Slashdot Mirror

I agree it's more the monopolies that bring things down. Government run departments and organisations are as prone as big monolithic corporations to seizing up and resisting change. Everything's great when they're new and fresh, the hard part is keeping the mindset and attitudes that way in the long term.

I'm in Australia, and I remember very well how bad broadband was under Telecom and Telstra. Even our dial-up internet was pretty poor, which is all we really had back then - the rest of the world was already rolling out cable and DSL. Telstra, as a largely government-owned "privatised" telco had no incentive to innovate, increase service levels or offer new products, were first forced to wholesale their retail access products to competitors and as the competition matured, forced to offer access to their physical plant (eg, copper pairs, exchange space, tower space) on a reasonable costing model. We had a huge, vibrant ecosystem of access providers offering fibre, wireless, DSL, etc up until ~2008, when everyone stopped investing and started waiting for the NBN to appear.

Now we have this NBN, which was designed completely contrary to the way it was marketed from the beginning (protecting the largest incumbent telcos' and NBN lead contractors' primary revenue streams) and has been built as a monolithic system with no competitive access to physical plant. Services that the NBN can't deliver aren't possible, successive governments have made the design worse (eg, the "MTM") and legislation prevents access providers from building competing retail networks over holes in coverage. To make it worse, the build-out contractors seem to be paid based on premises on-network rather than successfully connected, so up until late 2017 they've just been smashing out abysmal builds as quickly as possible and going back months later to fix the problems - the repair teams are a fraction of the size of the build teams, and repairing already requires more effort than doing it properly the first time.

This long rant (believe me, I've got a lot more and could go for a while) is just an example of a government network infrastructure project gone poorly. Contractors and design partners were chosen from existing large network operators and had a vested interest in not killing their biggest cash cows. Technology was chosen that would least impact their services and allow them to manoeuvre for competitive advantage down the track. Later governments took the remaining good parts of the design and returned it to the old Telstra model of using whatever can be scraped up, but this time we don't have competitors able to deploy their own equipment to underserviced areas or to supply missing features.

These successful municipal American ISPs have the right mindset (they've been created to fix problems and service customers) and are hopefully small enough to keep it. It doesn't really matter that they're associated with government.

It was never about saving money but making a point of differentiation from Labour at the 2013 election, which, like the last election, was a very close race.

Also, it was sabotage, but presumably as a matter of collusion between the Liberals and the higher-ups at Telstra and Foxtel, as well as Rupert Murdoch himself. Fun fact: Telstra and News Corp. (i.e. Murdoch) each own 50% of Foxtel, who hold a virtual monopoly on satellite and cable TV in Australia. Interestingly, throughout the 2013 election period News Corp. was highly critical of Labour while tending to champion the Liberal Party's policies--most likely a significant factor in them winning the election. https://www.crikey.com.au/2013...

In 2014 it was announced that rather than decommissioning Telstra's technically unsuitable copper network and HFC network (that Foxtel relies upon), the assets would be transferred to NBN Co for indefinite maintenance. https://www.gizmodo.com.au/201...

It comes to no great surprise that the former opposition communications minister during the 2013 election, now-Prime Minister--Mr. Turnbull himself--has been in communication with Telstra, through all of this and appears to have intentions to privatise NBN Co and possibly sell it off to Telstra for no apparent economic benefit. https://www.crikey.com.au/2016... http://www.nbnco.com.au/corpor...

Former ABC technology journalist Nick Ross, one of the few journalists who bothered to cover the NBN situation in any great depth, came out last year claiming he was "gagged" by his superiors for reporting critically of the obviously flawed Liberal NBN. https://delimiter.com.au/2016/...

Stephen Ellis, former NBN advisor under Turnbull's NBN, last year transitioned to a senior advisor role at Telstra with a spokesperson stating "We have engaged Mr Ellis as a consultant on a specific project to advise Telstra on longer term policy reform options. We will not be commenting further". http://www.theage.com.au/victo...

The Liberal NBN policy has been a knowing scam since day one. "Business as usual", indeed.

If they really want to live in a place with no infrastructure then exile to a barren island would be a suggestion..

We've tried that before... That island now has its own Fiber Network... :)

Hi ebno http://www.nbnco.com.au/nbn-for-home/how-it-works/how-it-works.html might help a bit
If you have a tech question have a text/google search of the NBN section at http://forums.whirlpool.net.au/forum/142

A good argument, although there are plenty of very densely populated areas in the US (on par with the more densely populated countries you mention) that still have poor speeds. I think the issue in the US is the inconsistency of the speed available. One guy might have 100 Mbit FiOS available, whereas his friend a few blocks away can only get 3 Mbit DSL. Competition is also a problem - I know for me in my mid-sized midwestern city, I have precisely ONE broadband option (one cable provider who fortunately provides a healthy 30 Mbit, though my previous place only had AT&T DSL at a maximum of 6 Mbit available which sucked ass).

The US could see a big jump in its average speed merely by switching out the ADSL1 exchange equipment in DSL-only areas and replacing it with ADSL2+ and/or VDSL (ala U-verse, but available universally, not just in select areas). Those people stuck on 3 or 6 Mbit DSL could then get 12, 18, 24 Mbit speeds using their existing phone line.

So yeah the US appears to be the fastest large (geographically) country on the list (though, one of the most expensive still) and I agree that it's actually doing quite well in terms of its ranking, all things considered. I expect it won't stay that way for too many more years though (Australia in particular should leapfrog the US in the 2015-16 timeframe once the NBN hits critical mass - this will provide 100 Mbit to 93% of the population and is already available in some areas, but not enough to make much of a difference yet).

Yep, mainly by introducing the NBN which the MSM see as a direct and imminent threat against their core business.

> 1. Analogy to water or electricity is flawed. Water and electricity are limited resources--but data is neverending. The only limit here is bandwidth

Bandwidth requires two components: pipe size and routers to move the data. Moving more data is expensive.

> 2. Past a certain point, more speed is irrelevant, but when you hit the data cap, your speed drops to zero (or your bill goes up in a way that is biased unfairly toward the ISP).

In Australia speed is capped at between 128Kbps - 512Kbps - still reasonable for browsing the web & checking email.

> 3. "With fibre that increases to 1Gbps and 324TB/month." You're arguing from the presupposition that users will max out their connection 24/7. That is silly.

> 4. A customer who uses 10% of his data cap pays x, but a customer who uses 101% of his data cap pays 1.2x. The 90% of unused bandwidth more than makes up for the extra 1% used by the other customer

The ISP still has to buy sufficient backhaul to support your 101% utilisation. In countries where caps exist (e.g Australia) the ISPs model on people using significantly less than their broadband cap. NBNCo provided a couple of example usages:
* 12/1Mbps 50GB plan with an assumed average download of 18GB
* 50/20Mbps 500GB plan with an assumed average download of 75GB. Remember these are averages so for someone downloading 500GB there have to be ~6 people downloading 5GB to arrive at an average of 75GB, except that it doesn't work like that because the RSP most likely has a 100GB quota plan and most of the people only downloading 5GB a month would choose that. Change the 5GB to 50GB and then it becomes 17.

Essentially what you are asking is that I suffer slow speeds to subsidise your heavy downloading with speed caps. As a technically literate person you would understand that the cost to lay fibre is exactly the same regardless of the speed and that speed caps are software setting. Supporting more data requires investment in larger backhaul and faster routers.

It is trivial to setup a computer to saturate a 1Gbps link. Sadly it requires only a couple of script kiddies to do it and performance of the ISP goes down the tube or their costs sky rocket.

> 5. Your contrived scenario is internally flawed, as well: what if the file the "someone" has to wait a month(!) to download is media from a family member overseas? a recorded video, photos...? Who made you judge of what is more justly important to random people?

By imposing speed caps you are forever denying the opportunity to experience a truly fast internet, where as with download quotas people can choose what content is more important, therefore I consider my point about it being socially unjust valid.

> 6 Yeah, my ISP (AT&T)...

So your ISP sucks. In Australia we can choose between multiple ISPs much to the disgust of Telstra. As you've probably guessed by now, I'm also making these comments in the context of Australia where the government is the the process of re-nationalising our telecommunications infrastructure.

> 7. "For me personally, I would much prefer 1Gbps with 100GB quota, than 12Mbps with no quota." That's a terrible example. What can you NOT do with a 12 Mbps download rate?

High definition video conferencing is one example. The througput recommended by NBNCo is 100/100Mbps. If you only use your connection sparingly for such services then 100GB is fine. I'm regularly under 100Mbps.

You might also be interested to know that the anti-fibre brigade in Australia use exactly your argument to stop the roll out of fibre.

> In conclusion, you're deluded or lying, a fool or a shill.

Or you have a fundamental lack of understanding, especially of the situation outside of the USA. I'll leave to others to decide.

• * NBN is costing ~$40 billion compared with Google Fibre estimate of $150 billion for a Nationwide roll-out.
  Now Australia has a similar land mass to the US, but less than 1/10th the population and is heavily urbanised.
• * Google is 1Gbps for $70. Currently the highest NBN plan is 100/40Mbps with 1TB quota for $164.95/month.
  1Gbps will be available in a couple of years on the NBN, but price is $150 plus data (100/40Mbps wholesale is $38/month so retail prices will be likely to be double wholesale).
• * 50% on the NBN are predicted to connect at 12/1Mbps (page 64 of NBNCo Corporate Plan (2012-15)) at ~$40-$50 for 30GB month.
• * Current NBNCo wholesale prices are discounted, as per page 67 of the NBNCo Corporate Plan
  - 1000/400Mbps falls from $150 to $90, while the average speed grows from 30Mbps to 230Mbps. So price falls by 40% while average speed grows by 7.6 times
  - price of CVC falls by 2.5 times, while the data usage grows by 20 times

Sadly, too many people in Australia have read the "1Gbps Fibre announcement made during the 2010 Federal election campaign in response to Google Fibre Announcement. Very little investigation has been done to appreciate what is being delivered for fear that like other government initiatives, the delivery will be a failure. The reality is that for 50% 12/1Mbps will be what they experience and for maybe 25% 100/40Mbps with download quotas. The rich will have their 1Gbps connections subsidised by the national roll-out. Rather than revolutionary the NBN will be little better than current 100Mbps cable services. NBNCo are also focusing on areas already serviced by cable to re-enforce their natural monopoly and reduce competition.

The alternative would have been to release uncapped fibre speeds (everybody receives 1Gbps) and charge for data. This is fair as low quota users put hardly any load on the network, while heavy downloaders do. We would then have had something closer to Google Fibre and truly revolutionary. To quote Simon Hackett (Internode):

"They could charge that average to everyone, and open the ports up to full speed for everyone (this is my personal preference, FWIW)." post by Simon Hackett

Instead we have a fibre network where anyone downloading less than 5GB month (national average is ~20GB (ABS) is probably better using a 4G connection and ditching the land line. If we accept that NBNCo are correct in predicting that 50% will connect at 12/1Mbps, then based on the fact that 25/5Mbps plans are only $5 extra a month there is a significant number of people who may find wireless better value.

The $140B figure seems low. Numbers like $40B are being bandied about here for a similar project already underway here in Australia (the NBN) an we only have about 1/15th the population of the USA.

"high speed internet everywhere in the USA and then eventually abroad"

We are already doing this in Australia http://www.nbnco.com.au/

And we have tons of people insisting that it's totally not necessary at all!

(which I find staggering in a country dominated by distance and sparse population, in a world where the largest growing sector is coming from internet business startups).

Woz, I used to have a great deal of respect for you. It's obvious that you've had the wool pulled over your eyes on more than one front.

Mr Wozniak told The Australian Financial Review in Sydney that he had spoken to Communications Minister Stephen Conroy and was in support of the federal government’s fibre rollout. “I spoke to him and they plan to roll it out to everyone in the country,” Mr Wozniak said.

Bullshit. 95% of the Australian population has cellular coverage at the moment because that proportion of the population is congregated in the coastal areas where it's easy for the telcos to provide cellular towers. Only 90% of that proportion of the population is going to be anywhere near the National Broadband Network. I live in Mount Gravatt (QLD, 4122), within a couple of kilometres of the nearest telephone exchange and have to put up with shitty ADSL courtesy of the encumbent ex-government telco, Telstra. The NBN is not scheduled to be rolled in our suburb any time within the next three years. Maybe 30-40% of the greater Brisbane area will have NBN coverage in the next 3 years. REF: http://www.nbnco.com.au/rollout/rollout-map.html

And Stephen Conroy wants to put internet censorship in place on the Australian networks. The list of censored sites are protected secrets under law and there is no process to appeal if your business happens to wind up censored. Doctors, dentists and family welfare groups got blackballed during the trials in Tasmania, so you don't have to be pushing porn or terrorist materials to get sunk. Is that the kind of country you want to live in?

“I’m not an expert on banking but bankers have told us how important this technology is to them and it is one of our big customer areas,” he said. “Some success in banking is all done in computers nowadays, not through humans, and milliseconds matter, the speed of transactions matter to them.”

More bullshit. Banks still take 3-7 days to clear a cheque (check, if you're a yankie), same as 100 years ago. Nearly-instant transfers are not applied for the benefit of customers.

Also, in case he does actually read articles about himself - I also hope he's checked out where it's been rolled out to already and where it is scheduled to be rolled out to before purchasing/renting a house... because otherwise he will have to deal with Telstra instead.

It's being rolled out as we speak (the purple bubles are working instalations) and it would be very hard for Abott (and lets pray he dosen't get in) to cancel it now that all the deals have been done with Telstra, etc and work has commenced.

How does 5 Mbps of "TC-1 speed" differ from 5 Mbps over an ADSL 2+ connection with QoS?

I googled some info about TC-1 below.

[ From http://www.nbnco.com.au/assets/documents/ssrs-product-technical-specification.pdf (page 24) ]

Traffic Class: TC-1

Frame Delay (one way): < 350 ms

Frame Delay Variation: < 25 ms

Frame Loss: < 0.01 %

Availability / Connectivity: > 99.5 %

The TC-1 CIR performance attributes are dependent upon the following traffic characteristics, as enforced by Customer:
  TC-1 CVC capacity operating at 70% utilisation
  A TC-1 AVC to CVC oversubscription of greater than 10:1
  A balanced distribution of CVC demand across the associated AVCs
  Periodic frame arrivals, every 20ms
  Frame length maximum of 150 bytes at NNI

Because I can see you live under a rock: http://nbnco.com.au/

Most people will have 1Gbps capable fibre in Australia. The rest get fixed wireless (LTE), & very remote areas get satellite, all for the same wholesale price.

It's actually cheaper to run FTTH these days with the polymer cables (what the NBN is using) than to run FTTN networks that rely on crumbling copper.

The actual goal is to ramp up within 2 years to 5,900 premises a day based on 250 working days a year. This equates to 1.5 million premises a year. This is 0.2 million less a year than the HFC rollout in Australia in the 1990s. You can find this information on page 78 of the NBNCo Corporate Plan.

The NBNCo press release contains some key wording missing from the news article: Construction to be underway or complete in areas containing over 3.5 million homes and businesses in 1500 communities in every state and territory. Underway means that NBNCo have started detailed planning for the area. For some suburbs named in the plan (i.e. Weston, ACT) detailed planning doesn't commence until September 2015 and homes in the suburb cannot expect to actually have an active connection until September 2016. So the actual figure will be significantly less than 3.5 million able to connect in June 2015.

Would love some form of fibre connection if it were available in my area. I'm in Australia so I should be able to get fibre within the next couple of years as the National Broadband Network rolls out (the area a few km north of me is getting it this year, lucky sods). But for now DSL + 3G is fine.

I'm mostly looking forward to the better upload speeds on fibre rather than download - that'll make a huge difference for checking large stuff into the SVN repo at work.

To my knowledge NBNCo is rolling out their own new fibre system to places which are not currently cabled for fibre. Including the original Tasmanian trial etc.

Telstra has their own fibre network which must be near 10 years old, using their own concentrator (correct term?) and their own modems. This older system is still of course fibre based (well to the street, I don't believe it is to the door)

If I recall, the Govt / NBNCo are trying to take a shortcut on wiring places up by using Telstras existing Bigpond infrastructure in locations they've already cabled up with fibre. I can only wildly speculate here but I'm pretty sure it's fibre to the street, not to the house. There would be significantly less fibre laid then 'proper' NBN installs.

So, seeing as the fibre doesn't come all the way to the house, they can't use the standard NBNCo equipment.
http://www.nbnco.com.au/assets/images/hi-res/truck-02-hi-res.JPG (4.5mb)

So basically I'm guessing anyone put on the "Telstra version" of the NBN is basically just getting Telstra Bigpond cable, at new NBN prices and I (guess?) with all the speed caps removed for the modems.
I'm also going to guess, for the sake of saving even more money, rather than issue a mini UPS / replacement modem with VOIP built in (basically what NBN customers get) Telstra can't be bothered replacing this equipment and that explains the rule to stick with copper (likely free / subsidised) once the deal is agreed to.
I mean logically there's no reason "Telstra Fibre" (cable) customers couldn't just use a new, standardised piece of equipment which offers full VOIP, Battery backup and a highish quality modem which delivers decent (although sub fibre) speeds

This information is absoloute speculation but something along the lines of what I think is going on, if this is not the case and someone has a better clue (very likely) let me know where I went wrong. Regardless, I suspect money is the culprit behind this one and saving a hell of a lot of digging up places.

Unfortunately NBNCo have successfully propagated a scarcity myth in both speed and data, so 50% will be connecting at 12/1Mbps (page 118 of NBNCo Corporate Plan). The government have been promoting 1000/400Mbps (which the installed hardware is capable of), but those plans aren't expected to appear until 2026, and eHealth, but HD Video conferencing won't work on a 1Mbps upload.

Telstra have just launched an LTE network and the other mobile carriers will also launch LTE networks shortly for the low-end user these networks will be $10-$20/month cheaper than the NBN. So much so that NBNCo have been trying to restrain the mobile operators from competing. If the wireless operators can take more than the 13% that NBNCo are predicting (page 116) then NBNCo will be in serious trouble and prices will rise.

In summary, Australia is building a world class fibre to the premises network, but it is going to be hamstrung by speed tiers which deliver little extra money to NBNCo (~$4/month) and reduce NBNCo's income from data.

Sacrificing modpoints to say this:

...providing broadband services with initial speeds of up to 100 megabits per second (mbps), rising to 1 gigabit per second (gbps) in 2012, and with the capacity for further upgrades in the future...

Taken from NBNCo's Overview PDF.

I like TPG's Broadband map. Very useful - http://www.tpg.com.au/maps/

For nation wide see http://www.adsl2exchanges.com.au/

The new National Broadband network also has coverage maps - http://www.nbnco.com.au/wps/wcm/connect/main/site-base/main-areas/our-services/coverage-maps/ ....

and I'm fairly certain that all of the above, combined, has cost less than $100 Million :-)

Yes, the interior is "sparsely populated" - Not much water in there, and the climate can be harsh. Amount of water and comfort of living roughly equates to amount of people.

After the NBN roll out continues across Australia, will it will be technically easier for the government to implement a filter (because NBN Co. controls the network hardware)?

After the NBN roll out continues across Australia, will it will be technically easier for the government to implement a filter (because NBN Co. controls the network hardware)?

You could just google NBN in Australia.

Given you cant be bothered try a look at the NBN
home page (gee that was unforseen eh?)

http://www.nbnco.com.au/

Yep it appears to be modeled at least partly on the Australian National Broadband Network, although it will no doubt be modified somewhat to suit the NZ telecommunications market, geography and requirements.

Incidentally, actual consumer plans on the new (Australian) network (which has several trial areas already wired up) have just been announced in the last week or two. And they are better value than comparable DSL plans (in terms of download quota), despite the far greater speeds. This will come as a pleasant surprise to those that feared that new, faster tech would also mean more expense.

Taking a look at one ISPs NBN offerings, initial launch speeds are 25/2, 50/4 and 100/8 Mbit (downstream/upstream), with a choice of quotas from 15 GB (entry level) to 200 GB. And these prices will almost certainly come down further once the NBN is available in more than just a handful of trial areas and more ISPs come on board. I actually suspect we'll eventually see true unlimited plans becoming common (some ISPs such as TPG and AAPT are offer this now, albeit expensively!)

I suspect though that NZers will get their network completed before Australia does due to their smaller land area though (and potentially less political infighting!). Good to see it happening on both sides of the Tasman. Copper POTS networks are on their way out. They have served well for ~100 years, but everyone knows replacing them with fibre is inevitable. Might as well start the job now.