The following is a rather extensive article I’ve written concerning the state of broadband in Australia, specifically to the proposals by both the Coalition and the Labor party to extend what they refer to as ‘high speed broadband internets’ to 98%+ of the population. I’ve touched on primarily technological problems and solutions here, and if I honestly thought it’d do any good, I write it up a little more objectively and send it to both parties. That said, I hope you enjoy.

I would like to congratulate the Federal Government of Australia, for again proving that they have the political flexibility and the technical incompetence to attempt to shaft the Australian populace on a relatively important issue, whilst trying to appear, to the voters, as if they were genuinely concerned.

I am, if you have not yet received the news anywhere, referring to the Coalition plan to deliver ‘high speed broadband’ access to ‘99 percent’ of Australians. This is, of course, rather ambitious, as indeed is the comparable Labor plan of delivering ‘high speed broadband access’ to ‘98 percent of Australians’. The difference between the two plans, apart from a percentage point of population, is rather crucial, and lies directly in the technology that the Coalition and Labor intend to deploy, and how much money each is willing to spend on its particular technological solution.

Firstly, Federal Labor announced their plan several weeks back now, amid a fair bit of fanfare and subsequent analysis that has basically gone their way. To recap, Federal Labor, if elected, will underwrite a nation fiber to the node network (FTTN), providing a total of $4.7 billion as a 50% investment with the private sector, resulting in a total network cost of $9.4 billion. They have claimed that they can deliver up to 12mbps per customer, and that their network will be completed in 2013 - 8 years after being elected.

Pros and cons? On the plus side, it’s nice to see a political party recognising broadband internet as a vital infrastructure for the future - and that’s what the Labor emphasis has been about, that this is about building a network that can later be upgraded and extended, to create infrastructure. Using fiber is itself a plus - from what I recall, the maximum speeds over fiber are at the moment somewhere in the realm of 10Gbps (standard, up to 1.6 Tbps - yes, terabits per second - using wavelength division multiplexing). Yes, ten gigabits per second - which in more understandable terms, boils down to around a gigabyte per second of bandwidth. This isn’t too shabby, and has every possibility of scaling further. Labor is claiming they can deliver up to 12mbps - which sounds crap, when you realise that it’s actually 1.2 megabytes per second, but reading up on the plan it seems they mean 12 megabytes and the journalists recording the speeches and announcements are just incompetent. As a technological assumption, it’s likely that they’ll be using SONET, being optical fiber and all, so at a guess the network will hopefully consist of OC-192 connections, which are commonly used as backbone by large ISPs - providing up to 10Gbps, minus overhead. Even OC-48 (2.5Gbps, minus overhead) would be quite useful. So that’s rather nice - still nothing on what South Korea or Japan have, with fiber to the premises, but then looking at maps of Australia, you might understand why. Another point in their favour is that Labor are promising to enable every node (exchange) in the country - not just the major cities. So every single exchange will, hypothetically, be able to push out high speed broadband. Additionally, from what I gather the pricing structure to access the network will be equal across the board - so no more artificial limitations and price gouging from the eponymous Telstra. Fibre to the node goes further than just the exchange, too - many exchanges are already linked via optic fiber. Fibre to the node takes the fiber cable and terminates it in a cabinet, from which connections are then made to houses. So assuming that you live in an area where placing a cabinet is feasible - i.e., you’re more than one house every few kilometres - it’s possible that FTTN is a good thing for you.

Where the plan falls down is in the last mile infrastructure, at least in rural zones. Currently, last mile technology in Australia is all based on copper, and is thus limited by the fun vagaries of DSL signal degradation. Since the greater part of rural Australia lives more than a few kilometres from their telephone exchanges, which is where the fiber will run to, this of course means that this upgrade will effectively do diddlysquat for them. Even the placement of cabinets won’t help much for the people who live tens, or even hundreds of kilometres from a source exchange. Of course, it is possible that at some point in the future there will be an investment in fiber to the curb, or even to the premises, but on the face of it, this plan is good provided you live within a DSL range of an exchange or possible cabinet site. Granted a great deal of exchanges probably don’t offer DSL at the moment, depending on the size of the two it’s located in - but also granted, there are thousands of people out of range of DSL, who will effectively gain nothing from this plan, and even more so, who won’t be in the running for a node to terminate close enough to their premises. I myself am one of them - I live, in a straight line, within five kilometres of two telephone exchanges, one of which is ADSL2+ enabled, the other ADSL with 2+ coming. I can’t get any sort of broadband over the phone line, because the copper to my house, which was laid in the good old years prior to 1965, runs around 15km before it hits an exchange. And since I live in a rural area, it’s very close to zero possibility that a fiber cabinet is going to terminate within range of where I live. Thus, for myself, wireless broadband is the only option - and I am lucky enough to have a local provider of such available. A final flaw with using copper for last mile is that the state of copper in Australia, quite frankly, sucks. Telstra, our national Telco and owner of 99.999999% of the copper lines, decided that in its infinite (money saving) wisdom that a lot of phone lines, when connected, could easily be split off of each other. This idea, apart from saving Telstra a few bucks, has also made it impossible for anyone of a split paired line to get DSL without Telstra going back and fixing their little nodes - a proposition which is limited in success at best, and has often required the intervention of the Telecommunications Industry Ombudsman.

So FTTN has its pluses, and it has its minuses. If you live metro or regional, or within a few clicks of an exchange, you’re laughing. Also, in terms of providing access, it doesn’t matter if you own or just lease the network, you’ll have fair access. But if you’re already screwed by last mile technology, well, I’m sorry - you’ll probably have to make do for the moment.

Now to the Coalition. The Coalition have a bit of a ‘me too’ aspect to their broadband plans here, despite claims from the Telecommunications Minister, Helen Coonan, that their plan has been in development for ‘eight months’. One might almost be tempted to scream ‘stopgap solution, election in three months!’ at the announcement, given two days ago, that the Government would begin constructing and deploying a ‘truly national’ broadband network. At any rate, here’s a bit of a breakdown.

Unlike Labor, the Coalition isn’t going to be running fiber to the node, or any kind of fiber at all, actually. Basically, they’ve elected to invest in a plan to deploy a two-tiered broadband service - one in rural Australia, and one in metro/regional. Now, the Coalition have earmarked around $2 billion (1.9 something) to spend on their plan, and are hoping that the private sector will jump in at the chance to spend more as needed. The metro plan basically consists of the government ‘allowing’ a private company, or companies, to build a fiber network within the Big Five cities - Sydney, Melbourne, Adelaide, Brisbane, Perth - and screw it for everyone else. Not even putting any public money in, just allowing it to happen, and of course regulating it. Yes, that sounds fascinating, and really long-sighted. The regional plan, however, is more interesting when it turns out that the Coalition is planning to roll out a Wi-MAX network across regional and rural Australia.

Pros and cons for this? Well, for those who don’t know, Wi-MAX can perhaps be summed up (rather inaccurately) as Wi-Fi, but with 70Mbps of synchronous bandwidth and a maximum of 50kms in range. Yep, fifty kilometres of range. Not too shabby, is it? Of course, the problems do crop up - being that, much like cable internet, all the bandwidth on a Wi-MAX node is shared between all clients, and that the laws of wireless transmission apply like nothing else - so forget getting full strength, or even potential signal at 50km from the tower, especially with hills and trees and other such things in the way, even if the WiMAX forum claims the technology is NLOS. According to the WiMAX Forum, a tower in a metro area will have a range of roughly a single mile. So less than 2km, depending on interference and radio spectrum used - although this is in heavily built up metropolitan areas. According to testing by the WiMAX forum, a more realistic expectation is approximately 10Mbps at 10km of range. Of course, if better line of sight and more spectrum is available, the transmission can push further. Assuming that there isn’t as much interference, maybe even up to 20-30km from the base station will be possible to receive a decent signal. However, the Coalition is promising 12Mbps connections (to match Labor, naturally enough), which means a station every ten kilometres or so. And no more than six clients per node. This sounds as if it could get very expensive, depending on the price of per node, and how many nodes can be installed on each tower - this is an area I’m slightly unclear on, and I can’t find much data in the WiMAX documentation. I assume that multiple nodes can be present, although it’s quite possible each would require it’s own breadth of spectrum.

So the Coalition plan basically involves building a network of WiMAX towers, and connecting them up to existing exchanges. As I’ve stated, this has problems with sharing of nodes and with the fact that, well, it’s still connecting to existing backhaul, which means that any bottlenecks that already exist between the major fiber links in the capital cities and the exchanges will only be exacerbated. Also, the standard of WiMAX that is being rolled out is 802.16d, which is nice being that it’s stable-point - but it appears to be more affected by such things as weather and LOS than 802.16e is. Another nasty is of course that the Coalition is planning to effectively treat non-capital city folks as, well, second-class citizens. Unless every five or six families are going to receive their own group WiMAX node, there’ll be no way that any form of ‘high speed’ broadband can be provided by this plan. ADSL2+ for the people with copper, and perhaps even cable for those within cities, but shared node WiMAX for those in rural Australia. Sorry, Coalition types - you haven’t exactly won me over either. Especially as rather amusingly leaked documents show that the most detailed rollout plans for WiMAX that you have are in marginal seats. Smooth, gentlemen. Especially since, from the OPEL (Optus/Elders) announcement, it appears that there will be 1361 wireless ‘sites’ (I assume towers) and ADSL2+ activation/DSLAM installation in 426 exchanges. Total coverage? Please show me how.

At the end then, we have:

a) the Federal Government plan of making some slapdash arrangements by dropping WiMAX towers in marginal seats as priority and then safe/opposition safe seats as a secondary consideration, and then allowing tenders to build fibre networks within capital cities from the major telcos, or any smaller groups that want to do so, and promises that high speed broadband will be enjoyed by 99 percent of the country.

b) the Federal Opposition plan of spending 8 years building a fiber to the node network, providing gigabit internet bandwidth to every exchange in the country, as well as to new nodes to extend DSL range from exchanges, but with no further information on how this is specifically going to help those people already stranded by the poor quality and distances that much of the prevailing last-mile copper network encompasses, and no details on where the nodes will be.

It’s a puzzle, to be sure.

So what should be done to give Australians decent, total broadband coverage, or at least as best as can be done? In truth, neither Labor nor the Coalition have hit the nail on the head. Where the Coalition has failed is in its provision of both backbone and viable last-mile technologies, but especially in backhaul from the users through to the metro nodes, and thence via international communications. Where Labor has failed is in provision of last mile technologies, as its network effectively stops at the node, and any further transmission is as yet unknown. So both plans, although interesting in their own ways, fall down in terms of providing universal access to the consumer. Potential access, certainly, but not actual accessibility, either in last mile (Labor) or in backhaul (Coalition).

Where does this leave us? At present, it appears that Federal Labor will win the next Australian general election, and will then begin their nation-spanning FTTN plan. All well and good, except it still leaves us with the problem of connecting clients to the backhaul network. This is perhaps where the current Coalition plan, with modifications, might come in handy. Consider - a nationwide fiber to the node network is built, connecting all regional, rural, and metro exchanges to a nice, high speed fiber network which provides a very substantial backbone to all Australian telecommunications, as well as pushing nodes out to clusters of consumers that are out of DSL range, but are grouped closely enough to make a fiber cabinet in the area feasible. All the data (apart from satellite customers) has to originate from a node in the area anyway, so providing this kind of backbone means that the only remaining obstacle is the last mile. So what is to be done for the last mile anyway?

Despite what you might think, I don’t think that WiMAX is the wrong answer. I feel that, however, WiMAX is not the necessarily only answer. DSL and its variants have limits to which they can be extended from the exchange, and it is these limits that are, of necessity, those that limit the access of hundreds of thousands of citizens from high-speed copper broadband. However, these limits need not remain. DSL lines can be extended by means of a repeater, basically a ‘mini exchange’. These are installed every three or so kilometres from the exchange, and basically extend the signal for another 3km. Since they can be installed along copper, there seems to be no reason why repeaters couldn’t do the same trick from fiber nodes, which are themselves extended from exchanges. If these could be daisy-chained along the existing copper, the last mile problem a lot less problematical. Coupling a solution like this with WiMAX for regions where installing repeaters are simply unfeasible (either completely uneconomical or too difficult in an engineering sense) and we have a rather interesting hybrid, a hybrid that makes more sense than either the Coalition or the Labor plan on their own.

There you have it then. Both the Coalition and Labor should be applauded for planning any broadband infrastructural increase at all - of course, Labor deserves more credit for their plan then the Coalition, due to a rather more ‘future forward’ idea, more funding, and of course releasing the plan as a standalone development rather than a quick response… but I digress. FTTN is well and good, but isn’t all that is needed. FTTP (to the premise) is far too expensive in the distances that are required, so copper must still be used for last mile technology at present time. Therefore - build a FTTN network, pushing the availability of incredibly high-speed access further from the exchanges, and to every exchange. Use repeaters where possible to extend the signal as far as possible, and where it just isn’t going to work, build a WiMAX tower. As I read in a rather informed article on the subject, use cable where it’s feasible, and use wireless where it isn’t. And you’ll build a better network.

There’s my challenge to both major parties. The really ironic fact is that they’re seen as competing at all, where deploying the WiMAX solution of the Coalition in the short term whilst the FTTN is built for the future isn’t itself a bad plan, is rather ironic. So go ahead, Coalition, and build your WiMAX network - in fact, I’m hoping that you do. But, afterwards, something more is needed. And Labor, at the moment, has the better base from which to build upon.

From the last mile, to the node, to the backbone, to the international backhaul, Internet access in Australia is currently deplorable. It really, truly is. As a voter now, I’m simply asking you both - look at it, realise it, and do something about it.

rnAlcata’riel.

-Andiyar