[ih] IP addresses are not phone numbers, was Some Questions over IPv4 Ownership

[Dave CROCKER]

On 10/17/2010 7:34 AM, John Curran wrote:
> There have been discussions for geographic addressing in IPv6, but
>    it doesn't improve routing (actually, the converse) unless the path
>    of connectivity actually follows the geography.

Deering conducted discussions as part of his original proposal effort (the first 
SIP) that fed into the current IPv6.  As I understand it, there was followon 
research in the area (via Lixia?).  Note that none of this surfaced all that 
publicly, which says a lot about how it progressed.

In order to have it work somewhat in the style of a locator, there would need to 
be topological constraints, with "local" internet exchanges.  This would permit 
routing to the area and then dispatching to the correct local carrier.

The existing scheme does not impose topological constraints, so this would 
constitute a massive paradigm change.

The alternative approach is to have the address operate strictly as an 
identifier, with a mapping layer down to an actual locator.  We'd get to invent 
another 'address', along with the mapping mechanism.  Whoopee...

My own view is that IP Addresses need to be treated strictly as locators and 
that the infrastructure needs to be tailored solely for that.  This is a variant 
of saying that the current Internet IP model is fine and we should not mess with 
it, except to make it more efficient at doing what it has been doing so well.


>    The problem with this model is that it's completely divorced from
>    typical reality, whereby numerous distinct international ISPs all
>    are vying for business in a country.

When Deering started his discussions, I felt that geographic addressing was 
essential, in order to buy customers freedom from ISP lock-in.

Per the above view, I now believe that the most powerful path is to get 
"identifier" functions entirely out of the IP Address, and allow IP Addresses to 
have rather late binding, to permit easier re-binding.  Freedom comes from not 
embedding lower-layer addressing information into hosts.  That puts me into the 
"move it out of the core infrastructure camp" which is not where most sympathies 
or efforts seem to lie.

That is, make host-level rendez-vous activities rely on an identifier and not a 
locator.  (Except for highly dynamic transitions, I remain a fan of having 
domain names fill this role.)  The work, then, is on dynamic registration and 
binding, and name-based applications.

d/
-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net