ISP's In Uproar Over Verizon-MCI Merger

Joe Abley jabley at isc.org
Thu Aug 25 03:41:24 UTC 2005


On 24-Aug-2005, at 22:43, Dan Golding wrote:

> I suggest you take another look at these numbers. Those countries with
> overall population densities lower than the US's all have something in
> common - they are really cold. Iceland, Canada, Finland, Norway,  
> Sweden.
> Folks in those countries are densely packed into relatively small  
> regions of
> their overall land area (near oceans or in cities). Sure, some  
> folks live
> out in Nunavut, but a relatively small number. Contrast that with  
> the US
> where the population is far more spread out.
>
> This is an issue of both distribution and density, not just density.

It's an issue of far more than either of those, and that is my point.

By way of example, New Zealand has a population density of 15 people  
per square km, vs. 30 for the US. It's a much smaller country by  
size. The population is much smaller. The population is also  
sufficiently centralised around few major urban centres that you can  
reach almost a third of the whole country's population if you decide  
to serve just Auckland.

The US is number 11 on the OECD list, with 13 broadband subscribers  
per 100 people. New Zealand is number 22 with 4.7.

If it was just down to population distribution and population  
density, New Zealand would be awash with broadband Internet. It  
isn't. (New Zealand provides a useful model of what happens when you  
privatise the incumbent telco and provide almost no regulatory  
assistance to competitors at all.)

While it certainly doesn't hurt to have highly concentrated, urban  
population centres when you want to build an access network, there  
are many more differences between the different economies on the OECD  
list than just geography.


Joe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
URL: <http://mailman.nanog.org/pipermail/nanog/attachments/20050824/33240d28/attachment.sig>


More information about the NANOG mailing list