Yet another plea for cost benefit analysis on a project planned to give benefits in 10 to 20 or more years ?

Which economic genius is going to tell us what lies ahead in 10 years in the normal course of events, let alone in a future still shaky with GFC after-effects and climate change implications?

Which technological or economic genius can indicate with any certainty where internet-type applications are going to be in 5 years, let alone 10?

The transport analogy both ignores and highlights one huge hole in most of the thinking so far about the prospects of the NBN :
we won't need expensive roads and transport systems nearly so much if we have an NBN as planned PROVIDED THAT our workplaces acquire managerial nouse and wake up to the fact that just because you have people in expensive CBD offices from 9am to 5pm doesn't mean you're actually managing their productivity.

IBM's boss in Australia has pointed out we need to think what we're going to be doing with the cable... He assumes we need it. It so happens his company already does a lot of what most arthritically-minded Australian "enterprise" managers won't wake up to: IBM actually lets people work from home!

Less traffic, less pollution, less energy required, lower office rent costs, better family life and leisure time, and usually, much higher productivity from the home office! Also less capital expenditure on expansion of existing transport systems, a greater spread or averaging of housing costs, assistance in regional development, etc etc.

Now let's see you build that into a cost-benefit analysis. What would be the take-up rate, in what circumstances ? How many variations on that theme would you allow for over 2, 5, 10, 15, 20 years ?

Will that cost benefit analysis be finished in 2, 5, 10 years etc ?

(Meanwhile make sure you add on the cost side, the cost of retraining and changing the bloody-minded incompetence of most Australian managers who can't see past their way to the corner office with a harbour view.)

Like PeterGM, I think we'll need the internet more than we'll need roads in the future. How to determine cost-benefit is another matter. I've just been to a peak oil conference in Washington and the overriding message is that we face a liquid fuel crisis within five years. It is imperative therefore that we maintain the grid and the internet in order to avert economic collapse. The secondary message from this conference, however, was get out of debt, quickly. So how do we get the NBN show on the road without going into debt? Geoff is right: we don't need high speed internet everywhere if it is not going to be cost-effective. I'm personally on wireless broadband as I live in the country and it works well.

Just because your wireless works well is a very narrow minded view.
People are going to get out of the citys, One sureway of this happening is with a broadband structure in place.
Think of the longterm future. Someone said technology might change in a few years , it probably will, that is no reason to do nothing.
Satelite tech, has been there for years, but it will not be used because charges can not be seperated. There is no long or short distance.
Infastructure should never be compromized.

For a detailed debunking of the please for cost benefit analysis of the NBN, see http://blogs.crikey.com.au/pollytics/2010/09/17/cost-benefit-delusions-of-the-nbn/comment-page-2/

Meanwhile, some perspectives:
1. Copper network is going to need replacement relatively soon anyway.
2. Wireless frequencies are limited, and like popnperish, I live in the country, but wireless internet is pretty hopeless. In any event, if you want to solve city problems, you have to give city communications to regional areas.
3. We currently spend about $14 billion per annum on road transport in this country. And it's rising.
4. $43 billion is disputable as a total cost to government of the NBN. Only about $24 billion comes from taxpayer funding. (ie, 2 years of roads expenditure, or 1 year of roads expenditure, plus the cost of replacing copper network (see below).
5. The equivalent in today's dollars of the cost of copper phone network is about $10 billion.
6. Hey, our government borrowings are minuscule and conventional compared with: a.) past history of deficits and b.) other countries.

And BTW, governments just aren't private business enterprises, nor do their finances or expenses or responsibilities or accounts even vaguely resemble those of private business or private households.
(Anyone printing their own money is in deep trouble, for a start...)

Sure it is nice to work from home. I did so since the early 1990's, using the newly-emerging internet and everything went most smoothly through E-mails, FTP and text-chats over a dial-up line.

The point is, you don't need a super-fast network to work from home, you don't need to see every freckle on your boss's face!

>The point is, you don't need a super-fast network to work from home, you don't need to see every freckle on your boss's face!

You do need it because speed means bandwidth and if everyone else is doing it rather than just a few users of the 1990s. For practical purposes, I could have done with lots more bandwidth when I was forced to use a satellite connection, and basic satellite bandwidth and speeds were huge.

Internet users increased by 150% between 2000 and 2010 in Australia. That trend isn't likely to slow, but even if the increase in the number of users eventually peaks out, the demand for different data and huge chunks of bandwidth isn't likely to slow at all.

I seem to recall. from a decade or two ago, PCs with 48k of RAM, expensive hard disks of 5 Mb and comms bandwiths to match. Today that's laughable. Expect more of the same. Don't allow for it and you welcome decrepitude. The faster the computers get, the more information they'll be asked to process, and the more information you'll be expected to upload and download to and from work. And the more businesses will be wanting constant cloud computing and deeper and deeper data analysis.

Meanwhile, you do need it if managers still hold onto a dream of seeing you at work, freckles and all (which many of them do) or if, having found out about instant communications, they start using them to the Nth degree, and want to share whiteboards all the time (entirely feasible now). And if all medical records move to clouds, including xrays and ultrasounds.... etc etc.

And if instead of a few billion bus trips per year, we get a few billion work connections, you'll need wide bandwidth Real Soon Now.

I just checked my download rates from my adsl connection. My "1500k" connection is currently getting me a peak of about 138kb/second, but mostly running in double figures. Tell me I don't need more.