Friday, 13 April 2012

What about the alternatives?

So, we've covered a little of what the NBN is and how much it will cost. These have been, by no means an exhaustive study and I, myself, learn more everyday I read about the NBN. The scope of this project is such that individual reports about it don't do much in understanding it as a whole. As always, more reading = more knowledge.  So get out there and read it all! (read mine first though!)

But, there is a serious question as to whether or not the NBN is a viable project, too much too late, too much too quickly or any combination therein. There are a number of alternatives that have been put forward and I'd like to examine them.

What the competition thinks......

As always when it comes to Government projects, there is going to be a competing project.....although of course with politics, they call this a "policy"...... Yes, because that makes it more likely they can cancel it and call it a "policy change" rather than a waste of money.....

Anyway, ranting aside, we have 2 major parties in this country and the Coalition in this case DOES have an alternative "policy" to the NBN. In their case, it has been rather difficult to pin them down on exactly what their alternative would entail (here's a rather tongue in cheek funny).  But, the one fixed point seems to be this; the Coalition would seem to be working towards a FTTN (Fibre-to-the-Node) rollout to the vast majority of Australians, although there is currently no specific statement about this. The gaps would be plugged by a combination of fixed wireless (both microwave and "wireless" ethernet) and satellite coverage, as with the NBN, to provide "up to 100 Mbps down to 12 Mbps".

So, firstly, the wireless and satellite sections of both policies appear quite similar. However, they vary quite significantly in one important way; NBN Co. will deliver FTTH to 93% of Australian premises and the other 7% will be covered by this combination in wireless technology. The Coalition's plan does not specify how many Australians will be covered by the, as yet hidden FTTN rollout in their policy, only that, " 2016 Australia achieves a national broadband baseline with 97 percent of premises able to be served by high-speed networks...." The baseline has been given, up to 100 and as low as 12 Mbps, but there is no indication of who will get what speeds. Under the NBN, 93% of Australians will have, conclusive, access to minimum 12Mbps and up to 100Mbps at their choice. Current plans work on 12 Mbps, 25 Mbps, 50Mbps and 100Mbps speeds everywhere the NBN (fibre) is currently available. Obviously these speeds vary, but measured speeds have been very close to 100Mbps (94-95 regularly as measured by NBN Co.)  The other 7% will have access to speeds of 12Mbps, higher in the future. (Edit: there have been recent articles about the NBN Interim Satellite Service providing significantly less than 12Mbps- this is true. For one significant reason- these are INTERIM services offered over existing satellites owned by commercial companies. NBNCo. is launching its' own dedicated satellites in 2014 and speeds will GREATLY increase at this stage. Even so, it's been reported more than 7300 people are already signed up to this service, most likely because it is subsidised by NBNCo. and are therefore, in general, cheaper than other available plans)

This, on the surface, then, appears similar, but is in actual fact quite a different strategy. The NBN offers 93% of premises 100Mbps from first connection, and this is just initial speeds. Recent speed tests in China have shown a single fibre core can handle 1.7 TERA-bps.....about 170 times as fast as current NBN speeds. The Coalition's plan will bring UP TO 100Mbps from initial connection......and yet even THIS isn't entirely accurate. Let me explain....

The FTTN rollout the Coalition plan may include will replicate much of what the NBN will cover; primarily, it will use backhaul (backbone) fibre across the country and install it to all large nodes and exchanges. It will also include running fibre, instead of HFC, copper or otherwise, to ALL exchanges within the 93% area. So far so good; but this is where the divergence takes place. 

While these fibre runs will greatly increase efficiency and uniformity and thereby decrease congestion across badly affected networks, this is as far as the fibre goes. The last (up to) 6km from the local exchange (some people are even further away and have essentially ISDN speeds) to the customers premises will STILL be whatever currently lies in the ground. To the vast majority of Australians, this is the Telstra CAN (Customer Access Network)- copper. There are several million premises now on Telstra and/or Optus HFC or fibre, but this is still only a growing minority; why would Telstra or Optus spend the money? It doesn't gain them much, as many Australians already baulk at paying for current broadband speeds, so they're unlikely to gain much revenue to begin with. This means the majority of Australians may well have fibre to their exchange......but be stuck with ADSL2+ speeds, around 24Mbps at best. This is a limitation of the copper, NOT of the exchange or backhaul. (EDITrecent tests by Alcatel-Lucent have shown speeds on copper "achieved downstream transmission speeds of 300 Megabits per second (Mbps) over distances up to 400 meters (or 100Mbps at 1km)"- READ- UP TO 1KM. More than 60% of Australians are more than 2km from their exchange (NBN Corporate Plan- page 40)

Also, a FTTN network would be incompatible with a FTTH network when we do, inevitably, upgrade.  Why? FTTN means, if you want any upgrade in speed from current setups (thanks to last mile copper still being used from the exchange) you would have to install many intermediary fibre nodes on street corners, between the exchange and the premises, to shorten copper lengths. These also have to be powered, as to change from fibre (non-conductive) to copper (conductive) you must have a power source. So these nodes will be not unobtrusive and use significant (once you add many up) power. And when a FTTH network would eventually come along, even under the Coalition? They all have to be ripped out or heavy labour carried out of splicing fibre done, because the fibre must go all the way to the premises and the nodes are useless on their own.....sounds like a massive waste of money if I've ever heard of one....

Part of the Coalition's plan also involves funding for "Providing a way forward to a higher bandwidth, more fibre-intensive Australia."  This is fairly vague and, when you read the statement, amounts to subsidising Telstra and other operators to increase the amount and decrease the timeframe of their own FTTN/FTTH rollout.  This, for fairly obvious reasons, doesn't amount to much, especially to unprofitable regions such as regional and rural areas.  And we'll still be stuck with the bewildering array of cross selling between competitors for use of their CAN's (anyone who has looked to see if their ISP has DSLAM's at their exchange will know what I mean).

So, to recap, the Coalition's plan "may" involve rolling out a FTTN network, which will result in the bottleneck moving even more from the exchanges, to the CAN's, and little extra speed for the majority of Australians in the next decade. And it will rollout wireless broadband/satellite to the other 3% not receiving FTTN. But those ON the FTTN will only be guaranteed 12Mbps, as will those on the wireless portion (EDIT: actually, on satellites, the Coalition have not said they will launch their own, like NBNCo., so they are likely to have speeds less than 6Mbps, as experienced by people currently on NBNCo.'s Interim plans because of usage contention on commercial satellites).  The speed bottleneck of copper (25Mbps now, with maybe 100Mbps within 1km) will be in place to premises until ALL CAN's are upgraded to fibre (or HFC), subsequently junking the FTTN and this has no timeframe. Only once all CAN's have been upgraded to fibre, can the true bandwidth of the FTTN plan be recognised.....because it becomes a FTTH network....

The NBN will bring FTTH to 93% of Australians guaranteeing minimum 12Mbps and most likely close to 100Mbps to that 93% by 2021. It will roll fibre PAST the exchanges, to the vast majority of premises, removing the bottleneck on speed. The other 7% will receive 12Mbps via wireless/satellite services to begin with, with more planned in the future.  These speeds are also only initial. Hardware upgrades can already achieve 1Gbps on the fibre NBN and much higher in the future. And NBNCo. has committed to further fibre rollout and cross-subsidies with communities to possibly roll fibre where NBNCo. isn't currently planning.

And the cost of this rambunctious rubbishing of our current cables compares?

Actually, yes.  The cost of the NBN, as I've discussed, is slated at around $36 Billion total and $28 Billion to the government.  The Coalition's costing of their policy? About $7 Billion.  Big disparity?  Yes, it is.  But, firstly, the service provided is much higher, as I've explained and further reaching, as the FTTN network would eventually have to be scrapped or heavily modified to keep up with speed demands.  Secondly, and, as is always the case in governmental spending, you can't simply switch from one to the other for free.  This is no different from tax savings in one governments budget and not in another's.....or a Carbon Tax in one, but not in the other.....If the Coalition were to be in power RIGHT NOW, it would still cost them several Billion just in penalty payouts to contractors for the NBN and several Billion more for penalties to Telstra (EDIT: they've changed their tune now but that means more NBN will be rolled out BEFORE their plan comes into affect....getting less like FTTN and more like FTTH everyday...) .....and that's not even possible as the next election isn't until 2013.

So while $7 Billion is a nice number out there all on its' own, as is always the case with political policy, it doesn't take into account real world costs.  It's been estimated that, should the Coalition get into power in the 2013 election, cancelling the NBN rollout in whatever its' current state, paying out penalties, doing feasibility studies, organising tenders and actually implementing their broadband policy would cost around $16.7 Billion.  Plus the already built part of the NBN that has to come off the budget, as it can't be payed back anymore with NBN Co. no longer another $15 Billion.  That's $32 Billion.....and we get what for it?  An inferior network, with many of today's problems and no increased reliability or real price difference to the Australian people.

Oh, and it wouldn't even START rolling out till 2018, 2 years beyond when they have said it'll be finished; because, first they need control of the Senate (mid-2014) and they've said they would require a CBA to be done and renegotiation with Telstra.  So it won't be done till 2022 with that timeframe.....a year AFTER the NBN is due for completion.... (EDIT: this last bit about timing is my analysed opinion, but it is shared by this Citigroup analysis and other industry experts)

What about competition?

The NBN not only brings with it the physical upgrade of the cables in the ground and the hardware to go with that.  It also brings a wholesale network back under government control, something not in place since Telstra was partially sold off in 1996.  Alot of this Telstra guff is vastly influenced by people's political preferences, but the message from the ACCC over the years has been clear; until Telstra is separated into wholesale and retail divisons, true competition in the marketplace is difficult.  Why is that? Currently, Telstra have contracts with the government to provide a certain level of telecommunications service to Australians (see here for some of those requirements).  But Telstra is also MOSTLY a private company.  Over the years this has severely affected the way they do provide services as an almost sole owner of government built infrastructure.  Currently, if a customer on a competitors service requires an additional line to be installed, a repair of their line or just a diagnoses on it, they MUST go through Telstra if Telstra owns that CAN (most of which they do).  Anyone who has had problems getting broadband or has had a dirty line knows what I am talking about.  

NOTE: OPINION/RANT PIECE APPROACHING!! Skip to the bottom of the paragraph if you're not interested....

My personal experience runs to the fact that, at one of my houses (back in 2003) I had dial-up and wanted broadband.  I rang iiNet, who seemed competitive and had good value-added services (such as VOIP).  They said I could not get broadband.  I had to ring Telstra.  Annoyed, I rang Telstra and was told broadband was not currently available in my area, but I could go onto ISDN with them.  I didn't know anything about ISDN, except that it was faster than Dial-up, required extra hardware, and allowed always on internet; what I wanted....for a price.  I agreed and they came out, the gentleman installing it had significant trouble, mumbling about his counterparts at HQ not having told him about a required "upgrade" before he left and he would have to come back with other equipment.  He came back a few days later, ferrying between the exchange and my house and eventually got me up and running.

All was well (other than the HIDEOUS cost of ISDN + phone calls dialling up all the time).  About 16 months later, I was getting rather disgruntled with my speeds and was talking to a neighbour across the road.  They happened to mention they were on broadband and had been for over a could this be I asked myself? I'm across the road from them??  I rang up iiNet again and was told my line was not eligible for ADSL, but after saying I was on ISDN, they did a Telstra check and found my line was capable, but it didn't show up because of the ISDN hardware on it.  I asked why they didn't tell me this over 12 months ago, that I could've gotten ADSL and after explaining what I had had done with Telstra, it became apparent what had happened.

ISDN could be run in situations where ADSL could not, back in the day, before advents such as Extended ADSL and ADSL2+.  This is mainly because of quality issues on poor lines.  BUT, hardware at exchanges has to be enabled for ADSL to be used on lines (this isn't even getting into pair-gain and RIM, both of which ISDN CAN run on, but ADSL can't).  What it transpired had happened, was Telstra had been in the process of upgrading our exchange, about 2km away, to be ADSL ready.  This was never mentioned, even though I asked.  About 2 months after I received ISDN, ADSL became available on our exchange, about the time our neighbours moved in.  I was never rung and never told this whenever I spoke to Telstra- until I told them I would be changing to iiNet....  iiNet told me this was commonplace; Telstra would provide a better service, when it became clear a competitor would get the business otherwise.  But it would NOT inform the customer if the situation changed, such as in my place, where I could now get ADSL.  Needless to say, I reported this to the ACCC/Ombudsman.   I was credited 1 months ISDN service......the month I cut it off and got ADSL with iiNet, which meant I couldn't use it.....

That long winded story should illustrate my point quite well; Telstra has no incentive to ensure a competitors customer has better or even equal access to services, although their contract stipulates they are to provide these services EQUALLY to ALL wholesale customers (retail competitors), they quite often didn't and still sometimes don't.  The Ombudsman quite regularly wrist slapps Telstra and publicly condemns them with the poor job after they were partially privatised.  The only way around this was if the competitor owned their own DSLAM in the exchange, but this was still limited if the actual line in the CAN (usually Telstra's) was damaged, degraded or, god forbid, a physical pair-gain line.

The upshot of all this is that with a vertically unseparated Telstra (retail/wholesale in one), there are major barriers to true, equal competition in Australian broadband and telecommunications services.  I, for example, would dearly like no home phone, and to use VOIP (I've used VOIP since I first signed up to iiNet and only use our home phone as a dial-in number).  On iiNet, Internode etc. this is known as Naked DSL and it saves you $30 a month on line-rental; another way of saying $30 a month of 'hiring Telstra's hardware'.  You pay for the broadband and VOIP service, but not the line rental.  This is ONLY possible if the company has a DSLAM at your exchange, which, living in a regional town, they don't.  Telstra don't offer Naked DSL, for fairly obvious reasons- they own the hardware and want the extra $30 out of you for line-rental, why would they allow you to circumvent that?

This sort of stifling of the competition is a result of a mostly private company, wanting to increase profits, owning critical public infrastructure shared across the nation by many companies providing services.  As part of the the NBN, many of Telstra's copper access routes/conduits and some fibre lines will be leased and repackaged to use/house NBN fibre.  Telstra will become wholly private and while they will will still own their copper, fibre and HFC networks, most of them the Telstra Financial Heads  Agreement will force to be decommissioned and have users migrated to the NBN.  The copper CAN will be largely left intact, but their conduits and access will be leased to NBN.  This puts them on the same footing as all other ISP's in Australia, albeit with their larger market share to back them.

The NBN enables Telstra's stranglehold on the wholesale CAN to be lifted, but it IS replaced by another government owned company.  Some say this will result in a repeat of the Telstra situation with the NBN down the line.  This is speculation.  There is talk of selling off the NBN after 10-20 years after construction.  If this happens, the likelihood of it being partly sold off like Telstra is slim- why repeat a mistake?  A wholly private company, regulated, owning the infrastructure has benefits and disadvantages, which I won't go into here, but suffice to say, it will NOT result in a Telstra-esque problem.

There has also been speculation that the NBN will stifle competition as it will provide the same service to all ISP's on the infrastructure and therefore there will be little to choose between.  This is true to a certain extent, but this will actually likely mean better competition between ISP's on pricing and VAS or Value-Added Services, such as VOIP, VPN's, cloud storage solutions and so on.  What it does mean though, is that all Australian's will be able to receive fast, effective, reliable and cheap broadband collectively, rather than the patchy nature in which we receive it now.

What about wireless broadband?

Malcolm Turnbull has been vocal in his disagreements with the NBN (a pity he's in the Coalition, my gut tells me he knows the NBN is a decent deal, but he HAS to rally against it) and one of his discussions has been about "Next-Generation Wireless rendering the NBN obsolete."  This is an interesting question.  Could the current crop of "4G" mobile technologies coming to market be better, or at least as good as the NBN, but with mobile convenience?  Let's take a look.

Current technologies of 4G are clustered around a technology known as LTE (3GPP Long-Term Evolution; or 3rd Generation Partnership Project- Long Term Evolution.....thanks goodness for acronyms!).  In Australia, the only current 4G network is Telstra's 4G LTE network (this is not technically a 4G network (LTE-Advanced and WiMAXX 2 are real 4G) from the original definition (1Gbps downstream), but it's enough to be going on with and it has been recognised under the general umbrella of 4G).  Optus have (Edit: just started) to rollout this month.  Vodafone appear to be waiting for the dust to settle and have their own problems to deal with in the meantime.  But 4G has some promising characteristics.  If used in the right spectrum (Telstra's using 1800Mhz, Optus have tested in the 700Mhz but are also rolling out 1800Mhz) such as the 700Mhz range, where analogue TV is due to be turned off in 2013, then the spectrum sold off, it can provide 15-20Km radius cells with speeds of up to 40Mbps.  Telstra has shown off speeds of between 34Mbps -18Mbps down, but in real world testing, Gizmodo AU has found it reaches anywhere between 18Mbps and 8Mbps, which is still significantly better than 3G (HSPA-DC) at around 8Mbps tops.  However, this is still worse than some CURRENT broadband subscribers can achieve.

LTE-Advanced, when it is implemented, should bring speeds approaching 50-60Mbps in real world use.  But that is assuming the ideal scenarios including enough cells to handle all customers in any given area, line-of-sight to the cell tower, weather, and any of the normal affects on wireless such as solid walls (and the materials they're made of) and interference.  This is an awful lot to go right, and that JUST gets the signal to the cell.  The operator then has to have its backhaul (fibre) and connections to the landline backhaul (what the NBN will be) are all operating at peak efficiency.  Now, it's true operators will have to make sure their nodes and servers are able to handle the load of their subscribers data requirements for the NBN to operate at its' optimal speeds.  But this is much easier seeing as their nodes and servers are the ONLY things the operators have to look after, unlike in wireless networks, where they must look after ALL hardware and interconnection backhaul, rather than just the servers and nodes.  They can spend less money, on less hardware but maintain higher throughput because they are only looking after their end, not the customers end.  And this doesn't even mention the fact that there will be many different frequencies of 4G in Australia, therefore many different devices required to access each network.  Or some sort of homogenising of the networks to make them all compatible (likely by government subsidy) which is unlikely.  

The NBN will be uniform with the same hardware provided to all customers for the same reliable level of service.  Operators will look after only their POI's (Points of Interconnect to the NBN) and servers (and their own backhaul if they choose to continue running it against the NBN) to maintain high throughput for customers; a much easier proposition than looking after an entire network.  Yes, they will have to pay NBN Co. to access the NBN, but all operators are charged the same, regardless of size or location, which puts the competition aside for providing higher throughput or more DSLAM's for access.  Wireless broadband uptake in Australia is very high, but this is to be expected- we have a large country and many of us travel large distances to work or to see friends and family.  However, many wireless broadband subscriptions are work related and many are because of the very reason the NBN is being rolled out; they don't have access to reasonable price, reliable broadband.  In my area, wireless broadband is a stop gap that barely allows you to check email and Facebook and forget streaming a youTube video at any sort of decent quality.  But this is ok for many older people, people who do not work with the internet or need it for work.  But the new generation is consuming data at an ever increasing rate:

This graph from the governments Department of Broadband, Communications and the Digital Economy's: Convergence Paper shows that the wireless consumption of data has decreased (but in fact over the same period wireless subscriptions increased) but the fixed line consumption has increased by a significant percentage (nearly 20% in fact).  This shows that while Australians have a love affair with being internet connected wherever they go (14.5 Million subscribers in June 2011) their heavy lifting, in terms of data downloading, is always done at home or in the office.  And the AMOUNT of data being downloaded is growing ever faster.  With the explosion of Internet TV, HD VOD, online movie rentals, music streaming, cloud-storage, online photos and video manipulation and catch-up TV, Australians are moving into the 21st century with open arms, embracing high data content on-demand wherever they are, but particularly from their couch or office chair. (EDIT: and this is just consumer value- we haven't even GOTTEN to business potential yet)

The NBN is as much about replacing Australia's ailing telecommunications infrastructure as it is bringing Australia, guns blazing, into the international digital battlefield.  It is, to a certain extent (as much as you ever can with technology) future-proofing Australia as we demand more and more from our internet connections.  We want fresh piped info, video, social networking, work and more all at our fingertips with no wait, no delay.  The NBN is the best equipped to do this, through its' total replacement of most of the current infrastructure.  There are cheaper ways to provide slightly better broadband to more Australians.  But Australia has always been at the forefront of most things we try to do.  Why should we kick back and watch as we're outstripped now?

No comments:

Post a Comment