Fractelle,

I believe that is the actual proposal - fibre to the home in metropolitan areas and supplemented by wireless in rural and remote areas. The target is only 90% coverage.

Before it became Telstra, Telecom's "commission" was a POTS service to a little over 98% of the population. It could have achieved most of that number in what was called "the Golden Boomerang". (Brisbane to Adelaide via the East Coast).

This announcement was pretty light on detail and was really only a statement of intent. It was made because the original Tender decision was due and they had to explain why nobody had been successful. The full costings are yet to come and some (if not many) things are likely to change.

It's way too early to get overexcited about much of this.

*It's way too early to get overexcited about much of this.*

I agree with Wobbles on that point. Alan Kohler of
Business Spectator, who is usually extremely well informed,
seems to think that with Sol moving on, Telstra will be
split. Much of the optic fibre is already in the ground,
owned by them. So Telstra shareholders will land up with
two stakes, one in the infrastructure company, the other
in the Telco for mobile etc.

So nobody would have to actually spend 43 billion and it
could turn out to be a win-win all round.

tomw: "obsolete before it is completed."

By what? Wireless? That sounds pretty unlikely. At best wireless scales from its current 10MBps to 1Gbps using experimental techniques. But is unreliable, as always. Fibre can scale to over 10 Tbps using experimental techniques now, and it is more reliable than our copper network. 10 Tbps is a factor of 100,000 over the proposed 100 Mbps. It is not something that obsoletes easily.

GrahamY: "I suspect that a version of the 80/20 rule applies to technology"

Well, others suspected we would need no more than 640K of memory. Modern computers have 100,000 times more than that. So what happened - did we do more of what we were doing when we have 640K of memory? No. What happened is new things sprang into life - like computers becoming so easy to use a 70 year old man could engage in a political discussion on a web site.

To put it another way, if we spent $50B on our hospitals what would we get - no queues? On our highways - an optimistic speed increase of 50% Well if we spend the same money on broadband we get a 1000%..10,000% increase now, and a 10,000%..100,000% percent in the foreseeable future. You say you can't see this changing things. As a comparison, if spending $50B on our roads meant the 12hour drive from Brisbane to Sydney took 7.2 minutes, could you tell me what the future would look like? If you start talking in terms of reduced hours for truck drivers you fail.

GrahamY: "it's a dubious proposition that internet speeds 100 times faster than currently available will provide much benefit"

I think characterising it as an "Internet" upgrade is misleading. We are replacing the copper wires current to our homes with optical fibre. Yes, we are replacing the Internet transport - but also the phone transport, and in the long term cable as well. It means everybody with a phone line will instead have a 100Mbps digital connection instead. In that sense, take uptake won't be optional.

(cont'd)

It is that last point that is most important. Right now we have an takeup of 50% of the population having broadband, getting an average of 2Mbps.

http://www.akamai.com/html/technology/dataviz5.html

This means if you are a large company, you can not build your business model on the assumption all of your customers have a high speed link to you. A purely electronic business model is impossible. You have to carry dual infrastructures.

If the NBN goes ahead, in metropolitan areas you will be able to assume 100% of your households have a 100Mbps link to your servers. If there isn't some strict need for a physical outlet, it can disappear without loosing most of your customer base. The shop fronts of banks, accountants, real estate agents, video and music stores will change into something unrecognisable to us now.

Despite, the $50B figure like this makes me pause. That would be true regardless of what it was spent on. It will always be a risk: an enormous bet on the unknown and unknowable. But to answer your question: what else could we it on that make me more comfortable - the answer is nothing. I can't see anything that has more potential upside than this. The rest is merely a case of making the cars go slightly faster, hospital queues shorter or whatever. Boring, predictable, and most importantly small scale stuff. About the best that could be said for it is you have some idea what the pay off will be. Evidently that isn't true for the NBN given most of the comments here.

I have seen the "Back of the Envelope" estimate of the cost to the
average user. It is quite a simple calculation actually.
Just divide the $G43 by around 10 million for the number of premises
amortise over a variable number of years and add on a 8% return on capital + maintenace
and it comes up at around $100 a month +- $10.

As far as the average user is concerned the vast majority will not
notice any improvement over a 1 or 2 Megabit service.
Instead of a page coming down in 2.000 secs it might arrive in 1.999 secs. If you are lucky it might arrive in 1.99 seconds.

There will be many on here that will argue about those figures but
the principle will apply. The speed you get is set by the server at
the other end and how many are connected to it.

TV stations will use it to move programs around the country instead
of paying licences for dedicated co-axs or microwave links.
They are doing this from overseas now when there is not a time critical
factor involved.

Thats where the benefit will be not to Jo Blow and his mate.

Good post, TomW.

Has anyone else noticed that computers are getting smaller and more mobile? Why will people want a home connection, when a wireless connection will give them a voip service and internet access, albeit slower than 100mbps, wherever they go? Speed isn't everything.