@Shadow Minister: What iinet is recently offering would appear to match what is presently available guaranteeing 5Mb/s. Why are we spending $46bn to get the same?

Three reasons:

1. People are maxing out their 5 Mbit/sec links now.

2. Internet bandwidth usage is growing exponentially http://www.useit.com/alertbox/980405.html so it unlikely people would be will be happy with their 5 Mbit/sec connections for long.

3. What appears to be the base level NBN connection being sold now - 25 M bit/sec, is already faster than ADSL can deliver, ever. And the NBN as installed can deliver to 1000 Mbit/sec. To get beyond that 1000 Mbit/sec we don't have to through the hugely expensive effort of replacing the cable again. We can get another factor of 1000 by just replacing the bits at the end of the cable - just like we did for ADSL.

In other words, this is a one-off nation building effort, which is laying down the infrastructure we will be using for the foreseeable future. It is same sort of investment our governments have made in building water supplies, ports, roads, schools, hospitals, electricity grids and the old telephone system.

Or to put it another way, your question sounds like someone 10 years ago asking why the telco's were going to the expense of installing ADSL, given everyone was happily using 56K modems.

This is another question you have asked repeatedly in several guises in this tread. We are going around in circles. I hope this is the final time I have to answer it.

@Shadow Minister: FYI voice requires a peak data rate of 22kb/s and compressed, 1GB of download capacity will give about 200 hours of talk. So why would you be mad to use VOIP over wireless?

I did try to explain Shadow. Evidently I failed. Unfortunately the textbook explanation I would normally fall back to is assumes the listener has a good understanding of network engineering concepts like latency, speed, bandwidth, reliability, error rates and quality of service.

When stripped of all the jargon and mystique these concepts aren't complex, and nor is the idea that different types of traffic (eg data, voice), require designs that emphasise different combinations of those concepts. Normally I would relish the challenge of presenting this in simple terms a layman can understand.

But the only person I seem to be talking to here is you, Shadow, and my efforts to convince of things that now seem self-evident to me have been startlingly unsuccessful, so trying to explain something that is even more complex does not seem like a very fruitful row to hoe.

Rstuart,

The reason you can't convince me about using VOIP over wireless broadband is for 2 reasons. The first is because as an electrical engineer I did communications and am perfectly well versed, secondly I have a colleague who has mobile wireless broad band with a Voip phone, and has no problems communicating. What seems self evident to you is entirely in your imagination. Try telling this to Skype.

Next you claim that there are people maxing out their 5Mb/s connections. Given that you showed that most people needs don't approach 1Mb/s I would like you to give examples. as I have a speed counter on my system and only very seldom exceed 1Mb/s whilst downloading several large chunks of data simultaneously, via bitorrent.

With myself and 2 teenagers, I have yet to reach 100GB, but there is no intermediate package.

The figures coming out of the states are for people using mobile only access. And considering that I have seen this in various articles, please provide any link where it shows that these mobile only households also have land line Internet.

While most of your criticisms are true for mobile now, the services were 10x more expensive 5 years ago and slower and less reliable. In 2020, the comparison between fixed and mobile is projected to be much closer. Try using your arguments against mobile phones and landlines, and you will see that the same applies.

In my work I deal with many high level network engineers all of whom think that the NBN is a colossal waste of money. Only those who have no concept of how networks function think it is a good idea. RS seeing the above would put you firmly in the later category.

@rstuart: But I'm guessing they will find a way to move all those cable customers onto their service.

I was imagining this would happen in 5 or 10 years time. It seems my educated guesses were wrong, but then they often are. From http://www.afr.com/p/business/companies/optus_on_brink_of_nbn_customer_transfer_5A3cfqZ3iuCmFLGdmkg4RO :

"Optus is on the brink of a deal with NBN Co worth between $500 million and $1 billion after tax to shut down its cable network and transfer customers the national broadband network."

@Shadow Minister: What seems self evident to you is entirely in your imagination. Try telling this to Skype.

It's not just my imagination Shadow, it's a collective delusion shared by all network engineering people.

@Shadow Minister: Try telling this to Skype.

Skype is free, has video yet people continue to use existing telephony. Perhaps this link http://www.evdoinfo.com/content/view/3456/64/ will help you understand why. A quote from it:

"The latency or delay can be very frustrating, as you can end up talking over the person on the other end of the line. This issue can be particularly problematic for business users, who obviously need to sound professional over the phone!"

Have you considered the possibility that computer networking engineers know more about building networks than an electrical engineers?

@Shadow Minister: With myself and 2 teenagers, I have yet to reach 100GB, but there is no intermediate package

So lets see if I have this right. Instead of the $700/mo I guesstimated, your current $60/mo package would only cost you $300/mo, and you are happy to pay this?

@Shadow Minister: I deal with many high level network engineers all of whom think that the NBN is a colossal waste of money

How odd. I personally a lot of network engineers, and it is the exact reverse - all think the NBN is a far sighted idea.

But since you do know so many, you might like to get them to explain to you how the voice and data are handled differently on the mobile networks, and why it is done that way.

Rstuart,

I use Skype at least once a week to have video conferences with business colleagues overseas, and have never had trouble with picture or sound. The occasional user using mobile wireless has been restricted on picture quality, but not sound. My kids use Skype with their friends constantly even though we have free telephone calls nationally on our land line.

Saying that I would pay $300 per month is assuming that I use 100GB of download, which I have never reached. Our typical usage is about 30GB which would be $90p.m. in your terms.

As 4G comes on line, and 5G in a few years, the capacity and cost difference between fixed and mobile will drop significantly.

Tomorrow's technology will be mobile Wifi, yesterday's is fibre.

@Shadow Minister: Our typical usage is about 30GB which would be $90p.m. in your terms.

Well a little over $100/month at vodafone rates, both uploads and downloads are charged, plus $20/Gb if you exceed your cap. But even allowing for that we aren't comparing apples with apples as if I understand what you said as your current connection includes a fixed line and the calls made over it.

@Shadow Minister: As 4G comes on line, and 5G in a few years, the capacity and cost difference between fixed and mobile will drop significantly.

So you keep saying.

- Never mind the fact that I pointed out 4G LTE as envisaged is currently vapour ware.

- Never mind that 4G LTE was meant to be available now, but they have given up on that and renamed 3G plus a few additional channel bonding techniques 4G in what I could only call a cynical marketing exercise. What was to be 4G has been renamed to 5G.

- Never mind the fact that even if the grandest 5G plans of the wireless engineers come to fruition, a 802.16m wireless cell will only be capable of around 1G bit/sec in total, which they say can be shared by 200 users in a minimum envisaged cell size of 1Km radius. http://en.wikipedia.org/wiki/3GPP_Long_Term_Evolution#Comparison

- Never mind the fact that 1Km radius will cover more than 2000 homes.

- Never mind the fact that optimistic 1G bit/sec is less than 10th of what existing ADSL at 5 M bit/sec can carry to those 2000 homes now, which is conservative given at 1 km ADSL can carry 20 M bit/sec.

So, as I said above of your vision for wireless: IT. IS. IMPOSSIBLE. Physics forbids it. Of course you didn't believe me. Which is why I gave link to Ziggy Switkowski making essentially the same point. You think he might know a thing or two about mobile networks as he oversaw the building of one. Apparently you don't agree.