[deepspace] Re: Understanding the difference in aggregation between a single and multiple RIR model

[Michael Richardson <[e-mail]>]

Remco van Mook <[e-mail]> wrote:
    > There is a third construction, in which the dilemma does not arise:
    > make aggregation a property of the address format. If an address
    > derives from where something is, then things in one place share a
    > prefix because they are in one place. The colony is one prefix because
    > it is one colony: whichever agencies built it, whoever they asked,
    > including nobody. One route per body falls out of the same arithmetic,
    > with no registry consulted below the initial designation.

Sure.
I'm not sure this is different than delegating directly from ICANN to the
_Mars RIR_, (once such a thing exists).

    > Tony quotes Rekhter's law, and I'd finish the thought: in space,
    > topology is orbital mechanics. Addressing that follows topology is
    > addressing bound to trajectories.

So, orbital elements as part of the prefix ?  :-)
I'm reminded of: https://datatracker.ietf.org/doc/draft-hain-ipv6-geo-addr/

    > I wrote a problem statement along these lines earlier this year
    > https://datatracker.ietf.org/doc/draft-vanmook-expanse-problem-statement/
    > , and a strawman architecture exists as input for whatever effort forms
    > here. For calibration, the initial IANA designation it needs is a /14:
    > smaller than either proposal spends on Mars alone.

Interesting.
Nice use of EXPANSE acronym. Very nice.  I like that you've called out:

   Traffic tromboning:  Architectures that implicitly treat Earth as the
      default interconnection point force space-to-space traffic through
      terrestrial gateways.  At interplanetary distances a tromboned
      path is not an inefficiency measured in milliseconds but in tens
      of minutes per detour, and in some geometries in availability
      itself.

I also like how you've linked this:
      topological adjacency changes continuously, relative velocity
      constrains contact windows and link scheduling, and binding
      addresses to orbital position repeats, at solar-system scale, the
      locator/identifier conflation the Internet has struggled with for
      decades.

I found your document to be a useful read.
Does the IETF need to push this to RFC before it can be considered by other
stakeholders?  I hope not.

}   This document makes no requests of IANA, which is precisely the
}   point: future documents should make well-formed ones.

!!

    > Happy to share once we sort out a forum. I was actually looking forward
    > to the interim meeting as it was communicated to people within the
    > address policy community, which is (apart from some notable exceptions)
    > mostly absent from this list.

Yes, I also am disappointed that this option was cancelled.

--
Michael Richardson <[e-mail]>, Sandelman Software Works
 -= IPv6 IoT consulting =-                      *I*LIKE*TRAINS*