Live data from Hacker News

How Australia's $49B internet network came to be ridiculed

bbc.com

161–170 of 170 posts

Re: How Australia's $49B internet network came to be ridiculed

#161
It’s interesting to compare Australia’s grand experiment with Stockholm’s. Australia’s proposal, before retrenching, was 93% coverage at 100 mbps (later, one gigabit), with a government investment around $30 billion AUD. Completion was promised on an eight-year timeline.

Stockholm, in comparison, followed a very different model: https://www.stokab.se/Documents/Stockholms%20Stokab%20-%20A%.... A city owned company was created. No public money was contributed; the network was financed by the utility taking loans. The government imposed no coverage requirement. Service was expanded based on demonstrated demand, and it took 17 years to cover 90% of the city. The utility started turning an overall profit in 2008, after 13 years of operation. The Stockholm system is expensive--customers are charged a set-up fee to recoup construction costs, and although prices must be transparent, they are not rate-regulated. NBN Co., in contrast, is subject to rate regulation and does not recover construction costs as part of a setup fee.

Broadly speaking, there are three lessons to take from this:

1) Urban areas and rural areas should not be part of the same program. "Universal service" sounds nice in theory, but in practice the challenges of deploying in rural areas can torpedo the project. In Sweden, the government did help deploy fiber in rural areas: http://www.ftthcouncil.eu/documents/Opinions/2013/Rural_FTTH.... Sweden required dark fiber to be deployed to within a few kilometers of rural areas, but rural residents had to pay for their own hookups. The government subsidized 50% of the cost of those hookups, up to about $600 USD per property.

2) Coverage requirements are a bad idea. When you force the operator to cover areas where the business case doesn't yet exist, you're forcing them to charge their existing customers more to subsidize new customers. That can make the product too expensive for existing customers.

3) It's okay for fiber to be expensive, at least at first. Charging more to early adopters to recoup large fixed costs is a basic business model in the tech industry. Telia in Sweden charges about $120/month for gigabit, and in many cases customers are liable for construction setup costs. That's okay--it allows the operator to quickly recoup construction costs and use ongoing revenues to expand the network. Sweden does not regulate rates to make fiber affordable for poor people, nor does it force fiber operators to subsidize service themselves.

Re: How Australia's $49B internet network came to be ridiculed

#162
post #98
post #70

Earlier quoted context omitted.

I'm not personally sold on the baby bell model of regional monopoly. I think the whole FIOS debacle, and what happened when google started offering citywide fibre says a lot about how US telco industry operates. Lobby your way to the people who make rules and then get rules passed to shut down anything which freaks your own monopoly of indifference.

I am curious and not sure what to search for. What are you referring to? Link?

What was the "whole FIOS debacle, and what happened when google started offering citywide Fibre"? Why the down votes? This sounds interesting, but I am not sure what to search for to learn more.

Re: How Australia's $49B internet network came to be ridiculed

#163

This article didn't mention what Australians are choosing to do when they have a range of speeds. According to the figures NBN released 80% of customers aren't even paying for the top speed they could. Similarly, this piece from Vox notes that fast internet just is not nearly as valuable as proponents claimed. http://www.theaustralian.com.au/business/technology/only-17p... https://www.vox.com/new-money/2016/10/17/132…

Echoing the sentiment of other replies, this is because you don't actually get what you are paying for. I initially bought a 100Mbps Unlimited plan. I have never gotten close to that. I pay for the most expensive plan out of a hope that I will get prioritized by my ISP, but I have never seen anything faster than 26Mbps, ever.

Re: How Australia's $49B internet network came to be ridiculed

#164
post #87

Earlier quoted context omitted.

Hey Chris, I just moved from Ararat. I never would have expected to see another regional western Victorian on HN! For reference, Stawell has a population of 6,000 while Ararat is about 8,000. The towns are approx 30 minutes drive distance away. The world is smaller than I thought. In Ararat it’s almost better to go with 4G. While you get decent caps (1000G/m for AUD$90) the problem is that the internet drops out ever…

I grew up in Melbourne, but preferred the idea of living in the country, so I made it so. (I’m employed by FastMail and head into the office every couple of weeks or so; I told them of my plan to move into the country before they hired me.) When I was first planning it, I had been considering NBN essential, but shortly before actually moving here it occurred to me that regular 4G was actually quite suitable—better th…

You're right, I was using ADSL2+ and we were connected to a pillar where we were the only ones using it (50 pair). We reported a fault but rather than fix the problem properly they just hooked us up to another pair. The justification was "well you'll have NBN soon".

Gee, I can't wait...

Re: How Australia's $49B internet network came to be ridiculed

#165
post #102

Running FTTH to every home was always a pipe dream (excuse the pun). Some kind of technology mix was a necessary compromise however the balance is wrong - percentage of FTTH is way too low. My home town is literally (and inexplicably!), divided down the middle - one half with FTTH, the other FTTN.

>Running FTTH to every home was always a pipe dream (excuse the pun). Why do you say this?

Because of the astronomical cost involved in running fibre to some premises. And not even talking about city vs bush - two city dwellings in close proximity can have vastly different cost of installing fibre.

I would qualify my original statement with 'as part of a NBN-like project'. I didn't intend to suggest that not every home should have fibre, or in deed will have fibre at some point in the future, just not at the tax payer's expense.

Re: How Australia's $49B internet network came to be ridiculed

#166
post #162
post #98

Earlier quoted context omitted.

I am curious and not sure what to search for. What are you referring to? Link?

What was the "whole FIOS debacle, and what happened when google started offering citywide Fibre"? Why the down votes? This sounds interesting, but I am not sure what to search for to learn more.

I do not live in the USA. What I am told by my colleagues who do, is that the post bell regional breakup created monopolies by location in order to prevent national dominance but give each region a reliable income in a time of voice call logistics. Fast forward to the emergence of fiber as a viable technology: either you live in a location serviced by a telco who is willing to re-engineer to fiber or you don't. Verizon exemplifies the quandary, it's a huge cost, they have to carve out special data models for cable companies to stop them refusing to drop coax, it's impossible to charge sanely, their shares tank. I am sure there are happy customers but there is no clear equality of service. It's a patchwork minefield.

Re: How Australia's $49B internet network came to be ridiculed

#167
post #125

Earlier quoted context omitted.

> The NZ model, by comparison is gold: structural separation. Crown holdings for the fiber. A pot of money, re-circulating to deploy to the cities, each region re-fills the pot to move to the next town. Far simpler. And.. fibre to (most) homes. Only after A LOT of pain. Broadband uptake was pathetic in NZ until Telecom, formed out of the NZ Post Office and sold to overseas buyers in 1990, was split up into retail (Sp…

Wait... are you saying that dialup internet was slow until 1990 when ISDN came out? Or that broadband uptake was pathetic until 2007 with the Spark/Chorus spilt? One of those is really early for any internet, and the other 17 years later, is really late!

The latter. I think there was literally one overseas internet connection in 1990.

For years Telecom managed to get away with doing nothing other than charging exorbitant fees. It's just a coincidence the decision to unbundle came immediately after these comments in 2006:

> What has every telco in the world done in the past? It's used confusion as its chief marketing tool ... customers know that's what the game has been. They know we're not being straight up.

– Teresa Gattung, Telecom CEO

Re: How Australia's $49B internet network came to be ridiculed

#168
post #125

Earlier quoted context omitted.

Wait... are you saying that dialup internet was slow until 1990 when ISDN came out? Or that broadband uptake was pathetic until 2007 with the Spark/Chorus spilt? One of those is really early for any internet, and the other 17 years later, is really late!

The latter. I think there was literally one overseas internet connection in 1990. For years Telecom managed to get away with doing nothing other than charging exorbitant fees. It's just a coincidence the decision to unbundle came immediately after these comments in 2006: > What has every telco in the world done in the past? It's used confusion as its chief marketing tool ... customers know that's what the game has be…

In my option, New Zealand internet didn't really get good until 2011.

That is when Chorus was finally 100% separated from telecom.

From 2006-2011, Chorus were inducing congestion in the ADSL backhaul to put non-telecom ISPs (particularly ISPs with all the high-bandwidth power-users) at a disadvantage. An ISP could buy unlimited bandwidth to the outside world, and give each customer ADSL 2+ connections of upto 24Mbit, but the backhaul between Chorus and the ISP to terminate all those ADSL connections was rate-limited to roughly 64kbps times the number of connections that ISP had.

If all an ISPs customers went online at the same time and started downloading, everyone's throughput would drop to 64kbps.

And Chorus would absolutely refuse to sell any extra backhaul bandwidth.

It was in the best interest for all ISPs to have a healthy base of non-skilled users who barely used their ADSL connections, just so the rest of their users wouldn't get too congested. It was also around this time ISPs implemented various perks to encourage power-users to only download at night time.

This all stopped in 2011, and the bandwidth caps that all ISPs were offering suddenly shot up.

Re: How Australia's $49B internet network came to be ridiculed

#169
post #32
post #20

Earlier quoted context omitted.

Wait, if the Conservative party decided to destroy it how is that the fault of the Liberal Party?

The Liberal party is the conservative party. No Conservative party in Australia AFAIK. The Liberal party is broadly social-conservative, pro-business. The Labor party is broadly speaking socially liberal and pro-labour.

> The Liberal party is the conservative party. No Conservative party in Australia AFAIK.

You a little behind the times. That was true many years ago, in the Fraser years. Now it's "The The Liberal party is the conservative party. There is currently no 'John Stuart Mill' liberal party in Australia".

Re: How Australia's $49B internet network came to be ridiculed

#170
post #30

First common mistake: Yes, Australia is geographically huge. No, it is not sparce. it is a highly urbanized economy in terms of geographic concentration of the population. Therefore, the oft-quoted extreme expense "oh, its so much more expensive to do it in OZ compared to ..." is bullshit: its not. There are of course some very expensive locations to service, but thats a USO problem. Overall, the cost of deployment o…

Actually, the first mistake is saying the NBN was Rudd's idea.

The problem was created by the Libs in 1997, when they privatised a government monopoly, Telstra. Telstra promptly morphed into natural shape of private monopolies everywhere: a bloody great rent seeker, beholden to no one. They sat on the existing copper network assets, milking them for all their were worth and showing not the slightest interest investing to upgrade the asset when new technologies arrived. (The contrast with their actions in the mobile sector, where they didn't have a monopoly could not be more stark. We have a world class mobile network, and were the first to phase out 2G.)

10 years later, the prime minster who created this mess was still in power. It was only 10 years later, but by then it was obvious even to him something had to be done because we had this headline: “Australian Prime Minister John Howard on Monday announced a 2.0 billion dollar (1.68 billion US) plan to provide fast and affordable Internet access across the vast country”. http://www.smh.com.au/news/Technology/n/2007/06/18/118201899... Notice this was several years before Rudd "had" the same idea.

Turned out 1.68 billion was a bit of an underestimate, so it went nowhere. We've are now spending 20 times that amount to extract ourselves from the mess he created. I do hope the $14B they earned from the sale of Telstra (which is several times less than the cost of the fix) was worth it at the time.

Post reply on HN