Person profile
PersonPublic policy participant
Source description Public policy participant
New proposal by Kevin Loch merged with
public people profile
From : Kevin Loch <kloch () hotnic net>
public people profile
Last updated: 2026-09-26
Source facts
Viewpoint consistency: Consistent
The January sizing suggestion, the warning that actual routability depends on network routing policy, and the April fixed-range suggestion address complementary parts of making PI assignments usable. The April suggestion seeks a way to distinguish PI space in filters rather than retracting the earlier assignment-size proposal.
2005-1: Provider Independent IPv6 Assignments for End-sites
Assessment: Consistent · The January sizing suggestion, the warning that actual routability depends on network routing policy, and the April fixed-range suggestion address complementary parts of making PI assignments usable. The April suggestion seeks a way to distinguish PI space in filters rather than retracting the earlier assignment-size proposal.
Conditions: The /44 minimum and scaling formula are suggestions, not adopted policy. · The proposal for a large static PI range leaves open whether reservation should be regional or global. · RIR assignment policy cannot by itself guarantee that networks route the resulting prefixes.
To qualify for a minimum end site assignment of /44 you must either:
Whether it would work or not depends on routing policies that aren't really in the scope of RIR policy.
This is why it is important to have a large well known and static range (like a /16) set aside for PI assignments.
ARIN-prop-18: IPv6 Direct assignments to end sites · 2005-1: Provider Independent IPv6 Assignments for End-sites
Assessment: Mixed · The suggested minimum size changes materially from the /48 described in August 2005 to the /44 proposed in January 2006. The later message presents a revision informed by meeting feedback and adds qualification paths and a scaling idea. The evidence therefore supports an evolution of the design alongside continuity of purpose, rather than identical terms or an unexplained contradiction.
Conditions: These are different recorded proposals and dates; the messages do not prove adoption or a formal replacement relationship. · A larger proposed assignment is a change in design, not by itself a change in support for provider independence.
No guidance is given for how to utilize direct assignments other than you only get a PI /48 to work with.
These were my suggestions using feedback from the last meeting:
To qualify for a minimum end site assignment of /44 you must either:
I would gladly write up a formal revision or new proposal if requested.
2006-2: Micro-allocations for Internal Infrastructure · 2005-1: Provider Independent IPv6 Assignments for End-sites
Assessment: Consistent · Considering a small /96 for internal uses and a distinct PI assignment range for public route filtering serve different uses. Both comments explicitly consider unwanted announcements or filtering effects, so the different prefix sizes are not conflicting prescriptions for the same use case.
Conditions: The internal-use message is conditional and does not establish a complete position on that policy. · Do not equate an individual assignment size with the size of a reserved aggregate range.
Would a /96 be sufficient if the intended use is for /126-/128's? That might also reduce the concern about them being inadvertantly announced.
ISP's will have no choice but to have different minimum prefix size limits by range.
Viewpoint consistency: No explicit position stated
Viewpoint consistency: Insufficient evidence to assess consistency
The April question proposes considering a /96 for uses needing /126 to /128 prefixes and asks whether that would reduce accidental announcements. One conditional design question cannot establish a stable or changed position over time.
2006-2: Micro-allocations for Internal Infrastructure
Assessment: Insufficient evidence to assess consistency · The April question proposes considering a /96 for uses needing /126 to /128 prefixes and asks whether that would reduce accidental announcements. One conditional design question cannot establish a stable or changed position over time.
Conditions: The suggested size is not evidence that the final proposal adopted it.
Would a /96 be sufficient if the intended use is for /126-/128's? That might also reduce the concern about them being inadvertantly announced.
2006-2: Micro-allocations for Internal Infrastructure · 2005-1: Provider Independent IPv6 Assignments for End-sites
Assessment: Consistent · Considering a small /96 for internal uses and a distinct PI assignment range for public route filtering serve different uses. Both comments explicitly consider unwanted announcements or filtering effects, so the different prefix sizes are not conflicting prescriptions for the same use case.
Conditions: The internal-use message is conditional and does not establish a complete position on that policy. · Do not equate an individual assignment size with the size of a reserved aggregate range.
Would a /96 be sufficient if the intended use is for /126-/128's? That might also reduce the concern about them being inadvertantly announced.
Viewpoint consistency: Insufficient evidence to assess consistency
The August 2005 contribution explains provider independence as the main objective and discusses enforcement of the restriction on holding both an assignment and an allocation. Only one supplied contribution addresses this recorded proposal.
ARIN-prop-18: IPv6 Direct assignments to end sites
Assessment: Insufficient evidence to assess consistency · The August 2005 contribution explains provider independence as the main objective and discusses enforcement of the restriction on holding both an assignment and an allocation. Only one supplied contribution addresses this recorded proposal.
Conditions: Explaining enforcement problems does not mean rejecting direct assignment.
This proposal is primarily about provider independence and multihoming as a secondary effect.
No guidance is given for how to utilize direct assignments other than you only get a PI /48 to work with.
ARIN-prop-18: IPv6 Direct assignments to end sites · 2005-1: Provider Independent IPv6 Assignments for End-sites
Assessment: Mixed · The suggested minimum size changes materially from the /48 described in August 2005 to the /44 proposed in January 2006. The later message presents a revision informed by meeting feedback and adds qualification paths and a scaling idea. The evidence therefore supports an evolution of the design alongside continuity of purpose, rather than identical terms or an unexplained contradiction.
Conditions: These are different recorded proposals and dates; the messages do not prove adoption or a formal replacement relationship. · A larger proposed assignment is a change in design, not by itself a change in support for provider independence.
Viewpoint consistency: No explicit position stated
ISP's will have no choice but to have different minimum prefix size limits by range.
No guidance is given for how to utilize direct assignments other than you only get a PI /48 to work with.
These were my suggestions using feedback from the last meeting:
To qualify for a minimum end site assignment of /44 you must either:
I would gladly write up a formal revision or new proposal if requested.
