ietf
[Top] [All Lists]

Re: WCIT outcome?

2013-01-04 18:03:47
On Fri, Jan 4, 2013 at 1:44 PM, Tony Hain <alh-ietf(_at_)tndh(_dot_)net> wrote:

Ted Hardie wrote:
On Thu, Jan 3, 2013 at 4:59 PM, Tony Hain <alh-ietf(_at_)tndh(_dot_)net> 
wrote:
Like it or not, governments are fundamentally opposed to the open nature
of
'the Internet', and they always will be (even the 'reasonable' ones).
Managing information flow is how they derive and exercise power

Aside from the whole "consent of the governed" issue, this perspective
seems
historically fairly short-sighted.

And that consent is based on information availability. Manage the
information, and you manage the consent.


Possibly; the extent to which that management is obvious may, of course,
drive other behavior (cf. самизда́т [Samizdat] and similar efforts).



...
Yes, there were treaty-based efforts to set some common
understandings.  Some of those were quite useful, but whenever they
actually
impinged on workings of the real network you got smuggling.  In some
cases, lots
and lots of it, with encouragement from some of the participating
nations.
In
other words, some countries tried to control their participants in these
international networks very tightly.  Some were content to let their
merchants
get fat off it instead.  There was no universal response.

Shipping Merchant / Harbor Master == ISP    : As you see with the
resolution
signatories, there is still no universal response.


Given this, I am somewhat confused why you said "Governments are
fundamentally opposed to the open nature of 'the Internet', and they always
will be (even the 'reasonable' ones)."  If some governments approve open
harbors for shipping and (and its Internet parallel), how are those
governments opposed to the open nature of the Internet?  Did you mean "Some
governments"?


...
In this new effort at a multilateral framework, we are seeing a clash
between a
desire for sovereign control of the Internet and a desire to reap the
benefits of
open participation.  I think our role in that is to make sure all
involved
understand:  the benefits of the Internet's network effect; the risks in
allowing
nations through which traffic passes to assert sovereignty over the
flows,

As if professional information control practitioners do not understand the
risks of information control at a much deeper level than anyone in the IETF
could ever hope to ...


The pace and chance of topological change are pieces that are commonly
misunderstood outside the world of Internet engineering.  The amount of
surprise you get when you explain to someone why a routing announcement can
pull all the traffic for your company through $RANDOM_COUNTRY seems
perennial.



especially given both the pace and chance of topological change; and the
reality
that entities outside of governments control the paths that packets
actual
traverse.

The entities that operate and control the paths do so at the pleasure of
the
governments, just as the merchants and harbor masters did in your example
above.


This is not the point I was making.  The operator of an infrastructure may
do so at the pleasure of the sovereign power in whose borders that
infrastructure sits, but that does not guarantee that packets will flow
over that infrastructure at the pleasure of that sovereign power.  If I do
not wish my packets from Den Haag to Paris to flow through the LINX, for
example, I can ensure that they do not.  That works because there are
multiple paths over which I might send the flow.  Internet routing is not
as open as sea-born trade; there are  paths that provide choke points.
But single points of failure are not an advantage to Internet engineering,
and providing them to engage well with specific governments both harms the
Internet and, at least in some lights, harms the governments' interests as
well.  We can make sure that's understood and build out the infrastructure
to minimize the choke points as best as we can.


As long as governments are pleased, the operators can live in a
fantasy land where they are outside government control.


I don't think that this is a realistic characterization of operators'
attitudes.


The Dubai
discussions show what happens when a collection of governments are no
longer
pleased ... If that noise level gets high enough the non-signers will have
to respond just to maintain some degree of cooperation on other matters
they
care about.

Our role is to recognize that there are much bigger issues than the simple
process of bit delivery. Yes bit delivery is important, and dynamic, but it
is equivalent to laying the tracks. Standards must be maintained for
consistent interworking, but it is not the path that matters; just as with
rail cars, it is the content that provides the value.


I certainly believe that the value is in the content; packets with no
payload aren't terribly interesting, but I don't see how your conclusion
changes the aims of the IETF in this regard.



Being highly dynamic
makes things harder to control, but not impossible. Even in the 'open'
countries, a few changes in something as apparently disconnected as tax
laws
would dramatically change the decisions and behaviors of the operators that
are 'outside of government control'.
...


Patrik Fältström wrote:
On 4 jan 2013, at 01:59, Tony Hain <alh-ietf(_at_)tndh(_dot_)net> wrote:

Like it or not, governments are fundamentally opposed to the open
nature of 'the Internet', and they always will be (even the
'reasonable'
ones).

Because I do not think generalization is really a reasonable thing to do,
and even
dangerous when discussing governance issues, I disagree with this
statement.

I agree there is danger here, but I believe the greater danger is embodied
in this thread due to the lack of acknowledgement that governments always
have control, even if the control points are not obvious.


You've returned, though, to doing what Patrik noted was a problem: treating
all governments as if they were a homogeneous mass.  A government controls
a territory and has a set of interests.  They are not an homogeneous mass.



Governments want just like businesses success in whatever they do. That
can in
general be divided in two wishes. Short term, in the form of being
re-elected (or
not thrown out of their office) and long term, as in growth of the
revenue
of the
country they govern.

The second point (restated as 'prosperity for their constituency') is
really
just a continuation of the first point.


They of course have pieces of their operation that belong to law
enforcement
agencies, but they also have those that are responsible of finding rules
so that
non-public sector can grow (to later increase for example tax revenue).

Because of this, I encourage people to not generalize per stakeholder
group, but
instead acknowledge that there are different *forces* that are orthogonal
to
each other, and calculating "the correct" balance between them is hard.
Or
rather, different people do for different reasons get different results
when
calculating what the for them proper balance is.

That is why I personally am against generalization that a stakeholder
group have
one specific view.

I would agree that they all demonstrate a different calculation for the
importance of various aspects of controlling information flow. My primary
point here is that the IETF has to accept this as a reality. So far the
position has been 'this is the technology, & governments need to adapt'. At
the end of the day though, it is the governments that are really in
control,
and if the IETF does not want a 'forced adaptation', it need to evolve and
listen to the needs of a stakeholder group they have tried hard to ignore.


If I can re-state that as "engage with", I would agree; the IETF needs to
work with the Internet Society and others to engage in this process, or we
are pretty much guaranteed not to like the outcomes.  But this is not a
case where we need to lie back and think of England; we need to be active
parts of this process, rather than passively accepting what governments
want.  First and foremost because of the potential damage to the network
and its users, but also to avoid induced multiple personality disorder from
different governments saying different things.  Sweden is not Syria and
Chile is not China.




Specifically governments.

If when pushed to be forced to choose between two choices *all*
governments
wanted to have control, we would have had many more governments signing
the
proposed treaty that was on the table in Dubai.

Instead, when being forced to choose, they picked openness and multi
stakeholder bottom up processes.

As noted above, this is due to relative happiness, and limited noise. If
the
noise level grows or they become unhappy, the independence of the standards
process is a minor pawn in the game, and will be sacrificed to protect
something more important.


I believe we can maintain the IETF as independent.  Under some dire
outcomes, its output might cease to be called "standards" by the lights of
certain countries or treaty organizations, and there are certainly risks
there.  But good specs are good specs, and government mandates have changed
in favor of network effects before.

In the mean time, though, far better to thank those who are engaging in
this process, help them where we can, and keep our eyes on the real end
goal:  making the Internet work for everyone.

regards,

Ted


...

Tony



<Prev in Thread] Current Thread [Next in Thread>