Hi Nick, Angela, and Colleagues,
Thank you for this prompt review, Nick, and for highlighting the exact structural mechanism that makes this framework effective.
Nick suggests that enforcing contractual resource accountability represents a "disturbing" change in strategic direction for the RIPE NCC. What is genuinely disturbing is the operational reality playing out across our region this week. As the severe infrastructure attacks against Albert Heijn stores here in the Netherlands and major retail distribution networks in France demonstrate, infrastructure-level cybercrime has graduated from basic data theft to targeting the physical food supply chain.

These operations rely entirely on the systematic exploitation of un-remediated, bulletproof hosting allocations. Pretending the RIPE registry must remain a passive, blind directory while its resources are weaponised to sabotage regional food supplies is a luxury the market can no longer afford.

There was also the case dropped onto the Security WG list recently by Theo Plooij regarding 91.92.243.0/24 (Ticket 1112396). It exhibits the exact systemic loophole this policy aims to close: an active phishing block, blacklisted by Spamhaus, operating under an uncontracted offshore shell company via a Turkish LIR. Because the unmonitored abuse mailbox technically doesn't bounce emails, it is deemed 'valid' under ripe-705, leaving both the LIR and the RIPE NCC stating that they have no policy mandate to intervene. So nobody acts.

Michele raises an excellent technical point regarding standard client-side email read-receipts being routinely blocked or disabled in 2026. This exact failure of standard email communication is precisely why Article 3.3 bypasses traditional unmonitored abuse mailboxes entirely.

By mandating "electronic delivery with automated tracking" to the authoritative legal and billing contacts under the signed Standard Service Agreement (SSA), the policy directs the RIPE NCC to utilise contractually binding legal delivery mechanisms (such as secure portal-log timestamps or certified server-side delivery networks) where delivery verification is handled at the infrastructure layer, making it impossible for a non-compliant operator to claim ignorance.
Your concern regarding the high-level operational nature of Article 3.5 is completely valid, and it is precisely why this framework utilises a split-document blueprint. To satisfy the Working Group Co-Chairs' explicit formatting directive to keep the primary policy text entirely unencumbered by technical software dependencies, the granular engineering pathways have been separated from the core template. This keeps the text of the policy elegant and future-proof.
However, as explicitly noted in the Rationale section you quoted, a high-level policy mandate does not mean an undefined implementation. The logic of these restrictions must be discussed in parallel to ensure complete operational predictability.
In direct response to your request to see exactly what is intended by "strong, well-defined restrictions," here is the full draft text of the parallel RIPE Network Abuse Policy Support Document.
It details the explicit, community-governed parameters engineered to restrict operational registry agility safely across Hosted RPKI, Delegated RPKI, and non-RPKI configurations — additionally utilising Enforced De-aggregation and signed geofeed: ZZ metadata mapping parameters to isolate threats while aiming to protect adjacent innocent downstream assignments from collateral disruption.
Per the published Rationale, I look forward to discussing these technical parameters alongside the main policy articles.

RIPE Network Abuse Policy Support Document
ripe-nnn     Publication date: nnnn 2026    State: Draft (Proposed Technical Baseline for Parallel Discussion)
Authors
Denis Walker
1. Governance, Authority, and Maintenance Framework
1.1 Operational Mandate and Scope
This document serves as the authoritative operational technical logic for the execution of proportionate intermediate administrative restrictions authorised under "ripe-nnn Contractual Accountability Framework for Infrastructure-Level Network Abuse" (currently Policy Proposal 2026-02). The programmatic enforcement pathways detailed herein — including Hosted RPKI portal management suspensions, Delegated RPKI CA certificate adjustments utilising Enforced De-aggregation where necessary, and non-RPKI RIPE Database object overrides — constitute the community-approved parameters for the execution of these restrictions.
1.2 Working Group Consensus Control
This operational technical logic document is a community-governed RIPE document maintained natively along with the registry of Approved Infrastructure Blocklists. Any future structural updates, technical pathway additions, algorithmic shifts, or modifications to the Approved Infrastructure Blocklists contained within this document shall be subject to the formal consensus rules of the Security Working Group.
1.3 Implementation Alignment
The RIPE NCC shall schedule an update in a timely manner to its backend software and internal support interfaces to align with the revised document state once it is agreed by the community.
2. Registry Of Approved Infrastructure Blocklists (Active Restriction Triggers)
To be included in this registry, a blocklist must rely on objective automated telemetry, maintain a publicly verifiable historical archive, and explicitly filter out or exclude residential/dynamic subscriber pools (e.g., utilising internal alignment with metrics parallel to the Spamhaus PBL framework).
  1. Spamhaus DROP (Don't Route Or Peer) List
  2. Spamhaus XBL (Exploit Block List)
  3. [Placeholder for additional WG-approved telemetry feeds]
3. Logic for Applied Restrictions
There are many permutations for an RPKI (or not) setup and RIPE Database object structures. For the policy to be effective, each of these permutations may need a separate enforced action flow. This results in a complex set of options. In practice, for any specific case, only one option is followed. The complexity should be handled within an automated system behind a simple user interface for the support staff. The appropriate option will be determined programmatically. The RIPE NCC support staff will not be required to work out the right permutation.
Where an "abuse-status: NON-COMPLIANT" attribute is added in the RIPE Database, this will be to the resource holder's ORGANISATION object as well as the appropriate resource object. The "abuse-status:" attribute can only be removed by the RIPE NCC.
3.1 The identified operational restrictions options
4. Document History

Cheers,
Denis
Retired RIPE Database Senior Analyst & Engineer 

On Tue, 13 Oct 2026 at 15:37, Nick Hilliard <nick@foobar.org> wrote:
Angela Dall'Ara wrote on 13/10/2026 12:08:
> This proposal suggests a contractual mechanism to hold resource holders
> accountable for systematically failing to address infrastructure-level
> network abuse.

 From a brief reading, my take-away is that this proposal has a number
of fundamental structural issues, not least an almost complete lack of
clarity about what the policy is going to mean in practice, a suggestion
that the RIPE NCC ought to execute a substantial change in strategic
organisational direction from being a registry to using registry data to
impose business consequences on RIPE NCC members, and not least, an
unstated implication that by outsourcing some of the input process
decision-making to third parties, that the RIPE NCC could somehow
disclaim responsibility for operational outcomes of the policy. These
are all disturbing problems, any one of which should be sufficient to
justify rejecting the proposal in their own right.

The Rationale section includes the following:

> Operational Dependency: This policy will only be effective with
> strong, well-defined restrictions agreed by the community and
> codified into a separate RIPE document. The logic of those
> restrictions needs to be discussed in parallel with the policy to
> prevent administrative ambiguity or unworkable deployment workflows.

A good starting point would be to understand exactly what the author has
in mind for "strong, well-defined restrictions". Can this proposal
document be published first, so that the working group can at least have
some idea what the author is asking the community agree to?

Nick
-----
To unsubscribe from this mailing list or change your subscription options, please visit: https://mailman.ripe.net/mailman3/lists/security-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/