In his independent audit of the public policy process behind the national broadband network, former Telstra director Bill Scales suggests there was never an internal debate about different technology options for the NBN.
Picture: Getty Images
He also argues the panel of experts (of which I was a member) assisting the Rudd government did not properly test advice from the Australian Competition and Consumer Commission (ACCC) about the upgradeability of a Fibre-to-the-Node (FTTN) network to a Fibre-to-the-Premises (FTTP) network, and that the panel inappropriately relied heavily on this advice in making recommendations to the government about the development of the NBN.
In my view, all of these assertions are incorrect, and this taints the credibility of the audit.
In reality, the panel spent many hours discussing and analysing the technology options and the upgrade paths, including those set out in the various proposals submitted by companies in response to the government’s Request for Proposals (RFP). The panel also independently evaluated other models for upgrades.
The panel, which included telecommunications experts from both industry and academia, carefully scrutinised all advice it received, and drew heavily on its combined experience. Some advice was very helpful and some was considered to be of little value.
When the ACCC tabled its advice regarding the costs of upgrading FTTN to FTTP, it came as no surprise to the expert panel that the ACCC had a view that was similar to what the panel had already concluded. Simply put the panel did not rely heavily on the ACCC advice.
A fundamental flaw with the audit process was that Scales, by his own admission, did not have access to key information, with limited access to documents associated with the panel of experts’ activities.
Members of the panel, constrained by strict confidentiality rules, were also unable to share any further information with Scales about the details of panel discussions and deliberations. Without divulging any details, I was able to explain the situation regarding the ACCC to Bill Scales when he interviewed me for his report, but it seems he did not put much weight on my comments.
Scales seems to have missed the point that upgradeability was an important issue from the outset of the RFP process. The RFP for NBN Mark I included many requests for information from proponents about how they proposed to future-poof their networks by providing upgrade paths from FTTN to FTTP. It even suggested a 2020 time frame for this upgrade. The consideration of upgradeability and its costs was one of a number of factors that fed into the “value for money” criterion for evaluating the proposals.
In one part of his report, Scales suggests the ACCC ignored a key report by Analysys Mason, which shows that the cost of FTTP is around five times the cost of FTTN. But a little bit of information can be dangerous.
What Scales seems to have missed here is that the Analysys Mason report was based on the British broadband landscape, and assumed that the copper network required for a FTTN rollout was already owned by the operator rolling out the network. When parts of the copper network and/or the associated ducts might need to be purchased or leased, as required for Australia’s NBN, the calculus changes completely.
The Coalition’s NBN is proving expensive
[related title=”NBN 2014″ tag=”nbn2014″ items=”4″]Under the previous Labor government, NBN Co negotiated a deal with Telstra to provide access to Telstra’s ducts and pits, enabling fibre to be laid with fewer trenches needing to be dug. The cost to NBN Co of this access was $11 billion, and added a hefty overhead to the total cost of the NBN.
Even if we make the optimistic assumption that NBN Co will not have to pay any more than this $11 billion to access and maintain Telstra’s copper wires inside those ducts, the Coalition’s (predominantly FTTN) NBN will suffer the same $11 billion overhead. Consequently, the factor of five difference between the cost of FTTP and FTTN given in the Analysys Mason report drops to something around two or three. By world standards, the Coalition’s NBN is turning out to be very expensive.
What has Australia lost in the transition from Labor’s FTTP NBN to the Coalition’s multi-technology mix, with a large proportion of FTTP? In my view, a big danger is that future upgrades from FTTN to FTTP will be slow and that Australia will continue to lag behind the rest of the developed world in terms of broadband access.
With such a huge investment sunk in the Coalition’s multi-technology mix NBN, there is going to be very little appetite in government to fund future expensive upgrades to FTTP. If the government sells the NBN to a commercial monopoly operator, there could be even less incentive for upgrades to FTTP. There is a real danger that Australia will remain trapped in a broadband backwater.
Rod Tucker is Laureate Emeritus Professor at University of Melbourne. He receives funding from the ARC and has received funding and in-kind support from a number of telecommunications companies. He served on the Rudd Governent’s Panel of Experts that assisted with the development of the NBN.
This article was originally published on The Conversation. Read the original article.
Comments
33 responses to “The Argument For Fibre-To-The-Premises NBN Isn’t Over Yet”
Its probably too late to turn this boat around and go back to FTTP now anyways.
This whole thing makes me so angry. High speed internet is a crucial part the economy and having Australia lag behind the rest of the world seriously hampers us on so many levels.
Asshats.
Lag. I see what you did there.
Not too late yet, FTTN are only in trials, Telstra have signed to renegotiate the $11 billion deal but it isn’t a definitive agreement so even if we do end up paying more, potentially the copper cable can be used to pull through fibre (it’s not currently).
But everyone and i mean EVERYONE needs to make themselves heard, im not just talkin about a couple of hundred thousand on change.org i mean tens of MILLIONS.
Track record with the libs is they don’t handle technology too well, and act in the interests of their masters (dtv-leave no room for other broadcasters with a triple broadcasting policy)- a classic example, Kerry packer was calling the shots with that IMHO…and we are now at a point where we have to consider another switch (change codec away from mpeg2 to mpeg4 to accommodate HDTV streams as sdtv is dead) . That being said, here we are, at a cornerstone turning point for the nation, where a communications infrastructure is being held back due to political ideology (coincidentally which will be the broadcast medium for what’s left of television as we know it – I’m sure someone will come up with some good uses for the spectrum that’s currently being used by tv).
And we know where we are going to be put. Behind the rest of the world.
What I find incredible is that everyone is telling the govt that they have it wrong, they are so overtly and publicly controlled by private interests…and yet there is no accountability. Senate committees are ignored.
I think it says a lot about our process of government today. Hopefully some fresh blood will one day introduce some accountability into our political system in the future.
Has anyone actually seen one of these Nodes on any street yet..? If not then it should be a no-brainer for the coalition to swallow their collective pride, and just connect to the friggin home..!
Not going to happen, ideology trumps logic in politics.
unfortunately -_-
you mean logic, math and truth.
This article payed for in part by the Australian Government.
You did actually READ the article.. right? He essentially criticised the current (coalition) government plan… and pointed out issues with the “independent” by former Telstra director Bill Scales…
I have no idea why the coalition won, I though Labor wining was going to be a sure thing. To this day everyone I know had voted for labor. The only person I know who voted for liberal was one of my friends mum on the basis that “they favour people that work!” when I asked her what she thought about the NBN “What’s that?”
fucking old generation ruining it for us all.
Buy all the houses and now screw over our what was going to be future proof fibre connection.
Rant..?
As for “fucking old generation ruining it for us all” Fuck You man..!
They won because they got 60% of the vote. But in reality mate no one wins elections, the sitting government loses them. With 7 years of the crap with musical Prime Ministers, massive debt people had enough. Same happened up here in QLD. Newman (LIB) didn’t win Bligh (ALP) just lost. Same will happen with Abbott and then the same with his replacement. Until this current generation gets out of politics, and a younger group enter (I mean under 40 now) then it will be much of this old school policies.
LOL. The fact that you only know people that voted Labor says more about you than anything else.
They won because the nationalists defaulted back to the libs and everyone treated the election like a popularity contest since the MSM (Murdochian Machine) treated the labor oustings like front page news:
http://www.glennmurray.com.au/wp-content/uploads/2014/05/Anti-rudd.png
Despite the fact that Phoney Tony and his merry band of miscreants did the same thing to Malcolm Turnbull. They made it about representation rather then who had the best economic policy:
self-selection bias.
I tend to hang around people who share my values and attitudes, so it makes sense we tend to vote in similar ways.
I think our system of preferences has a lot to answer for.
I’m still working on the NBN – I can confirm there are 9-trial FTTN sites atm. Some on VIC some in NSW.
I can confirm that the industry is STILL shedding staff as NBN is not releasing any volumes of work.
I can confirm that NBN changed their management group on the 4th of July and the new group are tearing the shit out of the older NBN structure.
I can further confirm that we are in the process of designing further ‘trials’ — this should alarm everybody. We have had a new government for a year and we are still talking about designing trials.
I can also confirm the NBN section of the company has been quarantined from the other ongoing works we do as it appears NBNCo. is being slowly asphyxiated and I predict it will be closed down by Liberals and the profitable areas (urban) will be auctioned off to the private sector once the Taxpayer completes the Transit backhauls required to connect the profitable urban centres.
The only people that will remember the NBN are us workers, and the tech-savvy Australians that understand this is like NOT building the Snowy Mountain Hydro scheme in the 50’s.
No engineer worth his salt would ‘prefer’ FTTN and would also understand the complications in converting FTTN to FTTP as being so ridiculous that you leap-frog FTTN to save money in the longer term.
I fucking hate my job these days – no satisfaction at all.
“‘I’m still working on the NBN – I can confirm there are 9-trial FTTN sites atm. Some on VIC some in NSW.”
My concern is that they’re going to provide the trials in areas with either good quality copper, so they’ll get high speeds unmatched by real world testing elsewhere, or with little connectivity to begin with, so they can say “look, we increased their speed from 1.2mbps to 12mbps. 900% increase, woo!”
Also worried that FTTN trials aren’t far away from my house… because if we get that, there will be no FTTP for me 🙁
Your assumption is quite accurate – the existing trials are on shiny new pristine copper and really close to the exchanges.
So – “best-case-scenario”.
Another issue: telstra charges NBNCo ‘rent’ for rack space in exchanges – guess what just went up enormously in price? the rent on that space – now NBN are freaking out on the rent-bill across the country over 20+ years with inflation etc.
Telstra shares are not a bad option atm
That has to be a frustrating position to be in, i feel for you
This was always the frustrating thing for me. It wasn’t just the Coalition being shortsighted – it was their ability to convince the majority of voters that their shortsighted views wouldn’t cost us more in the long run.
By the time their FTTN plan is properly kicked off, we’re already well behind in comparison to other developed countries. AND they want to be known for being a government focused on “infrastructure” and promoting new jobs… only they keep hamstringing every bit of future-proofing presented
We already are behind.
As most of you all know – the volume of data on a specific web-page is designed to be delivered at average speeds so we can stream and surf. This generally includes, ads, photos, gifs, video links etc. etc.
I don’t know about you guys – but I am already experiencing being 10years behind the global average and I notice it – Youtube used to be able to cache the whole video, now it only caches 10% ahead of where u r at and on a shitty connection, you are guaranteed stuttering. (I still pirate due to this crap).
Actually there is a way round that 🙂
http://lifehacker.com/preload-entire-youtube-videos-by-disabling-dash-playbac-1186454034
Good top – I was unawares (work-around exhaustion)
Keep us informed mate, we all appreciate it, Thanks 🙂
“Could FttN replace FttP?
http://www.abc.net.au/technology/articles/2012/07/26/3554003.htm
http://www.abc.net.au/technology/articles/2013/03/26/3723953.htm
When you do the sums, the bottom line is that FttN architecture:
— isn’t less expensive to roll out (in Australia);
— will take just as long to deliver;
— is more prone to failures and service issues;
— costs more to run;
— provides an inferior grade of service, and
— has no realistic, long-term potential for an upgrade path to better-performing technologies.
the argument is over, regardless of which is the best way to proceed we’re going to get *nothing*. not fibre to the node, not fibre to the premises, NOTHING. the only people who will get hooked up are those who were already going to get hooked up before the change in government.
Labour shouldn’t have gotten the government involved, instead they should have enforced the ISP’s to roll out fibre connection to supply better competition for the consumer.
However FTTN is competently fine. One of the only benefit’s fibre offers is the lack of data lost OVER distance. meaning short lines of fibre will not supply any noticeable benefit compared to copper. There is a fibre optic cable running form Melbourne to Sydney as data run on copper over that distance would lose a fair amount of data. Copper itself is more than fast enough to transfer speeds of even 1tb/s the real issue is the internet exchange. The whole reason you can get 1tb/s on fibre or copper is because the technology of the exchange, not the actual cable.
Honestly FTTN is completely fine, and if anyone is really to blame this nonsense its Labour. Remember when Labour made an incorrect assumption about the ownership and cost to access existing infrastructure? I do.
Firstly on point of fact this was never about JUST taking over from the ISP’s. This was about realizing the privatization of the industry (by the genius libs) was a terrible mistake in hindsight (one which they admit themselves) and upgrading the network while AT THE SAME TIME creating structural reform.
If you want to run a utility (which by definition should be available to anyone in the civilized world), you dont privatize the monopoly that owns 90% of the asset enabling them to gouge other ISP’s and consumers while neglecting CAPEX/OPEX… which is exactly what happened over the course of the past 20 years and something that has only started to be corrected with the introduction of the NBN… evidence? Take a look at iiNet’s revenues…
===========================================
Oh really would you care to elaborate on the other benefits besides the distance vs attenuation argument? Never mind ill do it for you…
A – Bandwidth
There is no question that fibre beats copper, true you can squeeze comparable bandwidth out of copper networks if they are being compared to a passive optical network (PON) 1Gbps. Against an active optical network with wave division multiplexing (WDM) however it has no chance, get copper to beat 43Tbps…
http://www.extremetech.com/computing/187258-43tbps-over-a-single-fiber-worlds-fastest-network-would-let-you-download-a-movie-in-0-2-milliseconds
People at this point usually start bitching and moaning about the fact that we dont need that much bandwidth today and even SSD’s cant read/write that fast…
To these people i say:
– Even if we don’t need it today how do you know we won’t need it tomorrow? When you build infrastructure you build it to cater to all possible future demand. To give you an idea, even blu-ray movies you buy are not raw they are limited to 59Mbps (doubling the minimum bitrate election promise that cant be guaranteed anymore), but the actual bit rate required to view a native 1080p clip is 1423.83Mbps. https://www.youtube.com/watch?v=qbGQBT2Vwvc
This tech has been around for nigh on 6-7 years at least, what happens when we move to 4k or even holography? What happens when you have multiple 4k or holographic devices in a single dwelling? Contention will dictate that you must have increased bandwidth to deal with additional users… which wont be available over FTTN…
– It is also in part why the decision was made to only implement a PON not an AON. Yes we dont actually need 43Tbps, #drool much less 1Gbps right now but as stated before this point is irrelevant as when you build infrastructure you build it to cater to all possible future demand. Why? As Mr Turnbull and many other people have recognized, getting the fibre from a GPON cabinet to the premises is the most expensive part of the deal and as such once a solution is implemented it will not see significant upgrades for some time due to the scale and cost.
Evidence?… look at the copper network… it’s been around for 100 years… and FTTP may be expensive in financials, true, but more importantly expensive in TIME. Time in man hours that will never be recouped by our society if we sit around an twiddle our thumbs and pray that boffins come up with a way to defy the laws of physics (shannon limit) in regards to copper, as Ericsson found…
http://www.ericsson.com/ericsson/corpinfo/publications/review/technology_update/archive/2009/issue_2/articles/vsdl2.shtml
Note the date of the whitepaper 2009, 2 years after the first commercial deployments of VDSL2:
http://www2.alcatel-lucent.com/techzine/wp-content/uploads/2014/08/History-of-Copper-Infographic-FULL.jpg
The point is you GET THE FIBRE LAID with the intent of upgrading to AON when the time is right, and not wait till the 11th hour till everything falls over in a screaming heap.
Oh that was only benefit A…
++++++++++++++
B – Reliability / QoS
Here are the facts:
1. The FTTN network that the coalitions proposal was modelled after included that of BT and Chorus (in NZ, until they scrapped theirs because they found FTTP was a better way to go, lulz). The BT model which is still cited today as an example, firstly has better quality copper then our own both in terms of gauge (0.64 and 0.5 vs Australia 0.4) and in terms of insulation:
http://www.smh.com.au/technology/technology-news/gel-casing-degrading-telstras-copper-union-20140209-32aqk.html
This all translates to greater attenuation on the lines resulting in more interference and hence less bandwidth capacity, which is more then likely one of the reasons why Malcolm Turnbull had to reneg on election promise (despite the fact he knew it was the case and misrepresented).
2. Since transmitting signals over copper lines requires electricity to propagate, this also increases with distance (proportional to attenuation) by comparison. FTTP not only suffers from little to no attenuation as a result of not needing the electrical component but is hence more energy efficient provided you can implement electrical generation methods at the destination which is where solar/microgrids come into play.
3. Contention, compare…
– Standard GPON: 2.5Gbps (2,500Mbps) / 32 users = 78Mbps (50 / 28)
– NBNco GPON: 2.5Gbps (2,500Mbps) / 19 users = 131Mbps (100 / 31)
– FTTN: 4 x 2.5Gbps (10,000Mbps) / 384 users = 26Mbps
forget 25/5 that the coalition promised, they’re so backwards they cant even provision the access portion to receive any speeds much better then ADSL2 can get within the appropriate range… let alone an upgrading it… evidence?
The FTTN trials are as @danofmelb verified:
Furthermore if each node can support up to 384 users and there are 206,000 premises in the trial that would indicate the requirement of around 538 nodes… yet they’re deploying 1000 nodes… why? True you have to account for property development (medium to high density housing) but seriously? around 1/2 the ports on each node being unused… why?
Most likely because they want to fake the results of the ‘trials’ and use them to justify FTTN as being viable… how?
Work out the number of users per node: 206,000 premises / 1000 nodes = 206 users per node (give or take).
Plug it into the original contention equation of backhaul / users…
– FTTN: 4 x 2.5Gbps (10,000Mbps) / 206 users = 48Mbps
how interesting…
===========================================
Erm no… just no… you cant get 1Tb/s on a single copper twisted pair, it’s literally IMPOSSIBLE… evidence? Even XG.fast can barely manage 10Gbps (1.2Gb/s) over a 70m copper cable run and this is with 2 pairs…
Now lets state the obvious…
– If you have to run fibre to within 70m of every home, you’d pretty much be handing everyone their own line card at that point FTTP/dp… in which case you might as well make it worth it and do the full FTTP run and save on the OPEX.
– How many houses have 2 pairs to spare? As has been stated many times within the public forum much of Australian premises is on a pair gain network (Telstra’s cheap quick fix to let them expand into new area’s quickly). Therefore even if your wanted to implement this technology you cant do it without running new copper lines… and if your running new cable to premises… FTTP anyone?
1Tb/s… ROFL, don’t worry you were only off by a factor of 1000, similar to how they libs were only off by about $30 billion when they ran a scare campaign pre-election saying it was going to cost $94 billion to deploy universal FTTP.
This is also assuming the copper is pristine and of sufficient gauge (which most of Australia’s is not as stated above) and im pretty sure it was tested in a lab environment only, ignoring all forms of real world issues such as interference and crosstalk (FEXT / NEXT) which would most certainly be present here in this country.
===========================================
Mmk well you keep living in your fantasy land then… ill just stick to the facts… you know the actual facts not ones that have been shit stained by the murdochian empire…
Oh? Kinda like that $94 billion assumption pre-election? No? what about the assumption that Telstra would just give their copper / HFC away for free that is still being maintained?… No, no, no i got it how about the assumption that people will only need 25Mbps at most for the next 20/30 years?…
+9000
@demondownunder You seem deeply confused. An Ethernet cable can transfer speeds between 1 Gbps and 1 Tbps. IEEE ( http://www.ieee.org/index.html ) have been testing speeds ON copper cables to achieve high speeds.
You said here They were able to get that connection for only a few seconds but whats to stop copper from being able to achieve that, IEEE is already developing 1 Tbps to be a standard. I think your claim for copper only being able to barely hit 10 Gbps is completely and utterly a lie, enterprises right now are using 100 Gbps on copper. Copper is completely fine, but next time don’t lie and act so condescending. Constructive criticism relies on factual information. Really I can’t take your argument too seriously.
The copper in the ground is unshielded copper, not Shielded CAT-6 as you seem to assume.
We live in the real world and we have old unmaintained unshielded copper in the ground that is costing $1.3 Billion a year to maintain according to Telstra’s financial statement.
@Munix Of course the copper underground is shielded, the copper is shielded with layers of plastic and rubber covering. The copper network does have to be maintained and they recently renewed a lot of Australia’s copper network. They also wouldn’t use CAT-6 as it is not weather proof. I’m not too sure where got
Telstra is obligated to maintain the copper network but that’s not to say fibre optic cables are better. In England they had fibre optic cables since the 60’s and even that has to be maintained. The point I’m making is copper is completely fine, people seem to think you must have fibre but you don’t. Financially fibre costs more and would make sense for dense cities such as Melbourne or Sydney.@demondownunder I also found this article
http://au.pcmag.com/news/13100/bell-labs-hits-10gbps-on-copper-lines
10Gbps on copper.
And i think you’re an idiot that wouldn’t be able to hold a networking qualification if it was presented to you on a silver platter…
Firstly lets take a look at your claim that:
Hmmm ok lets take a look at the the standards of the IEEE:
http://en.wikipedia.org/wiki/100_Gigabit_Ethernet#Standards
Note the table in there? The first 4 entries are to do with copper cable… But then notice how once you get past 30m it switches to fibre?
MMF is Multi-mode fibre
SMF is Single-mode fibre
You think that copper would be relevant in a network that must have an access layer that has connections that spans HUNDREDS (plural) of meters if not more? DREAM ON
So corporations can use copper (facebook was the one initially requesting 100Gbps ethernet) to interconnect things at their datacenters… BIG DEAL?!
The same technology cannot be used in an infrastructure that has to be robust in terms of transmission in relation to environmental factors and deliver over distances vastly greater then just within one premises.
====================================================
Then you said this:
He said unshielded and you just pointed out the cables are jacketed… Do you even know what shielded vs unshielded means?
Go read up: http://en.wikipedia.org/wiki/Shielded_cable
====================================================
Then there’s this whopper:
I dont think anything here needs to be said to point out how delusional this person is…
====================================================
And Again…
Only as much as is required to facilitate voice/telephony, they are not responsible for the quality of data services (which require much higher frequencies) over the same copper lines… which you would know if you had any experience in comms at all and knew what Telstra’s CSG entailed…
====================================================
I was already aware of such things and it’s still irrelevant. You do realize that even with the latest XG.fast standards they have only managed to actually transmit at 10Gbps at 70m over copper and this is using 2 pairs…
What are you gonna do? Run new copper cable (for everyone on a pair gain) + a node every 70m?
The whole point of FTTN was to avoid running any new cable to the CPE in the first place, if you’re going to do that might as well go fibre because copper wont be able to do it for the attenuation reasons pointed out above…
Could you potentially get 1Tbps on copper? Yes… POSSIBLY for about 1-30m… (as indicated in the IEEE spec) and how does this apply to rolling out an infrastructure that must span a country?… It doesnt, it cant be used because of the distance / attenuation factor.
So unless you come up with some magic that can defy the laws of physics regarding attenuation of signal which has been present in ALL copper based technologies / standards commercialized thus far. It’s irrelevant what speeds you can get on it because it literally cannot propagate signal well enough.
P.S. I suggest you not comment anymore and actually get qualified to talk about this stuff, you’ve already put 1 foot in your mouth, best leave the other one out…
If money is the problem, just remember how the Opera House was financed when the costs blew out beyond the state government’s capacity to pay – a lottery was created to raise the extra money needed. It may not be politically correct to mention a lottery but people didn’t mind paying in.
Upgrading from FTTN to FTTP will double its cost. In the long run FTTN will not only kill the speed and reliability of the service but it will also double the costing due to copper rental from private sectors. The slow roll out of NBN would be costly.