Hosing down the hype on wireless internet technology
|
There are two reasons the opposition is such a fan of wireless broadband. One is political: fibre equals National Broadband Network equals Labor equals something to destroy. The other is that the Liberals are being sucked in by the wireless vendors’ glossy brochures. A core part of the opposition’s anti-NBN propaganda continues to be that a fibre-to-the-premises network is somehow both risky new technology and old-fashioned. Wireless broadband, especially the stupidly-named “next generation” technologies like LTE, is the safe bet that will soon surpass NBN capabilities. Their media cheer squad is only too eager to parrot this line. It even sounds like it makes sense — provided you ignore the technical subtleties. “Tied to cable yet future is wireless,” was the headline in The Australian yesterday on a typical anti-NBN rhetorical piece by analyst Ian Martin. There’s an appeal to authority. Barack Obama announced a National Wireless Initiative. If America is doing wireless it must be right … right? There’s an argument from incredulity. Martin discounts a government-supported US fibre network because it would have cost up to $US100 billion. “It’s unthinkable that Congress would have supported that kind of budget spending,” he asserts. Yet in 1956 Congress approved $25 billion for the Interstate Highway System. That’s $195 billion in today’s money. Even given the woeful US economy today, is $100 billion really that unthinkable? Martin even lards it up with an emotive yet utterly irrelevant image: “Obama’s firefighter is downloading the design of a burning building on to a handheld device, not knocking on a neighbour’s door to plug a laptop into the local fibre network. In fact, they would probably download it in the fire truck on the way to the building.” Compelling? Sure. Providing you forget that this scenario is about a mobile wireless broadband, not the fixed wireless — bolted to the side of your house — that’d be used for a reliable broadband customer access network. And, while I’m no firefighter, I reckon I’d download those building plans to my laptop well in advance, rather than hope I’d get a good data link from a moving truck. Martin acknowledges that “Broadband Minister Stephen Conroy and other NBN supporters argue mobile service isn’t comparable to the service potential of NBN because mobile capacity is shared among users in each cell site”. True. So why does he then ignores this, and laud the potential speeds of new wireless technologies without mentioning that these speeds are only achievable if you’re right next to the tower and no-one else is using it? As Crikey has reported before, if you’re building a wireless network that’s used by everyone, and you want it to perform as well as fixed broadband, you need a wireless tower in every street. In any event, as Martin says, LTE won’t be rolled out until “mid-decade”. All we’ve seen so far is demonstrations. And it’s not as if the NBN’s speed won’t increase over time as well. In short, Martin is arguing what the opposition is arguing. We shouldn’t build the nation’s telecommunications network using known-good fibre technology that we can buy off the shelf today to deliver a consistent, reliable experience. We should instead wait a few more years and then use wireless technology that we already know will suffer from interference and load-sharing problems. I think part of the problem is that the opposition has been seduced by all the talk of “growth” in wireless. Growth there certainly is. In June 2010, the number of mobile wireless broadband connections in Australia rose to 3.5 million, a 21.7% increase in six months. They now represent 36% of all internet connections, compared with 32% in December 2009. As Crikey has reported before, though, fixed broadband still does the real work. That’s not likely to change, given the different characteristics of fixed and mobile connections. Growth is where investment opportunities lie, particularly those that might deliver major short-term returns. That’s why vendors and investment bankers talk up wireless. The language of growth sounds good to a business-attuned Liberal Party ear. It’s far more exciting that the pedestrian incremental growth and maintenance of core infrastructure, replacing 20th-century copper with 21st-century fibre. But should the nation’s broadband policy be about taxpayer subsidies for profitable short-term business investments? Or should it be about delivering a consistent, reliable communications network? |
|
|
|
15 Comments
The coalition doesn’t know what they are talking about. I have wireless broadand in Five Dock in Sydney and performance is pretty ordinary - better than dial-up but no more than 256. The speed definintely slows in the eveinng when everyone logs on. It is not an option
Thanx for this. One might also speculate why wireless is growing. I at least use wireless at home cos I can’t get much better from the fixed line yet, but when the NBN is available I’ll be onto it like a flash.
My experience with wireless in Melbourne’s Brunswick East (lamentably on the wrong side of Lygon Street for the first NBN instalment) is similar to Ralph’s in Five Dock: fine for the off peak times but frustratingly slow when everyone else logs on.
Congratulations Stil. You’re getting better and better at putting the argument straight. Keep up the good work.
Gavin, let me provide one answer to your question.
Wireless is now the preferred connection of young and not so young people forced to live in shared accommodation. It avoids the old conflicts over who gets to use the phone line next and who pays what share of the monthly bill.
Once a week my daughter arrives with her laptop to get software downloads and changes via my ADSL connection, plus other housekeeping tasks that are far too cumbersome via wireless.
I have a wireless ISP for my laptop for use at other locations in the absence of Wi-Fi. But it is very slow getting a connection and slow on data transfer.
Both would be optimal. There’s times wireless is good. But wireless will not be good if it’s overloaded by all of those people at fixed locations that should be using fibre.
I was a fan of fibre until I read a New Scientist article back in November about WiGi whereby an antenna in a chip, cheaply produced, can be used to direct line of sight wifi broadband in a focused beam thus eliminating most of the attenuation that normal wireless signals suffer. This could be used for stationary transmitters to say building top receivers instead of cable.
Have you read about WiGi and I wonder what you think of it?
Provision of high bandwidth services is a key function made possible by the NBN. The services will be B2B or B2Consumer, including services offered by govt to businesses and homes. Businesss will not do this using wireless. Why not? Same reasons the services dont work well with ADSL: contention and factors affecting real speeds, such as line noise or reception. There are too many downsides and too much risk to entities wanting to provide high bandwidth services.
The NBN offers very reliable, contention-free synchronous connectivity for all. The people failing to comprehend what this actually means in the real world have usually not been at the pointy end of delivering reliable wireless and ADSL to business, particularly for high bandwidth services. Many of the anti-NBN brigade seem to willfully ignore the real world differences between the technologies.
It’s no coincidence that those with the most to lose (FTA and cable TV, mainstream media organisations) run every argument they can invent to try to nobble the NBN. It’s galling for them to see the writing on the wall: the NBN will render many existing media services redundant and irrelevant. But there is huge opportunity in the NBN for those willing to provide the revolutionary services it will make possible. High speed synchronous connectivity for all will reshape our capacity to obtain services and consume information in wonderful ways.
The NBN removes a large chunk of the risk associated with establishing the new services. This is definitely a case of: “if you build it, the services will come”.
Cant say the same about alternative methods.
Well this lightweight piece has the hallmarks of a pro-ALP article rather than one that talks about all the issues. A sort of “Conroy is right and Turnbull is wrong” polemic.
One example is the absurd statement “I think part of the problem is that the opposition has been seduced by all the talk of “growth” in wireless” Mate, the wireless revolution is on. Take a look around you. It’s driven by consumers, not “Short term investors talking up wireless”. Even some of the cheer squad commenters above admit to having wireless services. And while complaints about wireless speeds are legitimate, they can only get faster, significantly faster.
Another example is that Obama would never announce wireless broadband in his state of the union address if it isn’t technically and financially possible.. And it will be done by the private sector as well.
Ipads, laptops, mobiles can only grow and along with it, demand for wireless broadband.
I’m with Mr Pyjama Pudding. Make the NBN business to business. The vast majority of the cost of the NBN is the last km to 10 million homes. This would make fibre broadband commercially viable as well as a fraction of the price.
David,
So there will not be any businesses in houses?
Look closely at the number of mobile wireless broadband connections - don’t just take the number as being peoples preferred choice and therefore growth.
I moved house and couldn’t get any other type of internet connection - it was wireless or nothing. So that’s one connection that wouldn’t normally have been there.
I still have mixed views about the best way to structure and pay for the much needed upgrade to our communications infrastructure, but for the FUDsters out there laws of physics say in application fibre will always win and always be orders of magnitude faster than wireless can ever be.
Attacks looking to equate the two are fundamentally flawed and not based in fact.
With fiber, you get a dedicated data path between two points, and the bandwidth of that data path is limited only by the capabilities of the equipment at either end. Theoretically (up to the speed of light), there is no limit to the bandwidth of a fiber-optic connection.
With wireless, on the other hand, you’re not using a dedicated physical data path; you’re using an RF transmitter and (perhaps multiple) receivers to turn the air between two points as the data path. And while the ability to set up point-to-multipoint communication is a decisive advantage that wireless has over fiber, the bandwidth of a wireless signal is constrained by a number of variables, including the amount of spectrum (the number of frequencies) you are authorized to transmit/receive on), the frequency itself (how many cycles per second you can transmit/receive), and your modulation scheme (how many bits you can push over one cycle).
We need to view the investment as something that will last decades like the copper network has since the 1900’s.
So choice for your place of residence or business ( A known, fixed location that does not tend to move for decades:)
1) An infinitely upgradable direct speed of light based connection
2) A shared, RF limited spectrum that is a constrained, shared resource
Laws of physics back #1 strongly and if you have a known fixed location I would much prefer a fibre connection given the astronomical growth in data every year that we will have access to.
@Tamo: The share-house problem is a function of billing systems, not technology. Fixed internet links are sold as pre-premises not per-person. There’d be nothing stopping an ISP selling a product that offered a “virtual port” for each person in a shared house, with traffic and even speed billed separately. It could be done in cooperation with the vendors of the routers, and would be a software/firmware change, not hardware. Presumably they see no call for doing this.
@JamesG: I hadn’t heard of WiGi and it’s not turning up in a quick search just now. I’ll have a proper look later. My questions would be, though: Is this technology available to purchase today, or still something in the lab? How much does it reduce the effective size of the “beam” so that a frequency can be re-used in an adjacent area without interference?
@David Hand: Thank you. Mate.
@ Stilgherrian
I don’t think we are in any particular disagreement.
My post was an attempt to throw some light on the popularity of wireless internet.
Wireless internet is he partner of the mobile phone. Both are popular with what I call the nomadic generation. Mobile voice and data avoids those problems created by the single landline connection, but at a cost of inferior performance and higher cost. In the early days of my daughter’s mobile phone I taught her about the relative cost/price of voice vs text. The former is high cost and high price, and the latter is really really cheap cost, but at least lower price. I won’t claim that I invented the texting revolution, but…
Which also gets me to the “Presumably they see no call for doing this”. The history of telecommunications is a history of “we know best” so why worry about market research, customer service and marketplace competition.
The claim that the popularity of wireless internet is such that we don’t need landline is a fraudulent argument.
Wireless will be reliable? I still get my mobile phone dropping out in the CBD!!!! and not for internet…just phone calls!!!
Wireless is easily sabotaged by creating lots of loud noise at the same frequencies. This can be done deliberately (ie wireless is not immune to deliberate sabotage from sources external to the premises). But putting that aside, the worst offenders affecting wireless reception are hands-free phone systems and the appalling RFI often emitted by ubiquitous cheap wall warts (power adapters). Cheap wall warts wont be disappearing any time soon.
Anyone currently streaming video via wifi within their home (from their ADSL modem/router) will confirm that the performance is nowhere near as consistent as as wired streaming. Even EOP beats wifi, every time.
Business and govt need maximum reliability and minimum risk in order to invest in provision of services. This also applies to the vast number of businesses located outside of city or suburban CBDs. Many businesses are in homes or near homes. More particularly, many employees work from home (or could work from home due to the NBN). Provision of health and education services via the web will only work with reliable, synchronous connections which are immune from contention, line noise and reception failures. This not only about bandwidth - it’s also about mitigating the much greater risk associated with building those services in a non-fibre world.