Re: IPv6 cases sought

[David Farber <dave () farber net>]

Begin forwarded message:

From: Brad Templeton <btm () templetons com>
Date: February 5, 2007 11:15:42 PM EST
To: David Farber <dave () farber net>
Cc: huizer () cs uu nl
Subject: Re: [IP] IPv6 cases sought

On Mon, Feb 05, 2007 at 06:00:59AM -0500, David Farber wrote:
> We all know that incentives to implement IPv6 are few and far
> between. To investigate if we there is a solid case for IPv6 at all,
> the Dutch IPv6 Task Force (which I chair) has embarked on a study of
> real-life IPv6 implementations. If we succeed in getting a good

You probably know you won't find too many answers from the
real life implementations.

IPv6 has many features, but the main one the end user might
care about is it gives you the opportunity to have a real, reachable
IP address where before you might be behind NAT.

Who wants a real, reachable address?  That turns out to be largely
peer to peer applications, applications where any machine may
wish to be a "server" and receive connections at unpredictable
times, without introduction, from an outside party.

Who wants to run peer to peer applications?  Well, voice over IP
is one of most compelling apps today that doesn't have the taint
some other P2P apps have attained.  And NAT is indeed the bane
of VoIP (and other associated media applications like video calls.)

Yet, if I put up a SIP (VoIP) phone and give it an IPv6 address
what will happen?   Nada.  Only a tiny fraction of other people
running VoIP software will be able to make use of it to call you.
For the other 99%, you must solve the NAT problem.  So there is
no reason to even bother with IPv6.

Who else wants to run P2P applications?  Innovators.   The internet
begain as a peer to peer network.  Almost all apps on it were initially
peer to peer.  All the innovation was done P2P.  The IPv4 shortage
moved people behind NATs, and we moved to Peer-to-Server for many
apps.   Now the typical PC can only do peer to server easily, unless
it plays NAT tricks.

But it's hard for the rest of the world to deal with barriers that
are only seen by innovators, who are trying to come up with a
new application nobody has yet thought of.  There is no clamour or
demand yet, so nobody saying "I want my ipv6, in case somebody
comes up with something really cool I can be a server for."

It's so bad that the innovator HAS to come up with a NAT workaround
to survive, and thus no need for v6.  This is what Skype did, it's
what made them billions.

To make it worse, there are two negative pushes.

1) Many P2P apps are controversial.   They eat large amounts of
bandwidth, and many users are infringing copyrights or doing something
else of no particular interest to network managers.  They see no reason
to work hard to make it easier to innovate in the P2P space.

2) For many people, NAT was a poor man's instant firewall.  Especially
for Windows boxes.  Being a server on the open net is a risky
activity in a world of malicious software.  Again, people saw no reason
to make it easier to take this risk.  You can of course do firewalling
in IPv6 just fine, but people don't want to go out of their way to
make something easier that has risk they have to then work to mitigate.


It's going to be a long path.  I asked my ISP to route me IPv6.  He
said, "You're the first to ask.  So sorry, not until more customers
are asking."

However, when we block innovation, we often pay the highest cost
of all,  yet never know it until much later.


-------------------------------------------
Archives: http://archives.listbox.com/247/
Powered by Listbox: http://www.listbox.com