[governance] Has the technical community failed wrt IPv6' .... Governance Frameworks for Critical Internet Resources'

[Ian Peter]

Well, let me be radical about this and suggest that IPv6 has already failed
and will never be rolled out. What's more, it won't really matter (except to
a few embarrassed people and organizations).

IPv6 belongs to the year 1995. That’s the year that a piece of software
called Windows 1995 was rolled out. It didn't even include a browser because
Microsoft hadn't released Internet Explorer yet. The web probably had only
about one million users.

The world has changed since then and it's just possible that a more creative
way of expanding the number pool might be available to us now that wasn't
thought about then. It is also possible that we are dealing with an adoption
problem of a scale not anticipated at the time (particularly given the long
unanticipated lead time in rollout).

If it's consumers who are supposed to lead the adoption of IPv6, I suggest
it will never succeed. As everyone agrees, there is no business case. NO
business case, no rollout, no IPv6. That's the laws of the universe.

I will also say it doesn’t matter - because the problem is not non-adoption
of IPv6, as we have begun to believe - it's that numbers are supposed to run
out and probably will unless something changes.

If we think through a decent mitigation strategy there are a number of
things that can be done to ensure that doesn't happen for another 20 years
or more. That gives time for a more elegant approach to number pool
expansion than IPv6 to emerge.

To their credit, both Geoff Huston and Randy Bush have begun to think about
these alternative mitigation strategies - although both seem to still cling
at times to the hope that the laws of markets and the laws of human
behaviour will suddenly change and IPv6 will suddenly be adopted but a lot
later than first expected. (That's sometimes called denial). The last great
hope seems to be that when the last number runs out someone (probably in an
underdeveloped country) will scream and we will all change. Yeah, right
on....

Sorry to rain on the parade, but really the answer to the number pool
expansion problem requires us to be realistic rather than hopeful, and to be
prepared to be flexible with approaches rather than clinging to an approach
that hasn't worked.

So let me say it again - the problem is not that people are not adopting
IPv6. The problem is that we have not yet arrived at a strategy for dealing
with number pool expansion that is acceptable to all major stakeholders and
is scaleable to future needs. We need a major rethink - and I really don't
think it will be a difficult problem to solve if we put our efforts into
alternative approaches rather than "flogging a dead horse".

But back to Guru's question about what this has to do with governance. Quite
a lot. Neither IGF in it's current state or current "governance"
institutions of a technical nature are adequate as they exist now to deal
with this problem without further levels of involvement (that's
self-evident). Structural change is probably necessary to ensure the levels
of talent and skills and political and business impact necessary to deal
with this and a host of other (probably more pressing) issues.

WSIS has given us some good principles but little in the way of suitable
structural suggestions. It seems to me that a compelling question is how do
we get some decent structural analysis underway leading to some
recommendations for what might emerge from IGF.

Ian Peter
(Mar Ipanema Hotel and looking forward to a constructive dialogue at IGF!)

-----Original Message-----
From: McTim [mailto:dogwallah at gmail.com] 
Sent: 07 November 2007 12:05
To: governance at lists.cpsr.org; Guru at ITfC
Subject: Re: [governance] Has the technical community failed wrt IPv6' ....
Governance Frameworks for Critical Internet Resources'

Guru,


On 11/4/07, Guru at ITfC <guru at itforchange.net> wrote:
> Not very long back, the thrust of quite a few discussions and submissions
on
> ig was  "... Ig issues are really technical in nature and should be left
to
> 'neutral' technical bodies and experts to decide ... And should not be
> 'politicised' or become a playground for non-experts, including
Governments
> to influence or participate in decision making ...."

Is this an actual quote?? I am guessing not.  If you want to
characterize the viewpoint more fairly, how about  "IG issues are
largely administrative in nature, and shouldn't be "left" to anyone,
rather it's the province of those governments, CS actors and engineers
who choose to participate in the bottom-up policy making processes of
the existing neutral administrative bodies."

>
> So here we seem to have moved 180 degree to assert that the (non) move to
> IPv6 is really a political and policy failure than a technical one.

Who is we?  Paul's mail, which said "economical, financial and
political decisions (or lack of), without any technical issue
involved." was spot on.

The big drivers are economic and financial, with politics coming a
distant third.

For example, if you as a CS organisation or end-user demand that your
ISP give you IPv6 connectivity, then you might get it, but until
enough people demand the service, it won't be supplied.  Since their
is no "killer app" driving IPv6, very few are demanding it, hence the
slow rollout.


I agree,
> for how can the technical community make a move happen or not happen, they
> can only provide (valuable) substantive inputs on the need for the move
and
> the pros and cons ... And people / institutions who govern / play a role
in
> governance need to make the larger decisions of movement

Well yes, and no.  In the case of v6, it was the IETF who set the
standard over 10 years ago, so they clearly made that "happen".   In
the few notable examples where people / institutions who govern have
gotten involved (Japan/S. Korea and to a more limited extent the USA)
in urging IPv6 deployment, governments have "decided" to make it
national policy to push v6 usage, and these have been successful
initiatives in promoting v6 deployments.

Paul is entirely correct that in the absence of such initiatives, it's
then left to the marketplace to drive IPv6 usage.  In other words,
it's largely ISPs who make these decisions based on the market, and
it's largely up to the beancounters in these organizations, and NOT
the propeller-heads.  (If it were up to the engineers, I think we
would all have IPv6 enabled toasters by now.)

>
> An element of policy or politics is inherent in such movements, since
> invariably they have pros and cons and affect different sections
differently
> ... For e.g. one perspective could be that the move to IPv6 is critical
for
> nations as India or China that will need significant 'quantity' of these
> resources in the days to come ... While the move involves costs (of
> migration) for the current players, without commensurate benefits to them.

Well, it will be critical for all network operators certainly at some
point, and that point will be reached at the same time (roughly) for
all network operators whether they are in the USA or in China.  I
think it's nonsensical to say that "country X consumes IP addresses",
because it's actually organisations located in country X that do the
consuming, but that same org may also be in country Y.  LIR
assignments and sub-allocations are made across borders.

>
> So are we also saying that the original wisdom of internet being ideally
> "self-regulated" by an "internal / trade-association" is not really
> valid....

No, WE are not saying that at all.  First of all, the notion that IPv6
was designed by a "trade association" is misguided.   Second, and most
importantly, the bottom up nature of Internet policy making is "the
best of all possible worlds" from a CS standpoint.  It boggles my mind
why CS actors would reject it.  No other policy framework is going to
give CS folk a voice as loud as the one they have now.

And we need to explore governance structures and processes that
> would do both - get the best of technical inputs in, as well as negotiate
> amongst multiple stakeholders to arrive at decisions that are both
> 'desirable' and 'implementable' ....  ("War is too important to be left to
> the generals")

AFAIAC, we HAVE governance structures and processes that can do both.
It's just that some stakeholders don't come to the table(s) where
these decisions are made.

>
> I am also intrigued by Paul's assertion that "... which is that progress
may
> rely, in the end, on demand at the consumer/grassroots level". I am not
able
> visualise a billion Indians standing up to demand IPv6 implementation ...

It will probably only take a critical mass of corporate IT folk in
India demanding that their ISP give them native IPv6 connectivity.
This may only happen once they are unable to get more IPv4 address
space, which may make it harder for their customers to get the
services that Ray is on about.

Do you have IPv6 connectivity at your office?  If not, ask your ISP if
they can deliver it. If they can't, then shop for one who can.  Your
current ISP will get the message eventually.  That's one way to drive
IPv6 deployment.

> Policy making perhaps is more complex than 'meeting the demand articulated
> by consumers', though the needs of individuals is indeed a critical input
to
> policy.

Well, as Janis Karklins said in LA,
"Specifically, the GAC noted the important need for the continued good
management of the IPv4 address space in light of the depletion of the
free pool and urgent need for initiatives by all relevant stakeholders
to ensure the acceleration of deployment and use of IPv6 addresses."

In India, you have an active IPv6 body http://ipv6forum.in/, so
perhaps you can work with them (if you are not already).

-- 
Cheers,

McTim
$ whois -h whois.afrinic.net mctim
____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance

No virus found in this incoming message.
Checked by AVG Free Edition. 
Version: 7.5.503 / Virus Database: 269.15.21/1109 - Release Date: 04/11/2007
11:05
 

No virus found in this outgoing message.
Checked by AVG Free Edition. 
Version: 7.5.503 / Virus Database: 269.15.21/1109 - Release Date: 04/11/2007
11:05
 

____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance