Re: Open question and Critical dependencies

[Thierry Ernst <ernst@sfc.wide.ad.jp>]

> > Resistance to adoption of IPv6 in the planet where I live doesn't
> > exists. I use IPv6 on a regular basis, even to send this mail - you
> > can check this in my mail header:
> > Received: from shonan.sfc.wide.ad.jp (shonan.sfc.wide.ad.jp
> > [IPv6:2001:200:0:8803::53])
> 
> Unfortunately, the IETF servers (or the PSG one, in this case) don't 
> support IPv6 so this is not the case. 
Hum, is this reasonable that an IPv6 WG doesn't even have IPv6 resources
? What about having an additional page which could only be seen through
IPv6 ? ;-)

> However, there should be an IPv6
>  address in the headers of this email.  :-)
Yep, I checked !

Note I didn't say it would be sent with IPv6 all the way long from me to
all recipients, but it was sent using IPv6 between my laptop (I of
course use an IPv6 compliant mailer) and my mail server, which is
located 50kms from where I work.

> > FYI, in my office, we couldn't even get enough IPv4 addresses for
> > all desktop,
> 
> Strange.

Do you know how many IPv4 addresses have been allocated to each
countries in the world ? Some have several per citizen, some have 1 per
citizen, and some less than one. I met people from a Tunisian university
and they told me they only have 16 global IPv4 addresses. I'm quite
lucky to live in Japan, but this is technology-speaking a very highly
developped country, so there are no tradeoff between v4 and v6.

(we could have received more IPv4 addresses if we where willing to
renumber - which we didn't want).

 
> >> When something take 10 years to get to the point of starting
> >> infrastructure  deployment, there is usually a rather basic
> >problem. > In particular, the world  has repeatedly seen promises of
> >the next > great technology that will see grand  deployments "soon in
> >the future" > that somehow keep being in the future, rather  than
> >now.
> 
> > When this 10 years period started ? If you refer to the time it was
> > first introduced at the IETF, it doesn't hold, as it takes time to
> > publish RFC, test implementation and market produst.
> 
> Right. Both claims of success and claims of failure are premature. And
>  in my opinion, dual stack servers weren't really deployable until the
>  
> addition of the IPV6_V6ONLY socket option, and IPv6 only clients still
>  aren't deployable due to lack of a resolving DNS server discovery 
> mechanism. The fact that the first IPv6 RFCs were published 9.3 years 
> ago is only relevant upto a point.
> 
> The truth is that nothing similar enough to the transition from IPv4
> to  IPv6 has happened in the past to make any conclusions based on it
> worth  the air and spit necessary to utter them. The only thing we
> know is  that predictions are notoriously hard, especially when they
> concern the  future. But I challenge anyone to show me how we could
> reasonably still  be using IPv4 in the year 2030.
> 
> > IPv6 is already deploying.
> 
> Some random data points:
> 
> - of the "top 100 English language web sites" not one is reachable
> over  IPv6
> - of the "top 100 English language web sites" one times out for AAAA 
> DNS requests
> - of the 213 Amsterdam Internet Exchange member's main websites, 9 are
>  reachable over IPv6 (up from unknown / 4 just under a year ago)
> - of the 213 Amsterdam Internet Exchange members, 59 had an IPv6 
> address enabled on (one of their) port(s)

Would you let us know how you were able to collect such statistics ? 

Thierry