Re: IPv6 adoption behavior

["Eugene M. Kim" <gene@xxxxxxxxxx>]

I'm afraid this whole series of argument is somewhat misled.  Pardon me 
if the whole IPv6 landscape has changed while I was asleep, but I always 
thought that the main goal of IPv6 is not to replace IPv4, IPv4+NAT, or 
anything that stands well-established today.

There are some holes that the current IPv4 with NAT can't reasonably 
plug in a sound manner; as Jeroen pointed out, it could be PC/Xbox/PS2 
games, NetMeeting, or heaven forbid, P2P warez (I don't want IPv6 to be 
flagged as a pirate ship by the government =p).  IPv4+NAT can work 
around this, but only with dirty hacks.  And in many cases these hacks 
are not 100% complete either.

If some of you want to say those limitations are not really important to 
most NAT users, just talk to anyone who played StartCraft behind her NAT 
box and got frustrated how she and her boyfriend can't play online at 
the same time because only one machine behind the NAT box can connect to 
the Blizzard's game server.  Just in case you didn't know, StarCraft is 
still quite popular among kids and teens (it's even an official event in 
an international gaming competition; see www.worldcybergames.com), so I 
guess it's a bit hard to say it's not important to most NAT users.  *grin*

Anyhow, I think that these niche markets are the number one target of 
IPv6.  And one day, with people realizing what P2P is and how they 
actually need it (from their own practical point of view of course; see 
above), the niche markets won't be niche markets anymore.

Yes, it would be also good if IPv6 could `replace' IPv4+NAT one day.  
But it'd be like putting the cart before the horse to say IPv6 doesn't 
have much chance because it can't replace IPv4+NAT easily.  Because that 
is not the only purpose of IPv6.  I mean, if that replacement actually 
happens, good.  If it doesn't happen, who cares?  IPv6 will have its own 
market anyways.  No need to stir up the trouble by shouting `Hey folks, 
you're all doomed in a sinking ship, whose name is IPv4!'

That said, what actually bothers me is the classic chicken-and-egg 
problem.  Application writers are reluctant to add IPv6 support because 
they know that there is little to none of IPv6 infrastructure (read: 
ISPs supporting IPv6) out there.  ISPs, on the other hand, are reluctant 
to do IPv6 business because they know that there are few to none of 
IPv6-ready applications out there.  Two-way secret crush.  I guess what 
we need to focus on is how to become a messenger of love. =)

Cheers,
Eugene


--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------