IPv6 - dead or alive?

[Kevin Keane Subscription]

I came across this article that argues that after 20 years, IPv6 has missed its deployment window and is essentially a dead protocol.

https://arxiv.org/abs/1809.07836

I only agree 80% with the paper, but thought it interesting that by now the possibility that IPv6 will never arrive is being seriously discussed.

Unfortunately, that does not mean that we can ignore IPv6 completely; its very existence still has serious implications even if we aren't using it.

Kevin Keane
Whom the IT Pros Call
760-721-8339

[Andrew P. Lentvorski]

Borrowing from David Thornley: "IPv6 doesn't look any deader than usual
to me."

IPv6 continues to increase linearly.  I don't know why everyone keeps
bitching about that.

The fact that it keeps moving at all is nearly stunning to me given the
sheer incompetence in the technical staff of most ISPs.

You could have a switchover tomorrow.  However, none of the big guys are
going to risk market share by declaring IPv6 only before any of the
others.  So, we will continue plodding along.

From an end user, IPv6 simply doesn't buy them anything tangible.  And
it generally buys them some weird grief until IT sorts it out.

-a

Post by Kevin Keane Subscription
I came across this article that argues that after 20 years, IPv6 has missed its deployment window and is essentially a dead protocol.
https://arxiv.org/abs/1809.07836
I only agree 80% with the paper, but thought it interesting that by now the possibility that IPv6 will never arrive is being seriously discussed.
Unfortunately, that does not mean that we can ignore IPv6 completely; its very existence still has serious implications even if we aren't using it.
Kevin Keane
Whom the IT Pros Call
760-721-8339

[Tracy Reed]

Post by Andrew P. Lentvorski
From an end user, IPv6 simply doesn't buy them anything tangible.  And
it generally buys them some weird grief until IT sorts it out.

I believe that the end of NAT will buy end users a great deal.
Particularly in the areas of VOIP/video/P2P-anything.

[Kevin Keane Subscription]

Yeah, the fact that it is still somewhat growing is why I didn't completely agree with that paper.

From my perspective, the big problem is that we are measuring the wrong thing. The goal isn't to get IPv6 adopted, but IPv4 phased out. We'd need 100% IPv6 adoption before we can even get started with the IPv4 phaseout. At the rate IPv6 is going, we won't get there for at least another 50 years.

Which for me does beg the question this paper is implying: why adopt IPv6 in the first place?

Kevin Keane
Whom the IT Pros Call
760-721-8339 

-----Original message-----

Sent: Tuesday, October 9, 2018 2:56 AM
Subject: Re: IPv6 - dead or alive?
Borrowing from David Thornley: "IPv6 doesn't look any deader than usual
to me."
IPv6 continues to increase linearly.  I don't know why everyone keeps
bitching about that.
The fact that it keeps moving at all is nearly stunning to me given the
sheer incompetence in the technical staff of most ISPs.
You could have a switchover tomorrow.  However, none of the big guys are
going to risk market share by declaring IPv6 only before any of the
others.  So, we will continue plodding along.
From an end user, IPv6 simply doesn't buy them anything tangible.  And
it generally buys them some weird grief until IT sorts it out.
-a

Post by Kevin Keane Subscription
I came across this article that argues that after 20 years, IPv6 has missed its deployment window and is essentially a dead protocol.
https://arxiv.org/abs/1809.07836
I only agree 80% with the paper, but thought it interesting that by now the possibility that IPv6 will never arrive is being seriously discussed.
Unfortunately, that does not mean that we can ignore IPv6 completely; its very existence still has serious implications even if we aren't using it.
Kevin Keane
Whom the IT Pros Call
760-721-8339

--
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

[David Brown]

Post by Kevin Keane Subscription
I came across this article that argues that after 20 years, IPv6 has
missed its deployment window and is essentially a dead protocol.
https://arxiv.org/abs/1809.07836
I only agree 80% with the paper, but thought it interesting that by
now the possibility that IPv6 will never arrive is being seriously
discussed.
Unfortunately, that does not mean that we can ignore IPv6 completely;
its very existence still has serious implications even if we aren't
using it.

I'm fairly certain that their graph of IPv6 traffic is misleading, at
best. Primarily because there are several enormous deployments of
IPv6 that tend to sit behind smaller numbers of IPv4 gateways. For
example, an LTE phone network is all based on IPv6, with each device
having its own IPv6 address. It is up to the carrier whether they
make the IPv6 traffic available to users, but the network carries a
lot of voice traffic over IPv6. There are also protocols, such as
6LOWPAN that have IoT devices using IPv6, with a gateway. So the IPv6
devices are connected to each other, and other devices, but traffic on
the internet itself will be tunneled through IPv4.

There's also countries like Vietnam, which have anamolous results.
Some pages show them having low uptake of IPv6 (e.g. 4.6%) and other
sites show them having much higher (e.g. 75%). It may be that the
first analysis (by Akamai) only sees traffic that leaves the country,
and biases their results to countries they have a presence. I would
expect a large amount of traffic in Vietnam to be local within the
country.

Now, I am a bit annoyed that my new fiber-to-the-house provider (Ting)
still has IPv6 out 2 years on their roadmap. I'll be using
Hurricane's tunnelbroker.net service, since there are work things
where I actually have to have IPv6 routable.

David

[Kevin Keane Subscription]

-----Original message-----

Sent: Wednesday, October 10, 2018 5:48 AM
Subject: Re: IPv6 - dead or alive?

Post by Andrew P. Lentvorski
From an end user, IPv6 simply doesn't buy them anything tangible.  And
it generally buys them some weird grief until IT sorts it out.

I believe that the end of NAT will buy end users a great deal.
Particularly in the areas of VOIP/video/P2P-anything.

That ship has sailed a long time ago. Without going into the merits of the NAT argument, to get rid of NAT, you'd have to get rid of IPv4. And you can't get rid of IPv4 until IPv6 has achieved close to 100% penetration. I plan to live to a ripe old age, but I don't expect to see IPv4 disappear in my lifetime. Worse - now that IPv4 address exhaustion has already hit us, and we already developed new coping strategies, the window of opportunity for IPv6 has passed.

And that was really the point of this article.