On Feb 19, 2026, at 4:57 PM, Tony Li <tony.li@tony.li> wrote: ... A first draft did ask IANA for this, however, I have been told that I was grossly inappropriate and that I had to ask the RIR communities. So here I am.

Tony - To be quite clear, the earlier draft that I saw suggested "IANA designate an RIR” –– I don’t believe that I used the phrase “grossly inappropriate” or that’s another reviewer, but yes, I did indicate that there are actually some processes involved with delegation of RIR regions, and it would be better if the document read something like: "This document requests that IANA work with the Internet numbers registry community to provide for issuance of general purpose IP number resources for outer space in accordance with this document.”

We would like one RIR to manage this block. I suspect that if no one else volunteers, that ARIN will take up this mantle. Is that ok with RIPE?

So general purpose IPv4 address blocks are issued to RIRs accordingly global number resource policy, and then there’s some “regional" policy regarding from the block to the individual ISPs/LIPs. How exactly that happens for this new “region” is something that the ISP community should probably discuss. There’s an entire set of services (RADP, IRR, RPKI, reverse DNS, etc.) that operators may expect to be operating for these new allocations which need to be looked into, and then the manner of governance for same. Thanks! /John John Curran President and CEO American Registry for Internet Numbers p.s. As "I suspect that if no one else volunteers, that ARIN will take up this mantle.” – I don’t know about other RIRs, but ARIN would only take on additional duties if there was clarity on what was being asked, and then our members or Board gave indication that we should do so. As such, I have no idea if ARIN would perform such a role, since what’s involved in presently speculative.

Gert - Yes, the “IPv4” qualifier was unnecessary and likely not relevant; instead read that as - “So general purpose Internet number resource address blocks are issued to RIRs accordingly global number resource policy, and then there’s some “regional" policy regarding from the block to the individual ISPs/LIPs.” I.e, the question remains as to why a distinct RIR region is necessary here, rather than (as Randy said) just having the involved operators go to any RIR and obtain an appropriately sized IPv6 allocation. /John

On Feb 20, 2026, at 3:36 AM, Gert Doering <gert@space.net> wrote: Hi,

On Thu, Feb 19, 2026 at 10:26:18PM +0000, John Curran wrote:

...

So general purpose IPv4 address blocks are issued to RIRs [..]

why would anyone want to build this with IPv4?

Gert Doering -- NetMaster -- have you enabled IPv6 on something today...?

SpaceNet AG Vorstand: Sebastian v. Bomhard, Karin Schuler, Sebastian Cler Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: Dr. Frank Thiäner D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279 ----- To unsubscribe from this mailing list or change your subscription options, please visit: https://mailman.ripe.net/mailman3/lists/address-policy-wg.ripe.net/ As we have migrated to Mailman 3, you will need to create an account with the email matching your subscription before you can change your settings. More details at: https://www.ripe.net/membership/mail/mailman-3-migration/ <signature.asc>

On Feb 20, 2026, at 12:52 PM, Tony Li <tony.li@tony.li> wrote: Hi John, On Feb 20, 2026, at 3:42 AM, John Curran - jcurran at istaff.org <mailforwards@cloudmails.net> wrote: There’s multiple ways to make this work - E.g. a new RIR for the “region” - simple enough - form an RIR per ICP-2 (or the new RIR Governance doc), get organized, and then apply for recognition. Or just make use of an existing RIR and apply under their existing policies for ISPs/LIRs as necessary to get the space needed. Either of these work - as the Internet Number Registry System is more than a table of entries; it’s a mechanism for the community come together, interact and self-govern operation of an actual system - and either approach provides clarity of who the community is, and how applicable policies are established and updated. Either approach allow for the operators to interact with the rest of the community - fairly important for making sure services RDAP and RPKI work across the entire registry. The approach that lacks clarity is what Tony seems to propose - there’s would be a new region with a distinct set of policies and set of operators but they not have an RIR governance model (ala ICP-2/RIR Governance doc) or any interaction with this community…. They’d be run by an existing RIR as if they were a distinct RIR, but minus the community and the governance model. I apologize for not being clear. Let me see if I can do better. The point is to have a single point of contact where agencies can place address space requests for outer space. Whether that is an existing or new RIR is a detail. The goal is aggregation for efficient routing. How do we get there? If I need to say something differently, please send text. Tony - So the Internet numbers community is made of up of regions, and the community in each region organizes itself into an RIR… Each RIR community has a service organization (their RIR) which has services and fees, governed by a member-elected body. In addition, each community has a policy development process where they establish and update registry policies. Agencies may want to "place address space requests for outer space” but it’s a little more than that - self-governance of Internet number resources mean that they will need to participate in these things in order to to make sure that it meets their needs. If they want to use an existing RIR for such, that’s great - that RIR’s community will establish appropriate policies, fees, and services. If they want to establish a new RIR,, that’s also a great option, and the resulting RIR (once recognized) will perform these functions. As to whether that’s just "a detail”, I would say it is not - since the choice will determine how you go about getting the policy that you need established. Does that help clarify things? /John

On Feb 20, 2026, at 2:08 PM, Tony Li <tony.li@tony.li> wrote: Hi John, So the Internet numbers community is made of up of regions, and the community in each region organizes itself into an RIR… Ok, so which region does Mars fall into? Indeterminate at this time for an organization that is based on Mars. For organizations based on Earth, they can utilize their regional RIR for their number resource needs under existing policies. Each RIR community has a service organization (their RIR) which has services and fees, governed by a member-elected body. In addition, each community has a policy development process where they establish and update registry policies. Agencies may want to "place address space requests for outer space” but it’s a little more than that - self-governance of Internet number resources mean that they will need to participate in these things in order to to make sure that it meets their needs. That’s a tall bar that you’re setting. Does every ISP today have to participate? ISPs participate in at least one RIR in order to obtain and manage their number resources. The level of participation varies, with some also participating in the RIR’s policy development processes, some getting involved in RIR governance, and some doing both. Many ISPs don’t get involved until there’s issue of great interest to them, and in the meantime rely the efforts of their colleagues to handle more routine matters. If they want to use an existing RIR for such, that’s great - that RIR’s community will establish appropriate policies, fees, and services. If they want to establish a new RIR,, that’s also a great option, and the resulting RIR (once recognized) will perform these functions. As to whether that’s just "a detail”, I would say it is not - since the choice will determine how you go about getting the policy that you need established. Does that help clarify things? Not really. They are mission focused. Their requirement is address space. Full stop. They have about as much interest interest in RIR policies as my kids do. Which is actually quite similar to many ISPs… but then again, there are existing policies which clearly cover the requests that most ISPs need to make for number resources, and that is not the case with respect to policies for IP addressing for other planets. Thanks, /John

Tony, On Fri, Feb 20, 2026 at 1:08 PM Tony Li <tony.li@tony.li> wrote:

Hi John,

So the Internet numbers community is made of up of regions, and the community in each region organizes itself into an RIR…

Ok, so which region does Mars fall into?

The Moon and Mars are irrelevant to this discussion; there are no people, no legal system, no contracts, or no companies on the Moon or Mars. For missions to those bodies, all those things exist here on Earth, not on the Moon or Mars. Maybe someday in the far future, we will have colonized the Moon and Mars, and they will have all those things, and then we can talk about independent RIRs for the Moon and Mars, but that is mostly science fiction at this point. Today, and for the foreseeable future, missions to the Moon and Mars are operated from here on Earth, and their governance also exists here on Earth, and within various countries on Earth. I think the proper example for missions to the Moon, Mars, and other celestial bodies is Antarctica. While technically Antarctica is in ARIN's service region, entities that operate bases there obtain Internet resources from the RIR for the home country of the operating entity, or from the service provider they use for connectivity. This is so the administrative and technical people work with those they are familiar with, and contracts are formed in accordance with the legal systems in the RIR's service region. I apologize for not being clear. Let me see if I can do better. The point

is to have a single point of contact where agencies can place address space requests for outer space. Whether that is an existing or new RIR is a detail. The goal is aggregation for efficient routing. How do we get there? If I need to say something differently, please send text.

Tony

Are you really sure the organizations that will use these resource blocks want a single contact and a single contracting entity within a single legal jurisdiction? NASA and US-based commercial space entities would likely be fine with contracting through ARIN and under US Law. However, ESA and European commercial space entities might prefer contracting through RIPE with Dutch and EU Laws. Furthermore, China and India might prefer to continue working with APNIC as they do now. Unfortunately, today, the Earth's political and legal environment will shape operations and the governance of missions to the Moon, Mars, and other celestial bodies, and the Internet Technical community isn't going to be able to change that. Now, I would support allocating an IPv6 prefix outside 2000::/3 and a block of ASNs for Outer Space and for entities supporting spacecraft and missions to the Moon, Mars, and other celestial bodies. The distinction should be that satellites that provide service to Earth come from 2000::/3. However, satellites, ground stations, and other systems, such as ground simulators that support missions to the Moon, Mars, and other celestial bodies, should use this new prefix and block of ASNs. But allocations from this block should be administered and registered through the current RIRs, in my opinion. Unless you want to invite an international treaty organization to administer the blocks, however, the Internet technical community has to date resisted the involvement of such an organization in Internet resource management. So, my vision would be for the TIPTOP WG to produce a document proposing the allocation of an IPv6 block outside of 2000::/3 and a block of ASNs, define the qualifications for use of the block, and request that IANA and the RIRs develop policies to ensure allocations to organizations are consistent with those qualifications. Thanks. -- Thank you / Ho Pidamayado / Miigwech =============================================== David Farmer Email:farmer@umn.edu Networking & Telecommunication Services Office of Information Technology University of Minnesota 2829 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414 Cell: 612-812-9952 ===============================================

On Fri, Feb 20, 2026 at 7:28 PM Tony Li <tony.li@tony.li> wrote:

Hi David,

The Moon and Mars are irrelevant to this discussion; there are no people, no legal system, no contracts, or no companies on the Moon or Mars. For missions to those bodies, all those things exist here on Earth, not on the Moon or Mars.

From a routing perspective, governance, people, legal systems, contracts, and companies are irrelevant. What is relevant is devices that speak IP and their topological location. We obviously already have devices on Mars and the moon. More coming. Next month, humans will be briefly added to the mix.

People, legal systems, and contracts are necessary components of our society, which has organized itself to create spacecraft already on the Moon and Mars; ignoring them at this point is counterproductive, as without them, you would not have spacecraft on the Moon or Mars to route any packets.

I think the proper example for missions to the Moon, Mars, and other celestial bodies is Antarctica. While technically Antarctica is in ARIN's service region, entities that operate bases there obtain Internet resources from the RIR for the home country of the operating entity, or from the service provider they use for connectivity. This is so the administrative and technical people work with those they are familiar with, and contracts are formed in accordance with the legal systems in the RIR's service region.

And if there was the prospect of an interconnected network there with a proliferation of devices, I would be equally concerned. Unfortunately, I don’t think that the long-term growth prospects are similar.

I apologize for not being clear. Let me see if I can do better. The

...

point is to have a single point of contact where agencies can place address space requests for outer space. Whether that is an existing or new RIR is a detail. The goal is aggregation for efficient routing. How do we get there? If I need to say something differently, please send text.

Tony

Are you really sure the organizations that will use these resource blocks want a single contact and a single contracting entity within a single legal jurisdiction?

I know that the organizations (space agencies) that will use these resources are NOT experts in networking. For them, this is one minor technology that helps enable their missions. Hopefully, they will follow our recommendations about how to deploy IP and aggregate their prefixes. We owe it to them to make this possible and convenient.

The scientists and engineers who operate the missions don't care, but contract personnel, lawyers, and even network engineers across all those organizations do care and participate to ensure the organizations get what they need. You need to talk to the contracting personnel, lawyers, and network engineers in those organizations. I guarantee they will care, or at least they should. I come from a similar mission-focused organization. And my job is to ensure my scientists can transfer petabytes of data to supercomputers for processing and work with collaborators worldwide. They don't care how it happens, but they do care that it happens, and that my job and literally hundreds of other people in the Research and Education Networking (REN) community worldwide. We are subject matter experts who work with contracting personnel and lawyers in our organizations to obtain equipment and other resources to make that happen. Similar people exist in space agencies and their contractors; I know they do. I work with them from time to time. Because my scientists also consume data from space missions.

NASA and US-based commercial space entities would likely be fine with contracting through ARIN and under US Law. However, ESA and European commercial space entities might prefer contracting through RIPE with Dutch and EU Laws.

Agreed, but then we end up with an ARIN block and a RIPE block in space (and more…). They want to interconnect their networks for mutual backup and suddenly we have a routing mess.

Those are the policies for 2000::/3, which call for hirecical infrastructue based aggregation. Define what you expect from a new block with some detail, and define the technical details of how the aggregation should work. Don't define who should do what or how; leave this to IANA and the RIRs.

Furthermore, China and India might prefer to continue working with APNIC as they do now. Unfortunately, today, the Earth's political and legal environment will shape operations and the governance of missions to the Moon, Mars, and other celestial bodies, and the Internet Technical community isn't going to be able to change that.

That seems defeatist. We haven’t even tried.

Maybe, but I don't think the Internet technical community can change the legal and political environment we live in today. I think we have to live within it.

Now, I would support allocating an IPv6 prefix outside 2000::/3 and a block of ASNs for Outer Space and for entities supporting spacecraft and missions to the Moon, Mars, and other celestial bodies. The distinction should be that satellites that provide service to Earth come from 2000::/3. However, satellites, ground stations, and other systems, such as ground simulators that support missions to the Moon, Mars, and other celestial bodies, should use this new prefix and block of ASNs.

We are in agreement this far.

But allocations from this block should be administered and registered through the current RIRs, in my opinion.

So then we end up with all RIRs jointly managing this block? How does that work? How does NASA get a Mars prefix that aggregates with a Mars prefix that RIPE assigns to ESA?

The RIRs already jointly manage 2000::/3 with policies optimized for hierarchical infrastructure-based aggregation, as charged by the IETF. If you want a new block with other policies, define what you are looking for, but don't get hung up on how to accomplish it; focus on the details of the aggregation properties you want.

Unless you want to invite an international treaty organization to administer the blocks, however, the Internet technical community has to date resisted the involvement of such an organization in Internet resource management

I don’t much care who does the administration, as long as it gets done in a coordinated and useful way. Having multiple players all trying to manage a common resource seems like a recipe for disaster.

I think you are focusing on the allocation function, which is just one small component, and it might be simpler if one entity handled it. However, there are other components to the administration, contracting, a service desk, etc. These components are better regionalised; this is why the RiR system was created.

So, my vision would be for the TIPTOP WG to produce a document proposing the allocation of an IPv6 block outside of 2000::/3 and a block of ASNs, define the qualifications for use of the block, and request that IANA and the RIRs develop policies to ensure allocations to organizations are consistent with those qualifications.

That is already the jist of my draft. Please help me with the wording.

I think you need to address George Herbert's question about transit. When I read the document, it seemed like a spacecraft magically appeared in orbit around the celestial body in question, yet it can take years for a spacecraft to get there. When I read your document, it sounds like the RIRs have ignored outer space. To the contrary, the RIRs haven't been charged with thinking about outer space. So far, the RIRs have only been asked to create a system that allocates resources to ISPs on Earth and with hierarchical infrastructure-based aggregation. Your document needs to have the IETF authorize IANA to allocate an IPv6 block outside 2000::/3 and a block of ASNs for use in outer space, specifically for missions to other celestial bodies. Provide details on how the desired celectual body-based aggregation is to work. Then ask the RIRs to create policies to allocate space to space agencies and their contractors based on the desired celestial body-based aggregation. Regards,

Tony

-- Thank you / Ho Pidamayado / Miigwech =============================================== David Farmer Email:farmer@umn.edu Networking & Telecommunication Services Office of Information Technology University of Minnesota 2829 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414 Cell: 612-812-9952 ===============================================

Hi, On Fri, Feb 20, 2026 at 08:56:00AM -0800, Tony Li wrote:

...

why would anyone want to build this with IPv4? Bandwidth is extremely constrained. Agencies may choose to optimize for performance.

You might not have noticed, but there seems to be a shortage of IPv4... So whatever you are aiming for, the changes for "you'll get a big chunk of unused IPv4 addresses, for free" are close to zero. Gert Doering -- NetMaster -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard, Karin Schuler, Sebastian Cler Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: Dr. Frank Thiäner D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279

mornin' tli,

...

You might not have noticed, but there seems to be a shortage of IPv4... Thanks, I’ve been noticing since 1991.

i thought gert was old enough to remember you led the CIDR fight as i know you are aware that non-trivial blocks of ipv4 space are no longer available from registries (except maybe afrinic), i do not understand how you think you are going to solve this at layer nine. how much space do you think you are going to need in a year, five, ten, ...?

Agencies are using prefixes from their existing allocations. Right now, that’s not problematic because their needs today are tiny. However, this builds momentum in a direction away from aggregation. As they hit address space constraints, they are then likely to resort to private IPv4 (RFC 1918) addressing.

i am starting, only starting, to understand where you're coming from

We can help them by giving them a clear path to a different alternative.

i sympathise with your wanting to avoid a trajectory which de-aggregates agencies' existing space. not a good start. i am trying to understand how you think this is going to scale. and what will intra-solar-system routing look like in 50 years? bgp-like? link state, with volatile links? we made the mistake of hacking addressing without routing once already.

...

why would anyone want to build this with IPv4? Bandwidth is extremely constrained. Agencies may choose to optimize for performance.

do you really think v6 header size is that big a problem? if you could use ipv6, i can imagine the rir system finding a way to give you a big chunk or whatever. it would likely take some discussion, but it leaves the deaggregation problem on your side of the net. otoh, there are some public service oriented folk holding large blocks of ipv4 space who might be willing to negoriate, e.g. ARDC. perhaps i am overly concerned by how this will scale in decades, with many agencies, countries, planets and moons, ... i can't help thinking it is déjà vu all over again. i do not like repeating the bad parts of history, only the good ones.. randy

hi tli,

...

i can certainly understand your desire for geographic aggregation in this case. those are some long thin wires. though i note that geo aggregation has not worked out in the past. and i still have the bad taste of the ipv6 attempt in my mouth.

but do you think there is real hope that, in this case, the cost of moving from one location to another makes moving so rare that it will dwarf the painful renumebering problem? is aggregation worth the renumbering pain?

Yes, I think that anyone changing planets will not care about renumbering.

and the elbonian, tralfamdorian, and barsoomian expeditionary forces will share the mars address block(s)? will all three announce the /42, or will they deaggregate? randy, who dreads travel time and costs to your address policy meetings :)

hi tli,

...

and the elbonian, tralfamdorian, and barsoomian expeditionary forces will share the mars address block(s)? will all three announce the /42, or will they deaggregate?

The hope is that they will cooperate. While this is, of course, optimistic, there is ample reason to believe that they will do so. The relevant space agencies seem to be cooperating so far.

i did not realise you were *that* much younger than i :) here on this self-destructing mudball, organisations with different external topologies almost always have rather different announcement policies. so, are you suggesting that the 'relevant space agencies' share the same external topology? and you think that this will stand the test of time? yes, the fuzzball backbone used to be homonogous, but then ... is curtis gonna dress in a space suit and update the solar system tables on wednesdays? a george santayana quote comes to mind randy, who is trying to wrap his head around interplanetary wednesday