to accept advice as part of review, or will instead doggedly deploy FTTN.

            So arguing which FTTH option is better is pointless when the guy in charge isn’t actually listening.

            tl;dr – it is no longer a technical issue, and hasn’t been for a very long time – it’s political choice (deja-vu); that will dictate the technology and the course of action.

      • tl;dr

        Turnbull will implement FTTN if NBNco and Telstra can re-tool to do so. Semantic debates around how to make FTTH no frills, to speed deployment, ignores that this is now a political decision, with a pre-determined set of outcomes, not one solely based on technical merit.

  14. “So run Cat 5 under the lawn”
    How is that relevant?

    We have done just that already but as this service is independent of the house provider a 2nd port would be required for the 2nd provider.

  15. I don’t believe for one second that people voted the libs in to stop the FTTH roll out, they voted that way because they wanted a change.. A 50/50 chance of obtaining power are pretty good odds in which most people would be happy to accept, that’s how our political system works.

    On the flip side if people voted on policy and not a party, I believe the NBN would have gotten through. Lets face it the coalition mislead the public on numerous counts to turn the public against the NBN, perhaps it’s time ALL politicians were held accountable for what they say.

Comments are closed.