Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
National Broadband Network Select Committee
04/03/2016
National Broadband Network

GLANCE, Dr David George, Private capacity

GREGORY, Dr Mark Alexander, Senior Lecturer, RMIT University

LOWERY, Professor Arthur, Private capacity

NIRMALATHAS, Professor Ampalavanapillai (Thas), Director, Melbourne Networked Society Institute, University of Melbourne

TUCKER, Professor Rod, Laureate Emeritus Professor, Department of Electrical and Electronic Engineering, University of Melbourne

Committee met at 09:05

CHAIR ( Senator McLucas ): Good morning everybody. I declare open this public hearing of the Senate Select Committee on the National Broadband Network. I welcome you all here today. This is a public hearing, and a Hansard transcript of the proceedings is being made. Before the committee starts taking evidence I remind all witnesses that in giving evidence to the committee they are protected by parliamentary privilege. It is unlawful for anyone to threaten or disadvantage a witness on account of evidence being given to a committee, and such action may be treated by the Senate as a contempt. It is also a contempt to give false or misleading evidence to a committee. I would like to emphasise that while the committee prefers all evidence to be given in public, under the Senate's resolutions witnesses have the right to request to be heard in private session. If you would like any of your evidence to be heard in camera, please do not hesitate to let the committee know. If a witness objects to answering a question, the witness should state the ground on which the objection is taken and the committee will determine whether it will insist on an answer, having regard to the ground which is claimed. If the committee determines to insist on an answer, a witness may request that the answer be given in camera. As noted previously, such a request may be made at any other time.

Gentlemen, that is a standard set of words that is said in front of every committee. We are not expecting this to be a difficult hearing at all. For the record, would you like to provide any additional information about the capacity in which you appear today?

Dr Glance : I am the Director of the UWA Centre for Software Practice.

Prof. Nirmalathas : In addition to my role as Director of the Melbourne Networked Society Institute, I am the Academic Director for the Melbourne Accelerator Program.

Prof. Lowery : I am an ARC Laureate Fellow at Monash University.

CHAIR: Thank you. I will now ask each of you if you would like to make an opening statement or, as Professor Lowery calls it, a 'speechette', which I think is rather cute.

Prof. Lowery : Good morning senators, academics and members of the public out there. I am an ARC Laureate Fellow in electrophotonic interchanges. They are the freeway network for optical communications—the backbone. I am a longtime systems engineer and a Fellow of the IEEE for leadership in software tools for communications systems design. More importantly to you is that I co-founded a company called VPIsystems in 1996, a company that has driven innovation and rollouts in optical communications around the world by developing software tools across the bandwidth value chain—that is, for component designers, network equipment manufacturers, operators and service providers. Why is this important? Because VPI supports optimised engineering by selecting the optimum mix of technologies to provide services—your new plan for the NBN. Many companies and countries have benefited from these planning tools, and I should add that I have no financial interest in VPI anymore.

I have read your report on the old plan and the new plan, and I agree with most of the text and recommendations. Good engineering is about optimisation by considering all available technologies, including considerations throughout the life cycle—upgradability, maintenance, suitability, return on investment and financing. Most importantly for complex projects like these, the engineering decisions should be supported by vendor-neutral software. From my experience of building even a smallish software team of a hundred people, it is extremely difficult to recruit enough engineering experts to make the right decisions. I do not know how you could have done this, as you need the experts to interview the experts, and the reports allude to this. It is also difficult to attract the best engineers when they know they will be redundant after the build.

For the last 12 years, I have been working on new types of optical system at Monash University. We are developing optical chips to lower the cost and energy consumption of complex optical networks. In 2008, I founded a company called Ofidium, which holds key IP in access and long-haul systems. I am also leading the development of a cortical implant to provide bionic vision and a recording brain interface to enable bidirectional communications to prosthetics, for example. My next mission in optics is to roll out a 10-terabit-per-second research network, which will connect RMIT, Melbourne and Monash universities, to allow us to perfect very high performance optical switching and processing together, so we can develop solutions for real-world problems. I sincerely believe that more thinking is required on the NBN. For example, it is easy to imagine hybrid methods that could cut billions from the cost of NBN while still providing the enormous download speeds associated with fibre. Thank you very much.

Prof. Nirmalathas : I would like to make a very quick couple of points. We approach the challenge society faces through the network society lens, where the ICT technologies pretty much decide every aspect of the society, from production to labour to the cultural aspect of the society. We see that as a very powerful framework to see how Australia could develop itself, both in regional areas as well as within the cities. The NBN is an important platform for activating that connectivity, because we are interested in using connectivity between people, places and things as a way of running society efficiently. In that context, I think the investment in NBN should be world class.

One of the things we see, despite the NBN implementation, is that it has been a bit of a struggle for us to lift our international ranking. If you look at it, there is a widening gap between the peak and the average speeds as the network builds up. The average speed is improving over the years, but other countries have been systematically able to improve at a much greater rate than we have been able to do. There is potential for a widening gap between people who have it and people who do not. While the NBN promises us a lot of opportunities for improving, accelerating that NBN rollout is a critical step so that we can reap the benefits offered by this concept called Internet of Things, where we have a pervasive network of devices, people and systems that can drive different industry sectors.

In that, if you look at the revised NBN plans—the greenfield sites and the brownfield sites—there have been significant down-projections of the rollout plan. I see that there are some statistics we need to publish more openly so that we can see how we are doing. For example, the greenfield sites—I think there are about 200,000 new premises coming on board, in terms of registration within Australia, but, in terms of the greenfield deployment, it is at a much lower rate. There is a risk of converting today's greenfield sites into the accumulating backlog of brownfield sites, if we are not making them NBN-activated pretty much immediately as premises come on board.

These challenges could possibly delay Australia's capacity to embrace the Internet of Things, in particular, in the regions and towns, because Australia cannot just see development by looking at only major cities. The major cities are surrounded by larger regional cities and the regional cities are surrounded by towns. We have to see culture, economy, labour distribution and also housing and development all as a connected city framework. It is great to see in recent times that connected city focus is on the government's priority, but we need to accelerate the NBN as a launching pad for being able to help the industry to reap the benefits. In that regard, even the satellite deployment plan already looks like we might, very quickly, hit a shortfall again, and we should be looking at a third satellite launch in order to really enable the IoT in regions and villages. With that, I might stop there.

Prof. Tucker : I had not planned to make a statement, but I have now been stimulated by my colleague, Professor Nirmalathas. Picking up on something that he said, I would like to make the point that we need to be planning for the future. If one looks at the capacity of broadband networks around the world—I have been doing quite a lot of work on this in recent years—it is true that Australia's broadband capabilities are falling behind the rest of the world. We are currently about 49th in the world and, by projections that I have been looking at, it is possible that we could hit 100th in the world, even with the NBN coming on stream. We need to be thinking about the future.

The important point here is that all around the world, demand for bandwidth is growing and always has been. Ever since the days of the dial-up modems, there has been an exponential growth in the capacity required by people, the bandwidth used and the data downloaded. That exponential growth is not stopping. Even though we might have sufficient capacity now or in the near future with the NBN, we need to be thinking further ahead. In that sense, exponential growth will continue and, in another 10 or 20 years, there will be a need to make sure the NBN can be upgraded to the capacity needed in the future. We need to keep in our minds that the engineering decisions we make now are looking, over the long term, well into the future.

Dr Glance : Following on from Rod's comments, I was not going to say anything, but I thought, just in case the question does not come up, in the past 10 years I have been involved in rolling out clinical systems to a variety of different locations, including rural areas. The problems and difficulties that we encountered in doing that are still prevalent today—having to rely on satellite systems to provide access for these communities and clinicians that provide the care there. These were not particularly exceptional demands. They were things like uploading retinal images, for example, so that someone in Perth could review and make a diagnosis. It was trying to do videoconferencing for a GP to access specialist care. In places like Kununurra, it might be performing an operation whilst getting assistance over videoconferencing at the same time. All of those things became incredibly difficult to deliver, even when we went to extreme lengths to get business-grade satellite provisions. We have been in a situation, for example, outside the detention centres in Darwin where, every time an aeroplane landed and everybody switched on their cell phones, the actual clinical systems went down because they could not get the bandwidth through the 4G network that was available at the time.

These are not particularly futuristic scenarios; they are scenarios that we have been tackling and systems that we have been trying to deliver today. They do not encompass the demands, as Professor Tucker has mentioned, of the future requirements. If we are going to deliver care to the home, for example, which increasingly is one of the requirements in terms of videoconferencing, we are not going to do it. Having just set up an ADSL connection for a neighbour and seen a commercial offering which is, at most, six megabits per second, it is clear that we cannot rely on private enterprise to deliver this infrastructure and this functionality. That is one of the critical futures of what any infrastructure needs to be able to deliver.

Dr Gregory : I would like to make a few comments a little bit more specifically about things related to the NBN, if I may. I would like to start with something, which I consider to be very important, that relates to the conduct of the NBN.

It is very important to understand that the current telecommunications market in Australia is very unbalanced. We made an error and we missed the opportunity to split Telstra into two companies a decade ago. We still have an opportunity to do that, as our colleagues across the ditch in New Zealand demonstrated to us in 2012. It would mean that there would need to be a re-organisation of how retail and wholesale work but this is something that the Senate committee should consider because it is most embarrassing to have New Zealand show us the way in anything.

It is important that the ACCC maintain its monopoly infrastructure role. I know that there have been discussion and talk about this role being removed from the ACCC. That would be a mistake. We have seen errors like this before that have caused chaos and have taken decades to restore. I also strongly support the ACCC's call for performance monitoring of the NBN. It is very much the case that we need to differentiate where the problems lie—is the problem something that is a responsibility of nbn co or the responsibility of the retail service providers? Technically, it is only due to performance monitoring and identifying where congestion is that we will be able to identify where the responsibility lies.

Now, if I may, in regards to nbn co, nbn co should not be able to discriminate on pricing of products. The way that nbn co has been trying to slowly do this—with the aid of recent legislation that has been pulled from the agenda—is wrong. At this stage the goal has got to be to get the NBN rolled out. Let's worry about finessing competitive markets and the implications of what that does later, after we have got a firm basis upon which we can move forward. We need the nbn co to roll out a wholesale wi-fi network and we need this desperately.

Unless the government is asleep, everyone should have noticed that Telstra has been expending huge amounts of money on Telstra Air. Telstra has been signing up councils around Australia, exclusively, to the Telstra Air network. This means that Telstra are expending funds in such a way that they are gaining a creep on what you would set call 'community' or 'public' or 'national' wi-fi.

Technologies coming down the pipe for wi-fi are going to be very important. In the next 10 years wi-fi is not going to be what we think it is now. Wi-fi and the changes in that wireless area, including the introduction of LTE into the 2.45 gigahertz and other bands, means that we are going to have a very complimentary environment between wireless and fixed networks, and the people at Telstra know what they are doing.

It is very important that a life cycle cost analysis be done for fibre to the node, HFC and fibre to the premise. If this were done at the start, rather than a fantasy cost benefit analysis, we would have very easily seen that there was no reason to build fibre to the node. We saw recently that AT&T conducted such a study and AT&T have published those results. Their results clearly show that fibre to the premise is the way forward. In the United States we now see companies are only talking about fibre, not talking about fibre to the node or anything else, and that is because they carried out the correct engineering study.

NBN Co unfortunately has some problems and the media and communications group is leading the way. The relationship between the media and communications group at nbn co and organisations that are not responsive to their point of view is very negative. This has been highlighted by people before, particularly the leadership of nbn co's media and communications group. In January this year, after four years of a very good relationship with nbn co and being able to visit many NBN sites—I have been to probably 15 or more over that period of time including locations such as Darwin, Geraldton, Queensland, Sydney, Melbourne, Tasmania and so forth—I was sent an email telling me that I was no longer to be provided with any support by nbn co, that I was to be no longer provided with any responses to questions that I had for nbn co and that nbn co was not interested in academia. In their media and communications group around the country, nbn co has 60 or more people. I find it amazing that they cannot afford one person out of such a large organisation with such a large budget to be a relationship manager to deal with academia. If we do not research, if we do not inquire, then we academics are failing, but, more to the point, the government has failed us.

I would like to move from that very negative point to something that I believe is very positive. I think it is very important that we look ahead—one way or another we are only five years away, hopefully, from the end of the NBN rollout. I would like to encourage the committee to recommend to government that we form a broad panel, now, consisting of people from industry, government, consumer advocates and academia to look at what should happen to the NBN post 31 December 2020.

This is going to be one of the most important public infrastructure decisions in this nation's history, if not the most important public infrastructure decision in this nation's history. We need to talk about it now. The next five years are going to pass very quickly. We need to form a consensus view on the future for the NBN. It is time that the NBN stopped being a political football. It has been the most disappointing thing in my career to see the NBN turned into something that detracts from what the engineers are doing at nbn co—which is excellent work. I have expressed, in a number of different forums and ways, very strong views on the NBN. All I am concerned with is that we get the best value for the Australian public, moving forward.

CHAIR: Thank you very much, Dr Gregory. There are a number of themes that have come out of your original comments and I wonder if, Deputy Chair, we could think about having a conversation around the NBN and the comments you have just made, Dr Gregory. I want to pursue Dr Glance's comments around health and the NBN—that is my personal interest but I do not want put my personal interest in front of everyone else. Are there other themes that people might want to progress?

Senator LUDLAM: Today?

CHAIR: Yes, with the group of academics that we have, here, today.

Senator LUDLAM: I have two things that I would like to progress. The idea of this hearing was to put fresh eyes on the NBN. The debate has got pretty stale and pretty stuck and it is right down in the weeds. It was to have you guys give us a bit of a helicopter view, anything that you can tell us by way of global comparisons of what our contemporary or like-minded countries might be putting into the field, and where you think we should go from here. All of you have written, extensively, quite sharp critiques of where we have got to. If we could pull the political football off the field today and talk about where to from here—you are in Parliament House, it is a political environment, but this is a project that has got to survive political cycles.

That was the intention. We have not called Telstra, we have not called nbn co, today, we have called people who do not normally get to front these sorts of forums.

CHAIR: Let us talk about the 'now', then we will talk about the future stuff and then we will talk about health and whatever else comes up as a theme as well.

Senator JOHNSTON: I would like to talk about what other countries are doing.

CHAIR: So the global comparisons. NBN now—I am particularly concerned at your comments, Dr Gregory, about non-engagement with nbn co and them not being interested in talking to academia. Let us progress those conversations that Dr Gregory raised, about what we need to do now, and pick up on the comments that you made. Dr Gregory, when you were engaging with NBN what was the value to your work, to the NBN and to the community of your engagement?

Dr Gregory : I feel that my engagement with NBN was very positive because not only could it be folded back into my teaching and research at RMIT University but also it was in discussions with my colleagues. It provides value in terms of providing a different perspective on what is happening with this project. I have looked back at how academics were involved in the Snowy Mountains Scheme and other large projects that we have had over our history, and there seems to be less interaction today between academia and some of the major things that have gone on over our history. We have recently seen reviews of the Sydney Opera House and, at the time the Opera House was being built there was a huge amount of interaction between academia, the builders, the government and so forth—a mountain of interaction, both positive and negative and so forth. I think that the debate is important. It is not that one side is wrong and the other side is right; it is that through the debate we all learn, and knowledge provides strength.

CHAIR: Are there any other comments? Please, let's turn this into a conversation.

Prof. Nirmalathas : Over the last two years we have been closely engaging with NBN more on the technology side of it, and I think they found some of the teams very responsive and interactive. I think that very soon we will host a major technology delegation back into the university to look at some of the emerging challenges and emerging applications that we are doing inside the lab. I find it easy to approach NBN technical teams, and we have been collaborating with them.

CHAIR: Good. Any other comments?

Senator JOHNSTON: Chair, I think this is a great opportunity to talk to some people who have a very broad view both domestically and internationally. I am interested in talking about fibre-to-the-node and fibre-to-the-premise. In the very brief but intense period of time that I have tried to come to terms with what is a very deep technological cultural subject, I have noted that some countries around the world—the UK, Germany, the US, Switzerland—are doing fibre-to-copper. I am interested in what potential there is to meet Dr Gregory's and Professor Tucker's ambitions for the future. My questions are: how are those rollouts going? How do they differ from ours? And what is the upgradable potential for us, given what we are doing right now and into the future—the 2020-2030 conundrum and challenge that I think has been put on the table particularly by you, Professor Tucker? I am happy for you to try and tell us where you think we are at in terms of what they are doing and what we are doing.

Prof. Tucker : A moment ago you mentioned the fibre-to-the-node in the United States, and I think it is important to recognise the fact that in the past three years, since the change of management in nbn co, the whole scene in the US has changed. Up to the point of three years ago in the US there were two companies that were taking two different directions. One was Verizon, which was rolling out fibre-to-the-premises, and the other was AT&T, which was rolling out fibre-to-the-node. This was happening at around the time of the global financial crisis and Verizon tapered out their rollout, whereas AT&T continued with fibre-to-the-node.

Senator JOHNSTON: And they have done this engineering report that we have heard about.

Prof. Tucker : But since then it has all changed, and in the last three years there has been a total turnaround. AT&T has totally stopped fibre-to-the-node and have done a complete conversion to fibre-to-the-premises. This has been driven by a recognition of the fact that they were, I guess, primarily on competitive grounds, because there was a number of other companies in the US such as Comcast and, in particular, Google, that had been rolling out fibre-to-the-premises. Google is well known now for its gigabit fibre-to-the-premises rollouts.

AT&T were losing market share. They have just now gone to fibre to the premises. I would like to give a quote from something that AT&T put to the FCC—the Federal Communications Commission—just a few months ago:

Demand is growing faster for broadband speeds than AT&T, or anyone else for that matter, can deliver with fibre to the node.

Their recognition is that fibre to the node is not able to cope with the increasing demands that people have for bandwidth so there has been that change to fibre to the premises away from fibre to the node. You also mentioned a moment ago the UK and—

Senator JOHNSTON: Switzerland and Germany.

Prof. Tucker : Yes, now those countries were actually mentioned in one of the corporate plans of nbn co as examples. The nbn co 2016 corporate plan had a small section on overseas activities and it mentioned those few examples and not much else, I must say.

I have in front of me here—I can show you later on if you like or put it on record—a bar chart showing the penetration of optical fibre in European countries. The bar chart shows the height of the bar as the penetration of optical fibre. There is something like 20 countries in Europe that have more fibre than the UK and Germany. It turns out that Germany and the UK are real laggers in Europe when it comes to fibre to the premises. It is not a good example to be picking the laggers when you are thinking about what is going on in other parts of the world.

My argument would be that that was not a good comparison and there are many other countries including Russia and many of the Scandinavian countries that have a very large and expanding fibre-to-the-premises rollout.

Senator JOHNSTON: So why is it that the UK, Germany and Switzerland are doing fibre to copper? What is the basis for that?

Dr Gregory : Time. May I come in on this?

Prof. Tucker : Go ahead.

Dr Gregory : The reason that those three countries are doing fibre to the copper was primarily because they started upgrading their networks far before anyone else. They started upgrading their networks in the early 2000s. Many other countries, like us, put it off. The issue is though: was their investment in fibre to the copper in early 2000s and subsequently valid? Yes. That was because they would be well into the lifetime of those technologies before new technologies came onstream. The important point is that back in 2009-10 industry engineers and academia could see the technologies that were going to be rolled out between 2010 and 2025.

One of the main reasons that AT&T did their spectacular backflip was because new technology called 'Next-Generation PON2' came onto the market. Essentially, that is a third-generation fibre-to-the-premises system. It is not a first generation, not a second generation but a third generation. It was the advent of this new Next-Generation PON2—the capabilities and the decreased cost of using that technology—that finally pushed the numbers, if you like, across into the fibre-to-the-premises column saying that this is advantageous.

People like myself, colleagues and others have been arguing since 2009-10 that if we are going to do a new network it has got to have a 30-year lifetime. We could see that by 2015, five or six years after 2009, new technologies would come onstream that would make investment in fibre to the node unwarranted.

Given that, essentially from today, 2016 on, every dollar spent on copper is 20c more expensive than a dollar spent on fibre to the premise. Over the next 10 years that gap is going to grow. AT&T have demonstrated clearly, even though we are late, that we should not continue down the copper path.

There has recently been a lot of hoo-ha made about a lab test that was done by Comcast about 10 gigabit per second HFC. It was a lab test. That technology would not hit the streets for 10 years, if ever. We have 10 gigabit per second next generation PON 2 being rolled out in Portugal, the United States, China and around the world. It is here today. Lab tests, HFC and Comcast and all that sort of stuff is just unfortunate fantasy. It is really disappointing to see senior executives from nbn co running around carrying on as if all their problems have been solved.

Prof. Lowery : I get annoyed a bit about the NBN debate, just as a private citizen. As an engineer, what you need is a requirement specification. The requirement of the NBN was primarily to get a more even distribution of bandwidth across the country. I entirely agree with that. That is a laudable political aim. There are a variety of ways to do that, but you should compare the life cycle costs of these various technologies. Sometimes satellite wins and sometimes fibre wins. I am broadly in agreement with the rest of the panel that if there is a new path or greenfield site then obviously you should put fibre in, because it is demonstrably more future proof than other technologies. But I also think that if you are going to dig up everybody's front garden in the suburbs just to put a new type of electromagnetic carrier across the garden—a carrier with relatively small bandwidth—that could be a waste of money in some people's eyes. Why not keep that cable going across the front yard, under the house and up through the floorboards? It all works fine. Then all you need to do is split up the HFC network a bit more on a finer granularity and put more fibre feeds in, and for a lot less money you have improved the internet of many, many people.

My point about engineers is that when you are running a telecoms company you should not be planning for what is going to happen before 2020; you should be planning how you are going to have a telecoms company that works forever. That involves having a workforce that is confident that it is going to be with you. Otherwise you get the situation like I have, where one student has gone to Apple, one has gone to Google Fiber, another has gone to Bell Labs—all the good brains disappear from Australia. Yes, fibre is good, but I think these comparisons with other countries miss the point slightly, in that we need to think about what capacity is needed where. Forty megabits per second is probably good for streaming TV. I doubt that any new apartment in Melbourne would need a 4K TV, because the pixels would be bigger than the room. You need to sit a long way away from it to appreciate it.

Then there are the arguments that we are going to have about a family with five people who are all going to have their own TV and be streaming all the time. My daughter went to Japan and there was such a family—the son had not spoken to the mother for 20 years. Is that the society you are trying to create with all this internet? Or should we be solving the problems that really need fixing? When I download myself every day to go to work it takes an hour on one of Melbourne's linear car parks. Perhaps we should be spending a bit of thinking time on how to free up the freeways?

Senator LUDLAM: The public transport committee's hearing is in the next room.

Prof. Lowery : You are the government—you do everything, you are omnipotent.

Senator LUDLAM: Can we unpack some of this a little bit? I think all of you at different times have mentioned this concept of future proofing. I feel like you just downloaded a couple of quite different ideas on us there. Can we pull apart this future proofing idea? One of you, in your opening statement—I forget who it was—was talking about—tell me if I have got this wrong—a 10-terabyte-per-second network between research laboratories. Who was that?

Prof. Lowery : That is what we are doing in our individual labs at Melbourne, RMIT and Monash—

Senator LUDLAM: Ten terabytes?

Prof. Lowery : We want to send signals to each other so we can work out the real world problems like what happens when a tram goes past the fibre. It affects the polarisation in the fibre. Can we invent cheap methods of solving that at the receiver? We can do some really great stuff just using some dark fibre that is plentiful down St Kilda road. That is one thing. I think Rod's point has always been that if you put a fibre to the home you could get 10 terabytes down it. John Laws gets very excited about this when the Swiss put 26 terabytes per second down the fibre. But what would you do with it?

Senator LUDLAM: It is 2016. We have invited you guys to tell us a bit about what you would do it. I am always a bit uncomfortable when we are comparing potential alternative technology solutions and the metric that we use is how many 4K TVs people can have on at a time. I feel that that fundamentally misunderstands how these networks are going to be used. I do a certain amount of 3-D modelling and rendering, so I use a render farm in Switzerland, for example. You spend a lot of time tearing her out sending these big files up and waiting for them to come back. I do not have a number of 4K TV metrics by which to gauge that. The question I want to get to, when we come to this concept of future proofing, is how far can this legacy copper network take us? How far can it be pushed before we hit the capacity and say that we had better just get on with upgrading it. That is where the debate in Australia has been stuck. If you can help us unstick it, that would be great. Should we persist with a fibre-to-the-node and HFC hybrid build, or should we jump to what appears to be scalable and future proof?

Prof. Tucker : I agree with Prof Lowery's comment about copper being good in a home. There is a technology that we have not mentioned here yet, called fibre to the distribution point, using a standard known as G.fast. The idea there is that the fibre comes almost all the way to the home, but you do not dig up the front garden. There is a little box outside the house which provides a connection from the fibre to the copper. It is not powered by any external source. It does not require a connection to the power lines, which has been causing problems for nbn co recently, I read, because it is reverse powered from the home by the copper pairs. This is a very new technology which is just being trialled around the world. It is highly compatible with the concept of taking fibre almost the home. You could either use G.fast with the existing copper pairs under the front garden or you could put a fibre in if you wanted to, for those people who really did want a fibre. I think that is a really good way forward. There has been mention made of this by nbn co executives, but I am not sure whether it is in any of their business plans just yet.

To answer your question, Senator Ludlam, my feeling is that to go to fibre to the node is a bad idea because fibre to the node requires an expensive piece of equipment in the field powered by the power grid, which is causing problems right now, as we know. In fact, the fibre-to-the-node network is going to consume something like 20 megawatts of power as an additional drain on the Australian electricity supply, when you add all those nodes up. Then to upgraded it is necessary eventually to bring the fibre from that node through to either out the front of the house and use fibre to the distribution point, or to the house. My feeling is that fibre to the node is a bad idea, but there are solutions that do not require necessarily digging up the front garden of every house.

Prof. Lowery : That is brilliant. We need to think about some of these hybrid solutions that give the advantage of the fibre but also reduce the civil engineering costs.

Senator JOHNSTON: I am very interested in what you just said, Professor Tucker. Could you talk to me about vectoring, Vplus and G.fast as you have mentioned, which I think is quite embryonic but very exciting in some respects. Can we talk a bit about those three things. Can you tell us about how vectoring and Vplus is supposed to work, in line with what you said about G.fast?

Prof. Tucker : Vectoring is a finetuning, if you like, of what is known as VDSL—very high speed data transmission, which is an upgrade from today's ADSL. Vectoring is a finetuning of the technology that has been around for a while, which provides just a little bit of extra capability. The issue with ADSL and with fibre to the node is that a lot of those copper pairs that go from the node to the homes travel alongside each other in a cable, and there is interference from one to the next. It is that interference which is a major issue in terms of reducing the capacity of the copper to handle the data. The clever thing about vectoring is that the box in the street knows about all the data going down all those lines. When sending data to your house it knows what data is going to another house, and it is able to change them both of them a little to reduce that interference. So vectoring is a finetuning of VDSL that gives a bit of extra capacity out of copper, but it is about at the limit. The companies that produce vectoring—Alcatel-Lucent, which has recently taken over Nokia,— have pointed out that vectoring is about as far as you can go with copper. There is really no way forward in terms of making things any better than vectoring.

Senator JOHNSTON: And that accounts of Vplus 2, which I gather is Alcatel product?

Prof. Tucker : Yes.

Prof. Lowery : That is a great explanation, but the problem is that people say, 'Copper twisted pairs are terrible because they are at their limit.' But the copper twisted pair across your front garden is not. We have to be very careful in these discussions to not just say that it is copper versus silica; it is copper in a certain application over a certain distance with a certain number of channels.

Prof. Tucker : Precisely. To answer the second part of your question, the G.fast technology uses a much shorter length of copper, and very little of it, maybe only a small piece of it, is travelling next to your neighbour's. Most of it is by itself in your front garden, so it is not experiencing interference from other signals. In addition to that, it is quite short, so there are fewer losses and other imperfections that affect the capability. So it means that with not very sophisticated electronics in that little box that goes on the street, you can get up to a gigabit from that fibre in the street along the short length of copper into the home. So it is much, much better than you could ever get with VDSL.

Dr Gregory : If I may, I would like to just correct a couple of things that my colleagues have just said. Professor Lowery may have been in another country for the last 30 years, unfortunately, but the idea of digging up front lawns has been done in this country before. We rolled out gas to every home in the country, which required a dirty great big gas pipe into homes.

Senator JOHNSTON: Not every home. There are still a lot places that do not have gas.

Dr Gregory : That is unfortunate. The point is that in other countries around the world putting the fibre up the existing infrastructure is not being done. The fibre is being rolled into houses, around fences, over concrete work and so forth using self-installed kits and sheathed fibre. People need to look at this. There are three cities in Europe where the people in the city is actually rolled the fibre themselves out of their homes into the street. I agree with my colleague on the point of fibre to the distribution point. I have written about it in the past. I proposed it in 2013 as an alternative to fibre to the node. Then it could be left up to the homeowner either to use the existing copper, if they wanted to, or to use fibre into the home. Digging up people's front lawns and so forth is a nonsense that is not sensible in this discussion, simply because the gas pipe explanation for the nation's benefit of using gas rather than electricity is justification for doing that.

If I can go back to the self-install kits, there is no reason why people cannot self-install the fibre from their home to the street. It has been done overseas. Australians are very competent and capable people; we can do it too. It would make the process of actually rolling out this network much faster. I think that Australians would get on board if they were given the opportunity, but it means you have got to get the fibre past the homes. G.fast is a valid alternative to what is being done now. My argument has been, and always will be, that fibre to the node is unjustified. G.fast is justified and fibre to premises is justified. My belief is that if you put fibre past houses and you say you can either do G.fast or self-install the fibre, 99 per cent of Australians will self-install the fibre.

Senator LUDLAM: By self-install, do we mean ring somebody with a van to come around and do it or do you mean get out there with a trenching kit and do it yourself?

Dr Gregory : You do not need to trench. You can do it yourself.

Senator LUDLAM: I can barely get my router to work some days. What do we mean by self-install?

Dr Gregory : It simply means that you ask your neighbours, if you have got two left thumbs, to help, which is what they did in Europe, or you ring a tradesperson to do it or you do it yourself. You do not need to trench to self-install fibre.

It is like the electric gates that everyone has in front of their houses now. You have got a cable from the electric gate running back to your house. If you look at how most of these are installed, they are installed around the fence and they are normally laid just a couple of inches below the top, around the fence—

Senator LUDLAM: So like installing a bit a retic piping?

Prof. Lowery : That sounds like a roof insulation type disaster. You have to touch electric cables and conduits.

Dr Gregory : No, that is the standard industry way, Professor Lowery.

Senator LUDLAM: I do not think that the cable is going to explode into flames.

Prof. Tucker : One important point to make here is that the current wiring rules forbid the public to make connections using the copper pairs. The reason being that if they make a mistake and connect to a high voltage—the high voltage can go into the system and cause damage elsewhere. But, fibre—

Dr Gregory : No rules.

Prof. Tucker : There is no problem with fibre, it is just glass. There are no rules and you can do it yourself.

Dr Gregory : Yes, absolutely.

Prof. Lowery : But certainly on your electric gate, that is a highly dangerous situation because when the wind blows the fence down it rips the wires out of the connection box and you have got live wires in the front garden.

Dr Gregory : They are low voltage—12 volts, Professor Lowery.

Senator LUDLAM: Let's come back to the NBN.

Dr Gregory : You obviously do not have an electric fence.

Prof. Lowery : That is why my neighbour's gate never opens.

Senator LUDLAM: Let's come back to the NBN.

CHAIR: I think we have canvassed global comparisons and I think we have gone to fibre to the distribution point. If you then moved to fibre to the distribution point, how future proofed is that?

Prof. Tucker : It depends on exactly how it is installed but it can be quite future proofed because fibre to the distribution point basically means that the fibre is very close to the home. It may require a little bit of extra fibre to the front fence and then the self-installation process. It is quite future proofed in the sense that there is not much extra effort to go to fibre to the premises if necessary.

CHAIR: I am moving to Dr Glance so I can talk about health for a moment. We will want, for someone who is ageing in their own home, for them to have capacity to talk through a television to their nurse, who may be in the same town or not, to have sensors so that people know that people are moving—all the technology that is going to be able to provide very good quality monitoring and care of a person living by themselves in their own home. What capacity is needed for all of those bits and the bits that we have not even invented yet to ensure that someone can live in their own home?

Prof. Nirmalathas : At the institute we play with a number of futuristic applications in the area of aged health care kind of things. We often find that a lot of people use 4K video as an example of showing what the capability of the network is. What we are finding is that latency, the time it takes to get a transaction done over the network, is starting to be the major limitation in terms of user experience and, I believe, to deliver a serious service. For example, we are trialling a virtual reality as a basis for helping quadriplegics to have a very systematic way of rehearsing their lung exercises because if they stop doing breathing exercises they have a very quick downward direction in terms of health because if the lungs fails, the heart fails. It is very quick. We are using that as a way of helping them to socialise with people who have similar things. Having music online itself requires significant latency restrictions. Providing bandwidth, in a way, deals with that issue.

It is the same with many other healthcare things. As I illustrated earlier on, we click and wait—and you cannot do it in that context. In the case of paraplegics, it has a very unfair distribution in the regional towns—because a lot of people use roads and they have a lot of accidents as a result of it. They are the types of services we want to do because it would significantly alter the footprint of health care or aged care in the future. When we look at serious applications on the back of the NBN, we start to see significant expectations on bandwidth and, more importantly, the latency—how long it takes to deliver that bandwidth.

We are fixated on the architecture and the technology options rather than saying 'Here is a gold standard that we would like to deliver to Australia in terms of connectivity and bandwidth' and letting decisions be made by city and by area on what is the proper technology to deliver it. That would probably see a lot of the politics pulled out and would have serious goalposts in terms of deployment. After the strategic review and after we accepted that we are going to do not fibre to the premises but HFC, I did not see a step-change in terms of the deployment plan. In fact, it has downgraded the projection of the NBN. So we are actually delaying Australians' chance to have access to broadband; we have even relaxed the architectural options and the bandwidth they have to offer. That is the game that we have to lift. We have to go back to what we promised. We have to find ways of accelerating, on average, what most Australians should have. And we should reduce the gap between peak and average speed of connectivity throughout Australia regardless of the individual architectural and technology choices.

Once we focus on the types of services we are going to deliver—we are going to rely on the NBN to deliver mission-critical services, we are going to rely on the NBN to deliver government health care to regional cities, we are going to rely on the NBN to activate regional economic and industry development using the IOT—then we need to be fixated on making sure that, as quickly as possible, we achieve the original goals of delivering NBN. We should move away from fixating on architecture because it changes from time to time. NTT figured out that fibre to the premises is the way to outpace incoming competitors when I was there in 2004. AT&T we will ask later. There are population density changes from country to country, city to city, region to region and village to village. It actually complicates that decision. If you are fixated on only future proofing, fibre to the premises wins—and other derivatives. But, as a country, if you say these are the metrics the NBN should deliver—this is the average connection speed that we aim to do and this will lift our average ranking—then that is what it is. Then the technology, the architecture and the delivery should be geared towards delivering that goal.

Dr Glance : One of the issues when we talk about capacity and speed is that we talk about it in the ideal setting of an individual sitting in their home unaffected by what everyone else is doing. The truth is that that is not going to be the scenario we are going to see. In the remote communities people are using satellite. So there is contention. There is the fact that you are competing with somebody whilst you are doing a videoconference. I know that there are mechanisms to try and get around this with other types of use of the network. If everybody is using video, telehealth and other things, and their house is using the internet of things to monitor whether somebody has fallen over, then it is not just one isolated house doing that, it is a whole neighbourhood. These are not necessarily futuristic things. I realise that companies use this as a marketing ploy, but 2020 is only four years away. We will see the wide-scale deployment of the Internet of Things and home monitoring. We have to. There are no alternatives to that given the current demands on the health system and other infrastructure. So the point is that it is not just about an isolated person sitting in their home. I might currently have a 100-megabit connection through Telstra BigPond, but that changes depending on what other people are doing in my street.

CHAIR: What worries me is that if we move to having people's lives dependent on a piece of technology, if it fails, the implication is not, 'Oh, I can't watch the telly,' it is bigger than that. I live in regional Queensland. I am a late convert to therapy over the internet. When we were in government we had a little program that provided therapy services for children in regional Queensland who had hearing and sight impairment. When I saw that therapy being delivered—from a therapist in Mount Isa to a little girl sitting in Townsville—I became a convert. That is a good thing, but we have to be sure that it is going to work. If I have got my dad in a house in a country town I want to know that the system that is looking after him is foolproof. So that is where I am up to in a conversation about telehealth. I have been on Saibai Island in the Torres Strait, the closest place to New Guinea, where there is a connection so that a person can get an X-ray of their chest to find out if they have tuberculosis and it can be sent to Brisbane. That is a good thing; when it does not work, that is a bad thing.

Dr Glance : Absolutely. Again, these are scenarios from the past 10 years, not future scenarios. For example, we can look at radio imaging. In WA we were talking at about having a centralised scheme of having all radiological images on a central platform to be able to distribute to hospitals or anybody who wanted them and from whichever provider. Those are the types of scenarios that you want to be able to deliver today, but you cannot do that on the current infrastructure. You want to be able to do that now, not in five years time. The images themselves are extremely large files. Even doing that over the network proved problematic within the new hospitals in WA. So there are real problems today in achieving things that we should really already be accepting as the norm.

Prof. Lowery : What I hear in the debate a lot is: let's think about the maximum possible use that someone might have of this network and then multiply it by the number of premises in Australia. It is clear from the hospital example that that is the wrong way of thinking—because hospitals do not move. MRI machines do not move; they are huge things and they need a lot of infrastructure. Radiology and path labs do not move; they all sit together in a big building—I have got friends in them. So that is where you need a link from A to B, or from A to B to C, or a special dedicated network for that, which you can buy from people who have dark fibre. It does not mean that everybody at home might one day get an MRI machine in their front lounge room.

I went for a job interview in 1986 and, for health monitoring, there was a thing you wore around your neck; you pressed a button if you had fallen over. That worked over the old telephone system. It does not require many bits per second to say whether someone is horizontal or vertical. So I think there have got to be some good metrics. Things have got to be fit for use. You have got to decide who needs what, who is in a fixed position and who might pop up.

I was talking to some people at the table next to me last night. They have a publishing business and they were complaining that, when they download a 500-page textbook, with all the beautiful graphics in it, it takes 10 minutes. Well, maybe they need to pay a little bit more than the rest of us. Maybe it ought to be like Uber: if everybody wants to download their movie at exactly the same time, they have to wait a bit. But, with a bit of forethought, they could plan downloading the movie—they could have a cheap night rate and that sort of thing. So I agree that shared systems are a bit problematic. I would love the Monash Freeway to come straight to my driveway and front door. That would be great—it wouldn't really! But, in reality, the amount of concrete used would mean that there would be no room for anyone's house. It is a bit the same with the NBN: you can have this dream of providing enormous amounts of bandwidth to everybody, every second, on demand, on whim, but it costs an awful lot of money and it takes an awful lot of time.

Dr Gregory : Does it?

Senator LUDLAM: It is costing them more to retrofit using HFC and fibre to the node. So I kind of don't buy that.

Prof. Lowery : It is costing much less because it is installed. Why rip up something that works?

Senator LUDLAM: It is not.

Dr Gregory : It is not costing much less.

Prof. Lowery : You are wrong. You needn't install stuff.

Senator LUDLAM: We are up to $56 billion to patch this thing together that is underway at the moment, which is actually more than Mr Quigley and the original NBN had estimated it was going to cost to put fibre to the premises to 93 per cent of the country.

Prof. Lowery : Wasn't there a report with a maintenance cost between $300 million and $700 million a year for the copper network and $200 million to $300 million a year for the fibre network?

Dr Gregory : That is wrong. Telstra admitted it was close to $1 billion for the copper now.

CHAIR: Can we just unpack this cost question.

Dr Gregory : Could I just address medical for a minute. The important thing is that we are all too focused on the old things—the old medical alarms and so forth. That is disappointing because what we are talking about is the future—and the NBN is the future. We are talking about therapy, we are talking about virtual reality, we are talking about being able to do more—terabit per second virtual reality systems where people with PTSD or some other type of problem are able to do in-home therapy. We need to move beyond thinking about systems that were connected to telephone lines. There are so many possible ways that we can use the NBN for health and education—in particular, health. There are systems being developed now for group therapy and so forth in a virtual reality world and so on. It would be good for people in remote Australia to be able to participate in these systems.

I would like to go back to something I forgot to say, and that is that we need to order a third satellite immediately. They take six years to get into the air. We need a third one ordered now. It is very evident to most of us by now that those two satellites are going to be heavily congested. I have strongly argued and strongly agree that we should be using those satellites to facilitate connectivity to planes, trains and boats. That is for two reasons—a person might be sitting on a plane looking at YouTube but, equally, they might be working. For a person on a boat, it is their place of work. For a person on a train or a truck, it is their place of work. They can have health and other problems and they need to talk to their families.

Talking about regional areas, I used to live on the De Grey River—and that is a place to find on a map! And, once you have found it, you will realise that it is about as far as you can get from anywhere. We had nothing. If someone got sick, that was it, that was the end of it. So I can only strongly stress that we need more satellite. But we also need to think about the future. These things are not a dream; these things are being developed today. I just wish the debate would move past 4K TV and medical monitoring systems that we have had for 50 years.

Senator JOHNSTON: In all of our discussions—and we have all got a view—when we talk about cost, I think we have to start from the position of a benchmark. We all see a data rate that we all, I hope, individually feel is adequate for each premise. I would like to discuss what the benchmark is and what we think the benchmark should be as of today—we can pontificate about the future later—and what the cost of that should be. Someone suggested to me it should be 53 megabits per second—I think that is the figure from you, Professor Tucker. What should we reasonably expect to pay for that? Let us look at the international benchmarks—and we have just discussed this with Dr Gregory. We talk about, particularly in Aboriginal communities, wanting health facilities. It is very expensive to the middle of Western Australia, I can tell you from personal experience. I take it, Dr Gregory, it was Goldsworthy you were living in on the De Grey River.

Dr Gregory : No, I was living on a station called Warralong.

Senator JOHNSTON: I know it well.

Dr Gregory : Not far from Goldsworthy.

Senator JOHNSTON: So the point about cost is: how much comparison do we make from a demographic point of view and a geography point of view from Australia? I think again, like always, we are pretty unique. There are the American, European, South Korean and Japanese examples—and I can virtually walk around some of these countries. We have a huge differential in terms of distance, services, electricity et cetera. What is a fair thing for our demographic in terms of data rate and what do we think is a fair cost for that?

Prof. Tucker : I would actually disagree, with respect, Senator, about the need to set a benchmark for today. I think that, because the world is constantly changing and because there is constant growth in new technologies emerging and constant growth in the amount of data that is being generated, used and stored, to set a benchmark at a particular point in time is perhaps not the right way to go about it. We need to think about what the growth needs to be and where we need to be in the future. If we set a benchmark for today and set up technologies that will provide it, we might end up having stranded assets that cannot be upgraded later on.

Senator JOHNSTON: Professor, tell me how we are supposed to budget for that.

Prof. Tucker : I think what we need to do is think over the longer term. We need to think about what we want in 2020, 2025—

Senator JOHNSTON: I have got to plan. I have got the forward estimates. I have got to have some—

Senator CONROY: You do it once, you do it right and you do it with fibre.

Prof. Tucker : Thank you, Stephen. He is there after all. He has not gone to sleep. Of course when we talk about costs we are talking about peak funding typically. There is capital cost and then there is peak funding, and the peak funding is the maximum amount you spend before the revenue starts to kick in and bring things back to a point where you are actually making a profit. It is a long-term issue and we need to think about the peak funding and how it pans out over a long period of time.

As new services come into play it is likely that there will be increased revenues through new services that will generate revenue streams that we have not thought of today. If you want to think about a benchmark, there have been a number of studies—and I really have not got time to go into it now—about what bandwidth people need. One of them was done for the Vertigan report by Vertigan in 2014. There was another one, which I have referred to in some papers I have written recently, by van der Vorst, a group in Europe—

Senator CONROY: A Dutch university did a good paper.

Prof. Tucker : That is the one, Senator Conroy. It is the technical university of Eindhoven. Their projections for future demand were more than an order of magnitude higher than what the Vertigan report had projected. In fact, the Vertigan report had projected bandwidth requirements which are less than what we have today, so it was really quite questionable whether that was appropriate or not.

There are graphs from these studies done by groups of academics, industry people and sociologists that have looked at future demand in growth. They have come up with these graphs. Typical numbers would be for 2020, about the time that the NBN will be completed, more than 100 megabits will be required in a typical European country, and of course fibre to the node cannot provide that. So the danger here is that we are building a network which might be good today in 2016 but by 2020 and certainly by 2025 will not be—

Senator JOHNSTON: So 100 megabits per premise?

Prof. Tucker : Yes.

Senator JOHNSTON: Okay. What do we get for that?

Prof. Tucker : What do we get for that?

Senator JOHNSTON: What does that enable us to do house by house?

Prof. Tucker : It will enable us to expand our access to the kinds of services we have talked about. Not everyone will use the full 100 megabits, but it means that it will be available for those who need it. We are talking about premises here, so it is businesses as well as homes. Small businesses will need it for uploading their 3D designs or whatever. There will be those kinds of capabilities available.

Senator LUDLAM: Isn't it the case that not every house uses the maximum electricity that the electricity distribution network provides, but you had better install it?

Prof. Tucker : We do not provide to each house wire which is just big enough for the average amount of electricity for a house. We have electricity wires that are big enough for the maximum amount.

CHAIR: I have a question around the cost of retrofitting. Rather than talking about it, is there somewhere you could point to where we could find the cost of retrofitting fibre to the node to get the 100 megabits that you are talking about?

Prof. Tucker : To upgrade fibre to the node, you need to take the fibres all the way to the home. The cost is the same as it would have been if you had done in the first place—

CHAIR: You have answered my question.

Prof. Tucker : and you have to throw away that box in the street that costs a lot of money; it is wasted.

CHAIR: Thank you, Professor.

Senator LUDLAM: We only have you gentlemen for another 10 minutes or so. You are in the position of steering policy. Given where we are in the electoral cycle, given where the politics are and given the stage of the network build and different kinds of technologies, what would your advice be to the government, if you got to choose which way to steer this ship from here? Maybe we could just hear briefly from each of you—and keep it as nuggety as you can—on what we should do with this network. What would be the smartest thing that we could do from here?

Prof. Lowery : We would keep the HFC. We would patch it up a bit where it is working. In the future, we could go to fibre. I like the G.Fast idea. I think meeting at the front gate is a good idea, like the water board does with me. I think we should concentrate on areas of the country that do not have reliable internet. Reliability is a key thing. You do not want it to be 95 per cent, as I saw in one report. You want it to be 99.999 per cent or whatever. I would use money prudently. In order to do this, I would extend the time frame of the NBN. This will give longevity for jobs, which is important in people's lives. I do not see that there is a huge problem in spending money on maintaining fittings. It keeps people in work. It keeps the network looked after and monitored. So less speed, please!

Senator LUDLAM: Less speed?

Prof. Lowery : Less speed in the rollout, but not less speed to the customer. There should be sensible use of the Net with probably some differential fees on charging for different services and software that guarantees that essential services always get priority. So there is some clever stuff.

Prof. Nimalathas : I think we should definitely make sure the greenfield sites are ready for fibre to the premises. That will prevent us from converting today's greenfield sites into brownfield sites. Brownfield has a widening gap in terms of what is needed to be deployed but is not currently deployed. I have not seen any real progress in the increase in activating NBN because of the urgency that is part of the mix. I am not that convinced that HFC is actually changing the deployment plan, but anything to accelerate the deployment would be good. There should be a third satellite, but with the new technology of flexible beams. At the moment, fixed beams are there, and two satellites have gone in. If there were flexible beam technology, it could add flexible bandwidth to areas and regions where there is a major issue or a major bandwidth requirement you can flexibly allocate. They are probably the three things that I would mention.

Prof. Tucker : My message is simple, and it is to cancel fibre to the node as soon as possible and replace it with fibre to the premises. We did not get into much detail about the costs, but I would like to correct my colleague on this side of the table—Professor Lowery—and point out that, if we learn from the New Zealanders and work out cheaper ways to install fibre to the premises using more modern techniques than are being used at the moment, the cost calculations turn out that a fibre-to-the-premises network would not cost much more than fibre to the node. I am also in favour of a third satellite, especially now that Qantas is going to start streaming videos on Qantas flights, sapping out the bandwidth from the NBN!

Dr Glance : I would agree with Professor Tucker on cancelling the situation and responding to the current economic demands that we find ourselves in. All of the decisions that were made even two years ago are no longer the case. We have an economy which is in a different situation, we have a Prime Minister asking us to innovate and we have technologies like Netflix and other companies coming in and changing the landscape about how we use those technologies. If we are going to be expected to do that as an economy, as a public, then we absolutely need the infrastructure to enable that. That may not have been justified five years ago or even two years ago; it is certainly justified now. We should follow the example of private enterprise in the case of AT&T in other countries and change the plan to fit the needs that we see today, in the current context, and in the next five to 10 years.

Dr Gregory : I would reiterate the need for a third satellite. It is very important because of the time lag. I believe that investment in the satellites has been extremely important for regional and remote Australia. I again support fully fibre to the premises. I believe that it is possible to get the build cost down significantly with approaches that I have seen around the world. As an alternative, I would fall back on fibre to the distribution point. I would cancel fibre to the node as quickly as possible. That would mean yesterday. I would immediately launch a broad committee to review what is to happen to the NBN after 2020 and also to look at what legislation and regulation associated with that would improve the competitive environment in Australia for the telecommunications market. I consider the current telecommunications market to be an absolute mess. I believe that we need a broad based review to achieve some sort of working way forward.

What we have been hedging around is: what is the purpose of the NBN? I would like to conclude by reminding us all about the NBN's purposes. We are spending all this money. The purpose of the NBN is not to simply provide people with bits of cable so that they can watch movies. The purpose of the NBN is to take us beyond universal service and take us to what should be known as universal access. We have now moved into an era where every Australian needs to be connected to the digital network reliably all of the time, every day. The problem that we have today is that a large number of Australians cannot afford it or they have no fixed address; they are itinerant, they are homeless, they are the poor, they are the socially and economically disadvantaged; they live in regional and remote areas; and they live in a shack. I believe that the reason for the NBN—my reason for the NBN—is that we need to define universal access as the goal. Speed and all the rest of it is one issue, and I agree with that, but the goal of the Australian government has to be to provide universal access to all Australians to the digital world, for health, for education and to access government services and so forth. Wi-Fi is a ubiquitous thing that many homeless and itinerant people have. They cannot afford mobile cellular, but they have a device that will do Wi-Fi. Nbn co should be asked to roll out a national Wi-Fi network, a national wholesale Wi-Fi network, immediately and, as part of that, councils, local governments and state and federal governments will be able to facilitate free Wi-Fi access for the homeless and the disadvantaged—the socially and economically disadvantaged. I feel that we are missing the point of the NBN debate.

CHAIR: Gentlemen, that was a very informative session. I very much thank you all for your contributions. I also thank you for the way you have interacted. That has been terrific. I really fear for the Hansard reporters having to work out some of the language that we have used, so you may be contacted by our people trying to understand what a terabyte might be and how to spell it. That is an easy one, I think! Thank you very much for your contribution today.

Proceedings suspended from 10:30 to 10:46