Satellite-hating Libs blow policy free kick

Full Duplex

David Braue

A view from the trenches of Australian telecommunications. As the name implies, it’s a two-way conversation and we ask you not to pull any punches ... we won’t.

It's a shame that Shadow Communications Minister Malcolm Turnbull reacted to the news of Labor's impending satellite launch with his usual knee-jerk negativity, since the news actually presented a rare opportunity to turn a Labor NBN policy move into a Coalition policy win.

The biggest issue with the satellite launch is that the birds are being purchased to service just 3 per cent of Australia's households, or around 200,000 premises that lie so far from rural centres that even fixed wireless isn't economical. Turnbull wasted no time before doing the maths: amortise the $2 billion cost across that limited number of premises, and you find out that this part of the NBN will cost over $10,000 per property. That's a lot.


Sputnik kick-started the space age; could NBN Co's satellites do the same for the Liberals' NBN policy? (Credit: NASA and NSSDC)

The thing is: the new Ka-Band satellites have way more capacity than that; the comparable North American ViaSat-1 satellite, based on the same technology, can provide 12Mbps connections to over 1 million premises.

That's a lot of broadband — and it would indeed be a shame to let it lie fallow. However, true to his private-sector-oriented psyche, Turnbull's instinctive response was to look past the numbers and simply blast the purchase of the two satellites as a lamentable waste of money. He also argued that it would flood the satellite market with excess capacity — destroying the private-sector satellite market as NBN Co pushes its excess capacity into new markets to claw back its expenditures. One could almost see him trying to dredge up the old nugget that NBN Co will kill private-sector fibre operators by retailing services to government departments.

It's the same argument that he's been using to fight the fibre to the premises NBN, but with the nouns replaced with satellite-related terms. Turnbull even dug up the old "Rolls-Royce" argument, which is always a semantically difficult comparison, because it implies that Australians only deserve a Hyundai Getz solution.

The common goal of the NBN, lauded by both Labor and Liberal politicians alike, is to bring broadband connectivity to areas that don't currently have it. Conveniently enough, satellites have coverage beams that extend well past the boundaries of this country — which means that the day they're switched on, Labor's new satellites will provide enough capacity to bring 12Mbps broadband services to not just 200,000, but over 2 million Australian homes.

The day they're switched on, Labor's new satellites will provide enough capacity to bring 12Mbps broadband services to not just 200,000, but over 2 million Australian homes.

That's an entirely different kettle of fish. Amortise the $2 billion across 2 million premises, and you're suddenly paying just $1000 per property. And that's pretty good, actually — only slightly higher than the amount that Kevin Rudd put as cash into everybody's hands a few years ago, and it would fix our most glaring broadband deficiencies.

Latency is higher on satellites, sure, so gamers wouldn't be impressed, but, as anybody will concede, some connectivity is better than no connectivity. And those two satellites are surely much cheaper than rolling the fibre NBN to remote areas, or even the bother of installing thousands of terrestrial towers to support a fixed-LTE broadband network that will deliver a similarly performing 12Mbps service.

Granted, it's not the 100Mbps or 1Gbps that the fibre NBN will provide, but it's still something. It's certainly enough capacity to not only provide decent broadband to remote Aboriginal communities hours off the beaten track, but to also plug coverage black spots in urban areas that paradoxically are still struggling along at dial-up or slightly better speeds.

In other words, it's a quick-fix solution and policy free kick, the likes of which most politicians dream of — and Labor placed it right in Turnbull's hands to be promptly fumbled. Had he taken a few moments to think laterally rather than just giving into his basest reflex instinct and not-invented-here party line, Turnbull could have come up with a pleasant and surprising policy winner. What if he'd said something like this:

It's ironic that Labor has turned to satellites to deliver the same kind of wireless services to the same places OPEL would have served. However, we welcome the government's investment in Ka-Band satellites, because it will immediately provide a 12Mbps baseline service to every Australian premise.

In so doing, it will obviate the need to roll out fibre to nearly 2 million of Australia's most remote properties. It will also expedite the delivery of services to urban black spots, which will gain a basic level of service that has been unavailable to them until now.

Most importantly of all, it will allow Labor's inordinately expensive fibre roll-out to be suspended and refocused. Rather than forcing a one-size-fits-all solution on every Australian, availability of 12Mbps satellite services as a common baseline will allow a Liberal or Labor government to pause the NBN roll-out, reassess the FttP business case and prioritise fibre or other infrastructure investments in the areas that really need them. In this sense, it is an exceedingly promising investment that provides far better value for money than Labor's current fibre-NBN white elephant.

Had Turnbull come out with something like that, he would have been able to parry the announcement while repositioning his party's policy as a rational step forward. His actual response, however — to stick with ageing satellites that are far slower and mostly obsolete, even in telcos' eyes — just comes off as predictable and wan. It's like someone's grandfather explaining why he wants to keep his 1950s-era carpet and curtains, and jealously hoards his collection of old bottle caps. Sure, they have value to him — but he can't expect everyone else to get as excited as he is.

Having access to a reliable if not warp-speed service would give all of Australia a leg up into the 21st century. And once all Australians have access to such a service, the government could let the private sector — with or without government support and guidance — work out the details on how to proceed from there.

What do you think? Could Labor's satellites be better utilised to immediately plug Australia's broadband black spots? Or are they just a pointless distraction from the real game?

Talkback

Even you will admit that spending that kind of money servicing 200,000 premises and close to $50 billion to satisfy the need of the rest even though the satellite can do the whole job, is lunacy in this day and age.

Vasso MassonicVasso Massonic February 10th, 2012
Report offensive content Reply (+4) (-66)

"...even though the satellite can do the whole job"????

Now I know you've gone off the deep end. Even these satellites, with a capacity of 90Gbps each, could only provide the tiniest fraction of the required broadband capability for all of Australia.

Let's see......10 million premises divided by 180Gbps..... That's 0.01Mbps each. But, let's be fair and assume an industry-standard 20:1 contention ratio. That's 0.07Mbps each. Well, I guess it's better than dialup. Should be perfect for the conservative side of Politics.

HazTechDadHazTechDad February 11th, 2012
Report offensive content Reply (+32) (0)

Vasso, welcome back! The $37.8 billion, ten year construction and operation budget of the NBN includes the design, launch and operation of these two satellites. They are not extra. They are part of the optimal technology mix to deliver the KPIs of the project, i.e. at least 12 Mbps uncontended bandwidth to all Australian premises.

umbriaumbria February 13th, 2012
Report offensive content Reply (+29) (0)

I don't think 'plugging' australia's urban black spots is the answer. iIf that was the intention, why not just re-lay copper cable in those areas. If it's econimically possible and I'm guessing it is in urban areas, then they should use a state of the art scalable solution such as Fibre. Only in the areas it's simply not possible or financially feasible should they choose another method such as Satellite.

gleffgleff February 10th, 2012
Report offensive content Reply (+10) (-1)

i.e, the NBN. It comprises fibre to premises where this is cheaper to provision than wireless, fixed wireless to catch outlyers up to the point where the cost of towers and backhaul become prohibitive, and universal satellite coverage. Full details of the methodology to arrive at the footprints for fibre and wireless is available in the May 2010 NBN Implementation Study, for which we taxpayers paid $25 million - money well spent to have all the facts and costs needed to get this project right.

umbriaumbria February 13th, 2012
Report offensive content Reply (+22) (0)

Your extrapolation to $1000 per property does not take into account that the end user equipment costs about $5000 per site so you should state $1000 plus $5000 per property.

Dave of NakaraDave of Nakara February 10th, 2012
Report offensive content Reply (+3) (-31)

Yeah, and let's not forget the $4000 in services costs per site to get it in. Thats $10K.

The we have 20% great big new carbon tax, that takes it to $12K per site.

As Labor are screwing the economy no matter what they do we will need to add another 25% great big new white elephant tax cost, that takes it to $15K per site.

Now as Labor only ever lie, we gotta add 50% cos the filter is going slow down the internet by 87%, so now we already have $22,500 per site.

So that is $22,500 of MY money per site so that jmill can pick up a few emails a week???

Nope, do not want it. I would much prefer to stay on my old and trusty 14.4K modem, it cost nearly nothing!

OckerOcker February 10th, 2012
Report offensive content Reply (+3) (-53)

And let's not forget that the sky is probably gunna fall in so there's another pile of cash down the drain.

And the NBN modem sitting on the floor is probably gunna affect the resale value of my house??

omegaomega February 10th, 2012
Report offensive content Reply (+34) (-2)

Ocker this is a comms blog.

So if you wanna campaign for a political party (particularly the one you endorse), please go over to BigPond or The Australian where they will welcome you with open arms and leave the rest of us here (trying to look past the idiocy of politics) to discuss the more pressing comms issues, thank you.

BetaBeta February 11th, 2012
Report offensive content Reply (+24) (-5)

Seems like the only thing NBNCo groupies are reticent on is the very subject of communications networks.

Mark Addinall.

addinall.maddinall.m February 12th, 2012
Report offensive content Reply (+1) (-46)

Mark two things.

The good news first...

You have cutting edge ideas.

Now the sad truth...

Pity it's not longer 1982, when they could have been used.

Just 30 years too late, never mind!

BetaBeta February 12th, 2012
Report offensive content Reply (0) (0)

Beta I think you might have missed the sarcasm in Ocker's post...

SupertedSuperted February 13th, 2012
Report offensive content Reply (+10) (0)

Yes I had hoped afterwards that no one noticed, but alas... d'oh.

Problem of course being, there are those who actually believe and "would" say that, as witnessed @ comment #1...

BetaBeta February 13th, 2012
Report offensive content Reply (+9) (-1)

We have satellite internet at the moment as we are in a black spot in western NSW.
We find it OK.How fast do we need down load to fetch my emails.My children and my grand children and my great gran children. will be paying for this white elephant.
JM

jmilljmill February 10th, 2012
Report offensive content Reply (+1) (-52)

Are your children, grandchildren and great grandchildren 'fetching emails' are they?

For every 1 dollar spent on the NBN, Australian tax payers will receive $1.07.

omegaomega February 10th, 2012
Report offensive content Reply (+40) (0)

Your children and grand children will thank you for your vision of envisaging that the only use for it is to download emails....

CammCamm February 13th, 2012
Report offensive content Reply (+31) (0)

...satellite isn't a solution for everyone. VOiP telecom is big here in North America, I use Vonage as my primary phone, and the huge bandwidth offered is no solution to the packet delay created by a geo-stationary satellite round trip time delay.. usually 1/2 up to 1 second. Most satellite phone calls wind up being more like 2 way radio contacts..one party yakks for a bit then says 'over' to the other ...an animated conversation is pretty much impossible.

MortimerSnerdMortimerSnerd February 10th, 2012
Report offensive content Reply (+15) (0)

I live in suburban Melbourne, not within 3km of an exchange. On a good day my Internet is about 3.8Mbps. It used to be closer to 5 but as time goes by it is getting worse. I know that for me the NBN could be ten years away, possibly never if the great wrecker gets elected. I would welcome a 12Mbps satellite link to tide me over, especially if coupled to an ADSL 2+ uplink.

GregEGregE February 10th, 2012
Report offensive content Reply (+8) (0)

I think David hit the nail on the head pretty much. Even as a Liberal supporter, I'll vote Labor just to get the continued investment into the NBN - Abbott's short-sightedness regarding this and Global warming is annoying to say the least. Politicians are too focused on arguing the issue then actually coming to some sort of solution. God forbid two opposing parties could actually work towards a common goal.

Vasso - Satellites cannot do the whole of Australia - that is lunacy. It's no one-size fits all solution. If this technology hasn't changed somewhat, it is a 'download only' service, and needs to be supplemented by a fixed line connection of some description, my knowledge of this may be out dated though.

GrahamKGrahamK February 10th, 2012
Report offensive content Reply (+38) (0)

What I've been wondering is if NBN Co can lease the excess capacity on the satellites to other players in the SE Asia/Pacific region and get some additional revenue that way?

redroverredrover February 10th, 2012
Report offensive content Reply (+1) (0)

+1, Very good point

BetaBeta February 11th, 2012
Report offensive content Reply (+1) (0)

I'm not sure if you were aware, but this was discussed on the Whirlpool forum. The general discussion led to the fact that while this may be possible, you would need to get hardware into the area to receive the signal, and then you would need to get electricity into a majority of the area to power the hardware.

Long story short, the take up rate would be small and would cost Australia more than it's worth, and most Australian voters would much rather keep money helping our own ecconomy.

djdoverdjdover February 13th, 2012
Report offensive content Reply (+5) (0)

David, the distraction is Turnbull's incompetence. Turnbull now believes he can predict 30 years into the future because he believes that all the unimaginable benefits from the NBN can be achieved with copper??? http://www.abc.net.au/iview/#/view/889941

Pretty clever...

When will someone conduct an analysis on average data speeds (past and present) and make an estimation on what it will be in 10 years time? Glenn Stephens of the RBA says that the best way to predict into the future is to look at the past.

It seems that on this topic we are looking back at history and saying, we're better off back there?? If Australia does get left behind, I wonder who we'll blame?

omegaomega February 10th, 2012
Report offensive content Reply (+25) (0)

The historical bandwidth demand growth is often documented. For instance:
13 October 2010, see slides 11-12 of http://www.nbnco.com.au/assets/presentations/commsday-melbourne-13-oct-10.pdf (includes 2009-10 ABS data)

17 November 2011 - see slide 4 of http://www.nbnco.com.au/assets/presentations/commsday-melbourne-13-oct-10.pdf (includes June 2011 ABS data)

umbriaumbria February 13th, 2012
Report offensive content Reply (+21) (0)

Glad you asked that redrover, I was going to ask that myself.

clive49clive49 February 10th, 2012
Report offensive content Reply (+11) (0)

David, your article is so poorly written and one sided that either you're incompetent or your post is a troll to gain plenty of hits for ZDNet.

Where has it been identified that the current satellite systems don't provide the coverage or the speed? If a problem hasn't been identified, then why is the government throwing billions at it to "fix" it?

I don't think Labor nor the Liberal's NBN / broadband plans are perfect and both have significant flaws in their assumptions and planning. It's a shame that Australia has had such poor IT ministers over the last 10+ years, we're just set to continue to waste money and miss the mark.

tjbtjb February 10th, 2012
Report offensive content Reply (0) (-36)

Please enlighten us all, what is the mark?

omegaomega February 10th, 2012
Report offensive content Reply (+16) (0)

This could be something you might want to read tjb: http://delimiter.com.au/2012/02/11/the-truth-about-nbn-cos-satellite-needs/

"The result of that investigation, Dawson said, led NBN Co increasingly to the answer of launching its own satellites. “The more and more you go into it, the more and more systems engineering you need to go into to use what little capacity there is anyway,” he says. “We have gone through this upside down and backwards. We know where there’s capacity.”

Solving the problem through leasing capacity, Dawson said, “just didn’t add up”. “The maths is pretty straightforward,” he added. “Network architects, network engineers do the maths [and] they know how to design networks. That’s what we’ve been doing for the past two years. When you start to examine what capacity they’re talking about, the answer starts to become obvious.”

CammCamm February 13th, 2012
Report offensive content Reply (+21) (0)

I see the significance of the NBN as being equal to building railways, ports and surfacing the roads. The efficiencies in the economy of having such blistering broadband speeds can't be underemphasised.
All this ridiculous talk about cost benefit analysis shows a misunderstanding as to the true benefits of having a 12Mb satellite link to the outback and 100Bb fibre link to urban areas. This is nation building stuff that can't be ignored much less "torn down".
Technologies like this may go a long way to mitigating the daily traffic gridlock in our cities by making telecommuting a reality for many people. The savings in greenhouse gas emissions alone if people only attended the office 2-3 days a week would be tremendous.

H.DigitalisH.Digitalis February 10th, 2012
Report offensive content Reply (+34) (-1)

You sed() ... "Granted, it's not the 100Mbps or 1Gbps that the fibre NBN will provide, but it's still something. It's certainly enough capacity to not only provide decent broadband to remote Aboriginal communities hours off the beaten track, but to also plug coverage black spots in urban areas that paradoxically are still struggling along at dial-up or slightly better speeds."

Why are we still condemming a large portion of the Australian public with sub-standard communications when it has been costed and proven that fibre can be extended out to rural and remote communities in a timely fashion, and reasonable price.

The web experience for satellite users is very poor, apart from watching movies, for which it is well suited. Anything that uses a Web 2.x paradigm or AJAX transactions is going to be painful to use. Even CSS and to the point, CSS3 web applications, so lovingly crafted over fixed line and wireless, are going to give a very poor result in the satellite world. What happens to an onMouseover or onHover event when some 20-200 HTML objects are sending information to the application when the latency is around 1 second? You are going to have to drag that mouse around pretty slowly to see the application working. What about keyboard events? Same deal. Regardless of how sexy a $2 BILLION dollars shopping spree looks, a satellite is just a repeater. Nothing more, nothing less. And the geosynchronous repeaters are parked quite a way away. Those of us that have been involved in the design and deployment of satellite systems understand the limitations of this communications media, and it is rather surprising that a ‘nation building’ mega spend is going to leave our mates in the bush with a poor solution. A mouse event on your iPod traveles from the iPod to your house wireless router, then to the NBN terminal equipment, then SHOOSH! via satellite 35,786 km UP to the satellite, then 35,786km DOWN to the application provider (and a few networks, routers, switches come in play during this level of the hop), the AJAX transaction returns the result of ‘SELECT account_name FROM clients WHERE account_name LIKE %ADD%” back UP 35,786km then back DOWN to Joe in the bush, some 35,786km away. That’s 143,144 km not counting terrestrial hops. The latency is a killer, regardless how new the space-based repeater. Several techniques have been investigated and invented to help overcome this problem, application engineering, modified TCP-IP stacks (SCPS-TP), and user education in exactly what to expect from satellite communications services. TV looks fine over satellite. Once the first chunk-O-data has been recieved, stream away. TCP/IP has a limitation in the packet size being sent, the ACK request, NACK re-transmission, packet ordering and the TCP “slowdown” that occurs when packet loss reaches limits. New applications are “chatty” with every HTML object having a potential communication/conversation with the application. And as the “chatty” transactions are usually quite small, more overhead is generated for the network to support. This is reality. NBN groupies are fond of mentioning Physics, well here it is. Latency sucks. People wanting to play games or use Web 2.x services are going to be left out (again).
There is an assumption in the article that anyone who is not an NBN groupie, is likely to argue that those out bush do not really NEED fast internet. I don’t know your social circles, but for myself and my peers, the opposite is true. I argue against the NBNCo model as it is NOT providing suitable services to remote users. What is the alternative? This has been modelled and costed in the past. Building the NBN on a FTTN architecture (using FTTP where required) provides a homogeneous network diagram. The main high-speed trunks are run out to node using existing fibre in the ground, with extensions and expansion as required. During 1999-2000 I build satellite Earth stations in remote Arnhem Land in aboriginal communites because that network architecture was the only choice. Recently, Telstra, the NT government and Rio have laid OFT into the same areas providing FTTN (and some FTTP) connectivity into these VERY remote locations. All the way over to Groote Island now.

“Arnhem Land connects to high speed broadband
On 2 December 2009, Telstra announced the completion of one of the largest optic fibre and broadband
infrastructure projects undertaken in recent times – the Arnhem Land Fibre Project.
The $34 million project, which received funding from the Northern Territory Government and Rio Tinto
Alcan and valuable support from the Northern Land Council, connects nine Indigenous communities and the township of Nhulunbuy to the nation’s fibre optic backbone. Stage 1 of the project (completed November 2008): incorporated laying over 800km of fibre optic cable between Jabiru and Nhulunbuy and building associated telecommunications infrastructure to provide access to world class, high speed broadband. Stage 2 of the project (completed November 2009): required a further 190km of fibre and five radio systems to each of the islands. Due to the project, high-speed broadband and services, equivalent to those in our largest cities, are now accessible to some of Australia’s most remote communities – connecting approximately 10,000 people. The project, which was completed on time and on budget, saw over 800km of fibre optic cable laid across fragile terrain, in difficult climatic conditions between Jabiru and Nhulunbuy.”

Seems strange the NBN wants to spend another $2 BILLION dollars to provide satellite broadband to:

1. Remote communities that now have Fibre
2. Remote communities that already have satellite broadband.

Telstra prepared a detailed case during 2009, and as part of the executive summary stated:

“Telstra estimates $250 million in funding, if allocated solely to extension rather than duplication, could connect 140 communities in Australia to the high- capacity, national transmission networks, removing barriers to deployment of high speed fixed and mobile communications. Some of these communities are the most remote and isolated in Australia. ”

The report documented these 140 areas as the rural and remote communications ‘blackspots’ in Australia and had a plan that would bring these locations high-speed LOW LATENCY network connectivity for a $250 MILLION investment.

And the model was proven to work on the Arnhem Land Project.

Full report here:
http://www.addinall.net/nbn/Telstra-2.pdf

‘Future Proofing”. We are all aware I take it that the life expectancy for a modern geosynchronous satellite is in the order of 15-18 years right? If we launch in 2015 they are the way back down by 2030-33. What then?

Ka band communications. I find it somewhat surprising that during the extensive research on satellite communications in Australia and worldwide that Ku band didn’t get a mention. Never mind. Ka band is the new(ish) kid on the block.. It has however some very real problems when it comes to rain and dust

High frequency bands like Ka-band are starting to be used in more often in satellite communications networks. The Ka band (30/20 GHz), however, the effect of or dust rain attenuation is more significant than in lower frequency bands, such as the popular Ku band (14/12 GHz) and Charlie band.
Some countermeasures for rain attenuation using site and/or satellite diversities are required in order to operate reliable Ka and satellite communication links. In a human sense this means over-engineering of the transmit/receive stations. If the network topology followed the homogeneous FTTN pattern, then large Earth stations would be built replacing the Fibre portion of the network feed with the satellite system. Over-engineering (dish size, computational power, energy requirements etc) can be made at a node. This is less likely when the satellite system is conversing with a home installation.

Given a decent network architecture that is FTTN at a national level, whilst still being able to use ‘best fit’ technologies in individual circumstances would reduce the number of people relying on satellite communications from “several hundred thousand” to perhaps 100,000 or less. These poor folk can be serviced by purchasing bandwidth from new space missions such as Jabiru-2 and other sophisticated Ka Band birds soon to fly, removing the requirement of the NBNCo to be the owner of space missions.

I really feel that the money spent on this portion of the NBN model is not money well spent, and the network is poorly planned.

Lament ye not the poor ol' Abos. I would have thought ZDNet would have been aware of these situations.

Cheers,
Mark Addinall.

addinall.maddinall.m February 11th, 2012
Report offensive content Reply (+3) (-77)

Do you really think anyone will be bothered reading your stupidity Mark?

Nice racism at the end though, as usual...!

BetaBeta February 11th, 2012
Report offensive content Reply (+27) (-1)

You NBNCo groupies sure are blessed with added vitriol. I suppose it makes up for a lack of education.

"Beta"?! WTF?!

Cheers,
Mark Addinall.

addinall.maddinall.m February 11th, 2012
Report offensive content Reply (0) (-66)

It's funny how when I create a OnMouseOver event on the ZDNet website, the website doesn't refresh, and the web site is still responsive?

How many remote communities are using terminal services?

I'm struggling to find the mouse event for my iPod though?? Why not call it a web request? Mr intelligent...

omegaomega February 11th, 2012
Report offensive content Reply (+6) (0)

"It's funny how when I create a OnMouseOver event on the ZDNet website, the website doesn't refresh, and the web site is still responsive?"

Now, that started out as a statement, and carried through to a question mark. What is it exactly you are trying to say here?


"How many remote communities are using terminal services?"

What are "terminal services"?
Regardless, dunno. Why?

"I'm struggling to find the mouse event for my iPod though?? Why not call it a web request? Mr intelligent..."

Both Android and IOS produce onMouse events that are mapped to the devices OS "touch" events. That is why you can surf the Web on your new device, and Web sites still work. Mostly. The onMouse events are generated a little differently between Android and IOS, and the different browsers do not always react in the same fashion.

"omega"!? ROTFLMFAO!

Mark Addinall.

addinall.maddinall.m February 11th, 2012
Report offensive content Reply (0) (-55)

"You are going to have to drag that mouse around pretty slowly to see the application working."

My point is you sound a little confused on exactly when a device communicates with a remote web server. That is why, for example, you can "click" in the search field on this web site and the web site doesn't refresh. Not all click or touch events are "url requests" regardless of the platform.

Unless of course you use a terminal services or remote desktop technology (but this is a completely different communication technique anyway).

Feel free to dribble on about latency all you want but you must admit that your lack of knowledge on web software development makes your comment quite misleading.

omegaomega February 12th, 2012
Report offensive content Reply (+25) (0)

Why do people assume that once the 93% of FTTH coverage is completed, that will be it forever? Just as there were still dirt roads being sealed for the first time last year - even though sealed roads have been around for decades - there will still be areas that received wireless or satellite in the initial build which will get fibre later on.

Of course it would be great if everyone fibre to start with, but imagine the field day the Liberals would have with the cost of that?

ChuqChuq February 13th, 2012
Report offensive content Reply (+23) (0)

Thanks for drawing attention to the deficiency of broadband via satellite in the bush. Beyond latency, the big issue in remote Australia will be that satellitea upload speed of 1mb effectively rules us out of participating in the digital economy. Given there's not much of an economy on remote Indigenous communities, this adds to the vectors the create and sustain poverty.

DavidN4DavidN4 February 13th, 2012
Report offensive content Reply (0) (0)

Mark, I am currently building a proper web 2.0 application with node.js and backbone.js the 1-2 second satelite delay will not be a problem, all the interactive stuff happens on the client ie in the browser. Applications generally support running on wireless lan's that will at times dissappear.

MunixMunix March 28th, 2012
Report offensive content Reply (0) (0)

"You are going to have to drag that mouse around pretty slowly to see the application working."

That is still correct. Any reasonable engineer will tell you that satellite broadband sucks for modern usage. Games are out of the question, unless like me you favour Scrabble and Chess. "Rool Time Video Conferencing" is something you NBNCo groupies shout for our coming Barve New World (apart from the fact that the technology is fourty years old). Well, try video Skype over satellite. Take a hint from my Army days and learn how to say "OVER" as a grammatical termination punctuation. I suppose one could go back to the Telegraph and use "STOP" but "OVER", "OUT" and "OVER AND OUT" have greater semantic use.


"My point is you sound a little confused on exactly when a device communicates with a remote web server. That is why, for example, you can "click" in the search field on this web site and the web site doesn't refresh. Not all click or touch events are "url requests" regardless of the platform."

Duh. I deserve a DipEd for this. As HTML has progressed from the CERN lab's many moons ago, now every HTML object CAN have events linked to the object. That does not follow that every HTML object HAS to have an event associated with it. The Web 2.0 paradigm is presenting Web pages (or applications) that are more "chatty". In this we mean that more of the HTML object can, and do continue atonomous communications transactions with the application server.

Some instances of events that are tied to objects share a semantic representation between Web 1.0 and Web 2.0 paradigms. A good example is an onMouseover or hover event where a portion of anchor text may be given an underline attribute, or have the background colour change. These types of event driven changes are usually handle by CSS and the processing is carried out on the client, therebye presenting no additional network traffic. Web 2.0 applications on the other hand are using the autonomous OBJECT->SOME_REMOTE_APP conversation model more and more. This can be seen where an onMouseover event generates a shadowbox like "details" concerning the location where the event takes place. If this information is static, it can be held in a variable or a hidden div that is presented by manipulation of the DOM. However, if the information required is dynamic (becoming the case in modern Web apps) then typically the event tied to the object will initiate an autonomous AJAX transaction(s) with one or more servers. This small packet exchange puts real overhead onto a network with a high latency. Capiche?

"Unless of course you use a terminal services or remote desktop technology (but this is a completely different communication technique anyway)."

No **** Sherlock?

"Feel free to dribble on about latency all you want"

Mate, latency is Physics. It is not going away because you or your doltish heroes would wish it so.

"but you must admit that your lack of knowledge on web software development makes your comment quite misleading."

Run along sonny.

Mark Addinall.

addinall.maddinall.m February 12th, 2012
Report offensive content Reply (+1) (-55)

According to my back-of-the-envelope calculation the round trip geostationary satellite latency for communicating between two locations in australia is approximately 240ms path latency + local network latency (10-20ms?). Based on pinging various offshore hosted websites this seems to be comparable to the latency for a fibre link between Australia and the US and less than the latency between australia and most other parts of the world (because of our most international traffic is routed through the US anyway).

Correct me if I'm wrong but by this logic the latency experienced by satellite customers would be fairly reasonable for national services (eHealth, distance education) and about 500-700ms for international services, which while pushing it a bit for real-time applications, isn't the end of the world.

redroverredrover February 12th, 2012
Report offensive content Reply (+8) (0)

"According to my back-of-the-envelope calculation the round trip geostationary satellite latency for communicating between two locations in australia is approximately 240ms path latency + local network latency (10-20ms?)."

Very fair indeed. 35,786 km UP and 35,786 DOWN + 20ms -> 260ms. You compare this to a ping in the USA. The program ping shows a RTT between you and a remote site. So, if I ping() www.ucsd.edu I should expect 2 trips of 11,500 km (23,000km) at c - (index of refraction degradation)(or about 200,000kmps) or
about a minimum 120ms round trip. Of course in real life it is slower than this.

RTT is 'Round Trip Time". If I were to ping www.ucsd.edu using a satellite service, the round trip will be

UP -> to the satellite
DOWN to the satellite
DOWN

addinall.maddinall.m February 13th, 2012
Report offensive content Reply (0) (-22)

It's not clear what your argument is here.

I completely agree with you about the path latency of a fibre link to the US being about 130ms but the total latency, empirically, is about double that. Presumably mostly due to numerous instances of electronic 2R or 3R signal regeneration along the submarine link, in addition to multihop router delays. The satellite latency should be a bit closer to the theoretical because it's just two distinct hops. However, I do realise that the effective performance will be influenced by the signal-to-noise ratio of the link, which will fluctuate depending on atmospheric conditions.

I wonder if Hawaii is within the satellite's coverage area and if satellite service international traffic could be beamed from/to a transmitter/receiver there to avoid half the submarine link.

redroverredrover February 13th, 2012
Report offensive content Reply (+1) (0)

Your doing a fine job of discrediting yourself but let me take it a step further.

H.264 is not 40 years old. Your video doom and gloom is misguided. Video compression techniques are constantly improving making HD video transmission faster. Improving the network hardware also helps.

The 'A' in Ajax stands for asynchronous. I'm sure you understand what this means.

"However, if the information required is dynamic (becoming the case in modern Web apps) then typically the event tied to the object will initiate an autonomous AJAX transaction(s) with one or more servers. This small packet exchange puts real overhead onto a network with a high latency. Capiche?"

Again this is misleading. Ajax enabled applications handle most dynamic events, such as data validation and manipulation, via the javascript that is loaded when the page first loads. At the same time that it is making display changes for the customer, it is sending data back and forth to the server. But the data transfer is not dependent upon actions of the user. Thus, very little lag on the application.

Ajax is designed to decrease latency by making small transactions to the server asynchronously, instead of your dramatised hour glass lag beat up.

Forget about the DipEd, you need to be a good communicator for that, and it requires talking about things other than yourself.

Run along tiger...

omegaomega February 12th, 2012
Report offensive content Reply (+24) (0)

"Your doing a fine job of discrediting yourself but let me take it a step further."

You sure of that sonny?

"H.264 is not 40 years old. Your video doom and gloom is misguided. Video compression techniques are constantly improving making HD video transmission faster. Improving the network hardware also helps."

You are the only one to mention H.264. I said "Video conferencing".
The ability to make a video call between two or more people, MKay?

"In early 1936 the world's first public video telephone service was developed by Dr. Georg Schubert and opened by the German Reichspost (Post Office) between Berlin and Leipzig, utilizing broadband coaxial cable to cover the distance of approximately 100 miles (160 km). The system employed mechanical television scanning and 8 inches (20 cm) square displays with a resolution of 180 lines operating at 25 frames per second.[30] Its opening was inaugurated by the Minister of Posts Paul von Eltz-Rübenach in Berlin on March 1, 1936, who viewed and spoke with Leipzig's chief burgomaster.[31][32] The same coaxial cables were also used to distribute television programming throughout Germany."

That makes Video calls 76 years old.

"The United States would not see its first public video telephone booths until 1964, when AT&T installed their earliest commercial videophone unit, the Picturephone Mod I, in public booths in three cities: New York, Washington, D.C. and Chicago.[27] Picturephone booths were set up in New York's Grand Central Terminal and elsewhere. Picturephones were also installed in the offices of Westinghouse in Pittsburgh, and in other technology companies."

That's the one I was thinking of. Fourty eight years old.

"The 'A' in Ajax stands for asynchronous. I'm sure you understand what this means."

I surely do. Do you have a point?


"However, if the information required is dynamic (becoming the case in modern Web apps) then typically the event tied to the object will initiate an autonomous AJAX transaction(s) with one or more servers. This small packet exchange puts real overhead onto a network with a high latency. Capiche?"

"Again this is misleading. Ajax enabled applications handle most dynamic events, such as data validation and manipulation, via the javascript that is loaded when the page first loads. At the same time that it is making display changes for the customer, it is sending data back and forth to the server. But the data transfer is not dependent upon actions of the user. Thus, very little lag on the application."

What? If I mouse over a control that solicits a response from a remote server, of course the latency time is going to have an effect on the user perception of the application! Idiot. Add ANY authentication handshaking around that transaction and the latency becomes so disturbing that many VPN or SSH applications just stop functioning.


"Ajax is designed to decrease latency by making small transactions to the server asynchronously, instead of your dramatised hour glass lag beat up."

AJAX is NOT designed to decrease latency at all. You have no idea what you are talking about. Depending on the nature of the Ajax application, dynamic page updates may interfere disruptively with user interactions, especially if working on an unstable Internet connection or on an Internet connection that has a high network latency. A user of an AJAX transaction may 'click' on a likely looking control in anticipation of a response. On a network connection with high latency, the user may have gone on to other processes or procedures by the time the AJAX request is returned to the calling object.
Web 2.0 or AJAX rich web applications increase the NUMBER of transactions the client communicates with the server(s) On a network that already has a slow latency performance, this engineering compounds the responsiveness of the application and the user experience.


"Forget about the DipEd, you need to be a good communicator for that, and it requires talking about things other than yourself."

Have to make do with my other Post Grads then. Teaching Primary seems not to be my forte'.

Mark Addinall.

addinall.maddinall.m February 13th, 2012
Report offensive content Reply (+2) (-38)

I mentioned H.264 because you may not realise and can be used to speed up the transmission of video, especially with "video conferencing".

Not sure where you were going with this anyway? Unless you were suggesting video calls haven't improved since they were invented?

"Ajax is designed to decrease latency by making small transactions to the server asynchronously, instead of your dramatised hour glass lag beat up."

Sorry, I should have said counteract the perception of latency to the user. And also disprove your argument.

"What? If I mouse over a control that solicits a response from a remote server, of course the latency time is going to have an effect on the user perception of the application! Idiot. Add ANY authentication handshaking around that transaction and the latency becomes so disturbing that many VPN or SSH applications just stop functioning."

That's why we AJAX enable web applications, so mouse over events can be embedded in the java script loaded with the page, so it doesn't make a call back to the server and the page remains responsive to the user. And don't just believe me, please see below.

"Web 2.0 or AJAX rich web applications increase the NUMBER of transactions the client communicates with the server(s) On a network that already has a slow latency performance, this engineering COMPOUNDS the responsiveness of the application and the user experience."

It's ok if you want to be stuck back in the old days where you could talk the ear off someone with your many amusing stories, but don't come to a forum like this arguing communications improvements are meaningless because you can think of disadvantages. You clearly bend the truth to suit yourself. There's been quite a few self absorbed people like you in this forum. But we're slowly weeding them out...

omegaomega February 13th, 2012
Report offensive content Reply (+27) (0)

+1

AJAX doesn't increase or reduce latency - latency is completely network related.

Good use of AJAX masks latency for the user by reducing the size and number of trips to the server though.

Also Mark, you're talking about a small percentage of the Australian population that will either be kept on copper or has existing satellite access.

$640m for the government is a drop in the ocean. It "IS" the Holden Ute for outback, because a Bentley simply wouldn't survive.

kachaukachau February 13th, 2012
Report offensive content Reply (+22) (0)

One of the big benefits reported by remote (6 Mbps) interim NBN satellite service users is that ABC TV IView and other catchup TV services now stream flawlessly. Another user comments on his two-year-old grandson pulling faces at Grandad over Skype. Sure, the 480 ms round-trip delay for geo-stationary satellite audio is a burden for voice calls on VoIP, but you can spend hours talking for next to nothing, compared to $90 per month for unlimited STD calls on a landline now, which doesn't even cover broadband or television costs for whatever service is available in the boondocks.

umbriaumbria February 13th, 2012
Report offensive content Reply (+21) (0)
Add your opinion

In order to post a comment, you need to be registered. (Sign In or register below)

Post your comment

Terms of Service - As a ZDNet registrant, and by using this service, you indicate that you agree to our Terms and Conditions and have read and understand our Privacy Policy.

ZDNet Australia Live

Before accusing me of fudging the figures, that was the percentage in April, the latest available. It seems that as the advantage of the ...

50 minutes ago by GregoryB1 on NBN FUD: will Abbott ever learn?

Currently about 50% of connections are at the 100Mb/s rate.
As a consequence, ARPU is significantly higher than the projected figures.

58 minutes ago by GregoryB1 on NBN FUD: will Abbott ever learn?

Currently about 50% of connections are at the 100Mb/s rate.
As a consequence, ARPU is significantly higher than the projected figures.

58 minutes ago by GregoryB1 on NBN FUD: will Abbott ever learn?

Wireless currently carries less than 2% of total internet data traffic. Simply to carry the existing traffic, we would need 50 times the ...

1 hour ago by GregoryB1 on Blowing the digital dividend on wireless NBN

The stupidest part about a wireless solution for the burbs is that it will actually cost more to put an antenna on the roof to get the si...

1 hour ago by GregoryB1 on Blowing the digital dividend on wireless NBN

The problem is not range of the cell in the urban areas where Turnbull wants LTE instead of fibre, it is the number of users. In urban ar...

1 hour ago by GregoryB1 on Blowing the digital dividend on wireless NBN

After the Second World War, the pursuit of pleasure domains the entire world atmosphere, Lancel (Lancel) to adapt rapidly into the demand...

2 hours ago by PokArrackpask on Spam sees Westnet blocked by BigPond

RT @DellEnterprise: Dell Secureworks talks with ZDNet about Android's biggest #security flaws - http://t.co/LSFLQVFq #infosec

NBN users opt for 100Mbps: Customers are picking the top fibre plan that is available on the National Broadband ... http://t.co/sjtFSU3g

"Customers are picking the top fibre plan that is available on the National Broadband Network (NBN), more than a... http://t.co/M3P24Htn

Another thing I found so misleading here is the step on how you assume to make the USB bootable . (The NTLDR needs to be renamed to USBNT...

4 hours ago by WindowsAnalyzer on Boot Windows XP from a USB flash drive

You can also use the help of these links, just incase your stuff failed, I probably got Windows build by using the Pebuilder as per the i...

4 hours ago by WindowsAnalyzer on Boot Windows XP from a USB flash drive

RT @CorrieB: An iPad for every child: Inevitable or impossible? http://t.co/I7uS8l9s Thx to @timbuckteeth for this; http://t.co/jxkqIRIp

RT @MADinMelbourne: roxon "will enable more families to access credit" @MLolderandwiser: Privacy Act amendments http://t.co/Mv4c7PC2 via @zdnetaustralia

NBN users opt for 100Mbps - ZDNet Australia http://t.co/fLfHMzPn #australia #technews

RT @konradski: Whaddayaknow - turns out Wi-Fi CAN interfere with a plane's navigation systems http://t.co/ospQCU2S

This story has been voted 5 times in the last 24 hours!

7 hours ago, NBN's Tassie upgrade to cost $1.3 million

Sorry no deal Cinders, I'd rather send my money to someone and watch them desperately try to stop the NBN as this has much better enterta...

8 hours ago by Hubert Cumberdale on NBN users opt for 100Mbps

What else can you expect from a Dodo customer?

8 hours ago by Hubert Cumberdale on NBN users opt for 100Mbps

NBN users opt for 100Mbps - Communications - News - ZDNet Australia: NBN users opt for 100Mbps - Communications ... http://t.co/btB9gKWg

NBN users opt for 100Mbps http://t.co/xKqEb4bE via @zdnetaustralia

Biometric bugs too dangerous for public? http://t.co/8JLz5tdF via @zdnetaustralia

Oh please dont be unkind, I gotta have some fan's. btw I agree I dont set the standard, but who does I wonder?

10 hours ago by Doubt on NBN users opt for 100Mbps

You agree but give him thumbs down... I think you'd better take the medication before one of your alter ego's Fred/Frank/Frergers appear...

10 hours ago by Beta on NBN users opt for 100Mbps

Exploring: http://t.co/rT7RPZLA

+1

10 hours ago by Beta on NBN users opt for 100Mbps

War talk dominates #AusCERT 2012 - http://t.co/SlBpMj0c - #security #cyber

So we agree it was a stupid idea and even stupider comment then ;-)

10 hours ago by Beta on NBN users opt for 100Mbps

Not you obviously ;-)

And stop giving yourself thumbs up FFS.

10 hours ago by Beta on NBN users opt for 100Mbps

Ok Beta, understand now, just one point who sets the standard?

10 hours ago by Doubt on NBN users opt for 100Mbps

Oh no Beta you misunderstand me. I like my waterfront home and deep water jetty, it's those "other" people who can move to Willunga.

10 hours ago by Doubt on NBN users opt for 100Mbps

I agree with you Magnus, but really most people like living on the coastal fringe.

10 hours ago by Doubt on NBN users opt for 100Mbps

Travel Tech Q&A: Skyscanner's Ewan Gray http://t.co/vYexrDwu #ipad

Exploring: http://t.co/YNVjdrct

Exploring: Travel Tech Q and A: Skyscanner's Ewan Gray: Ewan Gray, Skyscanner's director for Asia ... http://t.co/bNLCyobv #ICTChallenge

Exploring: Travel Tech Q and A: Skyscanner's Ewan Gray: Ewan Gray, Skyscanner's director for Asia ... http://t.co/HEPuJgyt #ICTChallenge

#NewSouthWales ditches registration stickers 4 light #vehicles in favour of #technology http://t.co/xX5N0Rp9

Another use is city based top surgeons using 8K resolution monitors to provide real-time assistance to country surgeons and doctors to op...

11 hours ago by Magnus on NBN users opt for 100Mbps

Anonymous hacks Reliance's Internet filtering server - ZDNet (blog) http://t.co/uObU1HBP http://t.co/0UBXxwX4

Which Windows will make for a better tablet? http://t.co/4mAHg850

Listening to @stilgherrian cover AusCERT and cyberwar, http://t.co/6lGUEz8H

Travel Tech Q and A: Skyscanner's Ewan Gray http://t.co/VN5tGJzC

#Westpac Board goes paperless with #Ipads with #Tabula #App http://t.co/duxuj2fd #Cybersecurity #Bank

Microsoft is serious about open source??? http://t.co/mqQGgta7

@joedamato just try varying caps randomly. Maybe they do this http://t.co/1FN5FwYv

NSW outlines datacentre migration plans - Hardware - News - ZDNet Australia http://t.co/OQfUl0D1

"on the new fast Internets everyone wants the fast plan" #orly #nareally #yarly http://t.co/kvfCa84A

Chrome overtakes IE: does it matter? http://t.co/e4SILk8a

A ZDNet study showed that British Facebook users are drunk in 76 percent of their photos.

The HDMI cable ripoff and why retail is really dying http://t.co/eFT7zEW7

Travel Tech Q and A: Skyscanner's Ewan Gray http://t.co/IUysbyKf

Travel Tech Q and A: Skyscanner's Ewan Gray http://t.co/V7vL5QB9

ZDNet reports Microsoft launches its own social service http://t.co/VJS5BkwF

by http://t.co/vmlLt4bh: Travel Tech Q and A: Skyscanner's Ewan Gray: Ewan Gray, Skyscanner's director for Asia P... http://t.co/4bfDRXo4

Travel Tech Q and A: Skyscanner's Ewan Gray http://t.co/CtNlVWN7

Travel Tech Q and A: Skyscanner's Ewan Gray: Ewan Gray, Skyscanner's director for Asia Pacific, shares some of h... http://t.co/ZxjpmqiM

This story has been voted 12000 times in the last 24 hours!

2 days ago, Is Bill Gates a great leader?

Facebook Activity

Keep up with ZDNet Australia

ZDNet Events Calendar

ZDNet Events Calendar