IPv6 content experiment

[Kevin Day]

Just a quick announcement that details of a new IPv6 experiment have  
been posted at http://www.ipv6experiment.com

As everyone's aware, there's the issue of not enough eyeballs to  
justify content providers spending the time to deploy IPv6. End user  
ISPs won't do it until there's content. Many say that there are ways  
(tunnel brokers, teredo, etc) end users can get on IPv6, and the fact  
that they aren't using them means it's up to the content providers to  
step forward. Either way, "lack of demand" is cited by many for the  
biggest reason why they aren't deploying IPv6. So, what if we put  
some desirable content up and made it available only on IPv6 and gave  
those who accessed via IPv4 detailed instructions on how to get on IPv6?

How many are actually able to get on IPv6 if they want?
What problems do they run into when trying?
Is their connectivity over IPv6 worse than IPv4? (number of hops,  
packet loss, overall transfer speeds, etc)
How many users have IPv6 configured, but don't actually have a  
working IPv6 connection? (i.e. how many people do you lock out by  
publishing AAAA records for your site?)

(I'm being intentionally vague here about the details of the content,  
as not to trip any email content filters. See the site for more  
details.)


Before we begin with this experiment, I'd like to ask the community  
if there are any other interesting metrics you'd like us to try to  
capture, what methodologies you suggest we use, etc.

If you have any questions or comments, feel free to contact me directly!

-- Kevin