Hi, On Mon, Apr 23, 2012 at 10:40:00AM -0700, Bill Moffitt wrote: > On the subject of being able to turn off V6, I agree with both sides, > but with a caveat. > > On one side, V6 is necessary today in Asia, but, on the other had, it's > not at all necessary or even desirable yet in North America and Europe, > so I think it is desirable to be able to turn it off.
I agree with that. Giving people options is always welcome (especially
on memory- or flash-limited platforms where every single byte counts).
Can I have --disable-ipv4 for everything, please? Most everything I
do these days is done over IPv6, except for web surfing to some minor
sites.
> Right now I don't know of a consumer ISP in North America that will GIVE
> you an IPv6 address, never mind require it, although some business ISPs
> tout V6 routing.
Comcast will give you IPv6 in America (though not everywhere yet, just
rolling out out). T-Mobile USA is beta-testing it on their 3G network.
Almost all business ISPs have it in their portfolio now - since smaller
ISPs have been asking for it for over 10 years now, larger ISPs had to
deliver.
> Because of the foot-dragging in rolling out V6, I believe that adoption
> is not going to be a gradual affair - I believe that, at some point,
> someone (probably in Asia) is going to invent the "Next Big Thing that
> Everybody HAS to HAVE," (i.e. the next new Google, Facebook, Twitter,
> Pinterest, Dropbox, whatever) and it will only be accessible via V6. At
> that point, everyone in North America and Europe will suddenly change
> from ignoring V6 to HAVING TO HAVE IT RIGHT NOW!!!
The "next big thing" is "being able to communicate with your customers,
or with your friends, or remote VPN users that happen to be in Asia".
Even in Europe, we're already seeing large-scale ISPs move towards
"customers will not get a public IPv4 address anymore, just a private
address and be NATted" - so goodbye to dyndns-provided home accessibility,
etc. (Telefonica announced this end of last year for Spain)
> I believe the ISPs will mostly be blindsided - the few who thought that
> this might happen will be able to move users to V6 and will be "winners"
> in this scenario, while those who didn't plan ahead (or thought they'd
> be able to move people gradually to v6) will be roasted alive in the
> court of consumer opinion. It will be a "crisis" that "no one could have
> predicted."
Indeed, for many people this will come as a complete surprise and nasty
shock. Others will just laugh.
> While we consumers in N. America and Europe can still afford to be
> complacent for a while, I think that we, as OpenWRT developers, need to
> be very diligent in ensuring OpenWRT "plays well" on V6 in anticipation
> of this event, should it come to pass. It may be a nice opportunity for
> OpenWRT to get some nice publicity by "saving the day" when the "crisis"
> occurs.
OpenWRT is already pretty good :-)
(Now, what I'm not sure whether OpenWRT already has this: to fully utilize
IPv6 over here, what you need to have is dynamic IPv6 prefix support
using DHCP-PD. As in "router queries ISP for a prefix, ISP assigns
2001:db8:1::/56, router assigns 2001:db8:1:1::/64 to LAN,
2001:db8:1:2::/64 for WiFi and informs radvd that this is the prefix
to be used". AVM's Fritz!Box does this nicely today, but not with open
source components...)
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany g...@greenie.muc.de
fax: +49-89-35655025 g...@net.informatik.tu-muenchen.de
pgpMdT5VSf5C3.pgp
Description: PGP signature
_______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
