[TriLUG] Adding to the list of topics: IPv6

[M. Mueller]

On Fri, Jan 23, 2004 at 07:57:28AM -0500, Reginald Reed wrote:
> I don't think it is a false generalization William.  The same has been said
> about new technology/ideas throughout history.  The point people are trying
> to make is the fact that even if its not an IP enabled toaster, there could
> be many other new devices that haven't even been thought of yet that would
> fully utilize full open connectivity to the Internet as a whole.  The idea
> is that IPv6 a good step towards making sure new innovation isn't stifled by
> the fact that there aren't enough IP addresses available to these new future
> devices.

Nice thread.  

The thing that eludes me in this discussion is whether or not the
proponents of IPv6 have implemented it in their own operations.  IPv6
exists today and it's available to anyone.  So why not deploy it?
Because there's not much you can do with it outside your own network
unless you know of another similarly configured network in which case
you'll have to bring up your own WAN facilities so you can connect your
IPv6 routers.  An alternative is to connect to an IPv6 service
provider.

Now if you're the one paying the bills you'll quickly conclude that IPv4
is much less expensive and the ROI on IPv6 is not justifiable to
reasonable people (notice I did not say Board of Directors because they
are simply not to be trusted any longer).

In China, where there is no infrastructure, of course they will use
IPv6. That may drive the change to IPv6 in the rest of the world.  In
order to connect to China (a vast business opportunity, or more likely,
where everything is actually made) you'll have to be connected to an
IPv6 network.  US based ISP will gradually turn up IPv6 networks as
demand grows.  At first it will be a premium service as it is now.
Later it will become pervasive.

It is clear that IPv6 will offer more recognizable points of presence on
the network and that is a good thing and a bad thing. It is clearly a
limited thing. The general problem represented by the IPv4 vs IPv6 
transition will occur repeatedly as we plunge into the future.  
Generation after generation will solve the same problem over and over.  
It seems to me that what we need is a network of networks that can adapt 
to new technology as it evolves without obsoleting current networks. For 
example, if I wish to browse the IPv6 world I will either need to attach 
to an IPv6 network directly, or subscribe to a service that acts on my
behalf in the IPv6 network and maps IPv6 addresses to an IPv4 temporary
address, or to an application that implements its own address space.
Now I have a choice - directly connecting to the new network or use a
proxy interface.  Not good for real-time data streams you say? The
walkie-talkie model fills the need  until laws of economics 
allows for change.

Mike