8 posts • joined 28 Sep 2011
Broadband tool inflates median peak speeds
Malcolm's Magic Broadband Pudding Gauge seems to inflate current speeds.
MAIT 73 is a pocket of my suburb where I have surveyed many residents, none of whom ever achieve sync speeds of 3 Mbps, yet the "median peak speed" quoted for MAIT 73 is 3.79 Mbps, with the assertion that some residents would probably achieve double this speed.
Yeah, right. The best we ever see is 2.5 Mbps, but the copper is so appallling that unrecoverable errors see it drop quickly back, normally to 108 Kbps, before I pull the plug and force a cold resync.
I contend that the numbers of premises experiencing inadequate ADSL speeds are demonstrably far worse than this tool says.
The view from government
"See, the problem with the project as it’s – as Labor framed it – they massively underestimated the cost, the complexity, and the time it would take to complete."
Sounds like what Mr Turnbull did when he said every day from April to September that the gross cost of his 72% FTTN and 22% fibre would be $29 billion, and that he would complete it by 2016.
The fact is that Labor told NBNCo never to use copper inside buildings. For mine, this is the only constraint that should be relaxed. Fibre should be laid to all urban premises because it will generate higher wholesale revenue to repay sooner, yet cost about the same (money and time) as buying and remediating ten million copper pairs, building/electrifying/maintaining 60,000 nodes and taking responsibility for asbestos, which is currently Telstra's contracted responsibility.
Every urban building should get fibre, and only where internal fibre is genuinely frustrated should copper or Wi-Fi be used within the building to deliver the service to the occupant.
This will reduce the cost of FTTP, compared to forcing internal fibre to be delivered where impractical. but if fibre is not laid to the building, the wholesale revenue stream will be inadequate to self fund the project.
Is this is a trick?
The only reason anyone will want to use the existing HFC infrastructure in a few years is for watching TV. It is almost as useless as dialup for outgoing data, with 1 Mbps upload speeds the norm.
Is this submission designed to impede the fibre rollout? Provided the issue is not used as a way to confuse the public over the merits of fibre, there should be no reason not to examine it.
My chief concern is that if the ACCC requires Optus and Telstra to grant at-cost access to HFC to its competitors, they will demand more compensation from the government for decommissioning HFC when fibre arrives, and that would be a major reason not to open it up. I smell a rat.
No cabinet, no problem
Far better not to have the cabinet there to hit in the first place, surely? FTTP rules.
Re: NBN - Veldan
Veldan, the flawed decision to sell 100% of the taxpayer-built copper to Telstra Ltd is the reason FTTN is uneconomic in Australia. John Howard recognised this in 2004 when he tried to circumvent Telstra's monopoly with his first national broadband extension proposal.
But it was the tax-funded nation-building work of laying copper to country towns, hamlets and farms in the 1960s and 1970s which first enabled basic communications for everyone. Much of that copper now needs replacing before it can deliver FTTN. Copper maintenance alone costs Telstra $1 billion per year and costs households and businesses a lot of downtime. Oncea decision is made to build national infrastructure, it inevitably means that government must run the project.
The past five years saw stagnant HFC numbers, Telstra bullying competitors by waiting for them to install a DSLAM before activating their own to reduce the profitability, and a third of Australians still unable to get ADSL at all, with the median spedd in June 2011 only 3 Mbps despite the price paid for 24.
You are not forking out anything for FTTP, which is built with project borrowings it will repay from wholesale revenue. Takeup of higher than expected speeds (only 16% are buying 12/1) proves the cost recovery will come even faster. It is Turnbull's FTTN which will cost you, firstly at least $11 billion, handed to telcos, to build massive cabinets and lay fibre to them, then at least $15 billion compo for the copper, much of which we will need to replace with new copper!
FTTP does not require FTTN. In fact, FTTN must be bypassed by new fibre starting at the exchange and running to the premises. The cabinets will remain as a stark reminder of the money wasted in building them only to be decommissioned before they even had time to rust.
The population density of the fibre footprint in Australia is quite similar to that in other countries with FTTP. Half the area of Greater Sydney comprises water and parkland which does not get fibre. Laying fibre right to the building in a suburban street here is no different from any town in Japan or South Korea, in fact the proportion of multiple dwellings is smaller here, reducing the work needed to get it to the individual premises. And our income per capita was double that of South Koreans when they built 100 Mbps FTTP last decade, though it has subsequently risen a bit. The USA doesn't bear comparison, as everyone there complains about the geographic turf wars of the big providers, who not only charge like a wounded bull, but have now moved to cap data as video now comprises 49% of internet traffic there.
In short, yes, we need the government to get adequate services beyond three cities; no, the track record of market delivery is poor in regional Australia; no, you are not forking out $40 billion, but you will fork out $25 billion for a white elephant under Turnbull's plan; and no, our population density in the FTTP urban footprint is perfectly comparable to our go-ahead Asian neighbours.
FTTP is the right answer, and the off budget funding model protects budget areas from losing funds, while guaranteeing no delays to the project in the event of changed budget priorities. The coalition must adopt it, or it will struggle to retain or win regional seats in 2013, as it did to most peoples' dismay in 2010.
I beg to differ
Having handled my first iPad under a year ago, I have since deployed a hundred to field staff. Few ever use their laptops any more in the field, with some leaving them plugged in on their desks at home base.
I do agree that MS Office alternatives have been shocking on the touch screen. Until now.
An app called CloudOn has just appeared, and provides real Word 2010, Excel 2010 and PowerPoint 2010 on a server, with a quite useable touch screen functionality. Free for now, expect them to charge a usage based licence fee next year. But it ups the ante for Microsoft to deliver cloud based Office to the iPads, and not only to its own tablets, whenever they appear.
A lesson in universal natural monopolies
Scotty, if you are arguing that privatising the NBN fibre after it is built would create another fiasco just like selling Telstra the universal copper did, then I agree with you, but I suspect you have a different view.
Right now, depending on where you live, your options are not limited to cable, ADSL and wireless - seven million Australians cannot even get ADSL and most of these cannot get useable 3G indoors or at all! Meanwhile, nearly 20% of capital city premises have both Telstra and Optus cables in their streets!
The reason for this major fail for regional Australia is that we left it to the free market for fifteen years. Universal fixed comms infrastructure is a natural monopoly. Localities that are more costly to build to will never be serviced by a free market. This is why South Korea spent $46 billion laying fibre to premises last decade. Now for just $2 billion we watch them coax corporations to upgrade the switches to gigabit and in some areas 10 gigabit speeds.
As to your flawed price comparisons. There is no reason to buy a phone service on the NBN if you have a fibre data service, so the $24 wholesale charge (which includes GST) is incurred once only. All calls will be VoIP (not PSTN) from either port anyway. So, for a pensioner who wants the phone and web access an iiNet $49.95 20GB NBN fibre service is three dollars cheaper than the cheapest Telstra configuration of $22.95 Home Line Budget plus $29.95 5GB ADSL. But you then pay Telstra for the STD and local calls, or you pay $10 or less for all your calls on VoIP. The legislation also provides that a phone-only service will be made available for no more than at present to pensioner health card holders and the like, by means of subsidies, as it is now.
Rest assured there will be Dodos and Exetels offering leech-grade services, possibly as loss-leaders as they do now with ADSL products.
NBN fibre enables a completely free retail market, using a natural monopoly infrastructure, and cannot be delivered to all of Australia in any other way, nor any cheaper than is being done.
Fibre to premises costs $12 billion, FTTN $26-30 billion
Mr Turnbull knows that a national FTTN rollout (ADSL extensions requiring 20,000 electrified street cabinets to extend the reach of the ageing copper) was costed at $11 billion in 2007, PLUS $15-20 billion compensation to Telstra for cutting their copper, which John Howard sold them against advice to keep the infrastructure and sell the retail business.
If FTTN (which is permanently capped at copper speeds) cost $26 billion in 2007, how can it be cheaper than building universal fibre to homes at a total cost of $12 billion in 2011?
To his FTTN folly, he must also add the cost of Ka-Band satellites for remote Australia, and 400,000 wireless services to the premises beyond fibre's urban footprint. Pretty soon he could be talking a lot of money for his inferior alternative.
Just as taxpayers built copper phone lines across regional Australia in the 1950s, a decade of market failure to upgrade it proves that only taxpayers will ever build fibre across Australia, a one-off cost for at least a fifty year life expectancy, compared to a five-year dead end of power-hungry FTTN cabinets in every suburban street and regional Australians again neglected.
Right-minded persons oppose borrowing to fund government expenditure, with the single exception of infrastructure which cannot but generate revenue sufficient to substantially repay the loans, and the NBN will do so in spades, while creating a level playing field for big and small providers to access any customer regardless of geography. The coalition should adopt the whole model and deprive Labor of its only remaining popular policy, or risk regional electorates again as they did in their 2010 broadband fiasco.
- Vid Hubble 'scope scans 200,000-ton chunky crumble conundrum
- Bugger the jetpack, where's my 21st-century Psion?
- Google offers up its own Googlers in cloud channel chumship trawl
- Interview Global Warming IS REAL, argues sceptic mathematician - it just isn't THERMAGEDDON
- Apple to grieving sons: NO, you cannot have access to your dead mum's iPad