Summary

  • WSIS+20 was an intergovernmental review of implementation, digital development and Internet governance. Its final resolution reaffirms governments, business, civil society, international organisations, technical and academic communities, and other stakeholders in their respective roles.
  • The review strengthened cooperation, follow-up, the permanent IGF and UN policy pathways. It did not establish a UN number registry, amend ICANN's top-level number mission, replace RIR policy and service systems, or bind autonomous networks to one routing policy.
  • Number operations are disaggregated: IANA maintains top-level registries and allocations; RIRs and LIRs register and distribute resources; resource holders create route-origin authorizations; relying parties validate them; and operators decide which routes to accept and prefer.
  • Greater government voice is valuable for law, rights, competition, public investment, security and diplomacy. It becomes a category error when speech in a diplomatic process is presented as the executable act of a registry or operator.
  • The appropriate post-WSIS+20 design is a documented handoff: forum and diplomatic processes identify public problems, the institution controlling the relevant lever responds through its own procedure, and the public can trace implementation, refusal and remedy.

"Outside the room" describes authority, not attendance

Technical organisations and private actors were not literally absent from WSIS+20. The preparatory process received inputs from governments, companies, civil society, international organisations, and technical and academic communities. ICANN briefed diplomats and submitted text. RIR representatives and the Number Resource Organization participated in the wider WSIS and IGF ecosystem. The final resolution acknowledged these stakeholder groups.

The operational layer was outside the room in a different sense. The General Assembly hall did not contain the production systems through which number-resource changes occur. Delegates did not log into RIR administrative accounts, authenticate transfer parties, alter IANA registries, issue resource certificates or configure BGP policy on thousands of autonomous networks. The outcome document did not create a procedure for doing so.

This distinction matters because participation can be mistaken for institutional integration. A registry representative may explain operations in a consultation without placing the registry under the consultation's command. A government may deliver a forceful statement without acquiring the private credentials needed to update a record. An operator may support a principle without accepting a route filter.

The room shaped language, expectations and future cooperation. The operational layer remained linked to it through people, advocacy and institutional relationships, but separated by authority and implementation. To say this is not to defend secrecy or private immunity. It is to identify the point at which public argument must be converted into an accountable act.

The review was designed as a review, not a registry migration

The formal process began with a request to assess implementation of the outcomes of the 2003 and 2005 World Summit on the Information Society. The 2025 modalities resolution set a high-level General Assembly meeting for 16 and 17 December and invited stakeholder input. The process took stock of digital access, development, human rights, security, data, artificial intelligence, Internet governance and the WSIS framework.

The final Resolution 80/173 is correspondingly broad. It addresses connectivity and affordability, gender and other divides, enabling environments, finance, skills, media, sustainability, cybersecurity, rights, data governance, AI and follow-up across the UN system. Its Internet-governance section is one part of a twenty-page outcome.

A registry migration would require a different kind of document. It would identify the existing operator, the successor, the assets and data to move, the law and agreements that authorize the move, the transition date, continuity requirements, security controls, appeals and the consequences for resource holders. None appears.

The review did create and reinforce institutions within its own field. It made the IGF permanent, requested stronger reporting and set a 2035 review. It asked UNGIS to prepare a joint WSIS-Global Digital Compact implementation roadmap for consideration in 2026. These tasks show that the drafters could specify actors and dates when they intended an institutional act.

The absence of comparable number-registry provisions is therefore informative. WSIS+20 reviewed the environment in which number governance operates. It did not replace that governance's executable machinery.

The 2024-2026 sequence was diplomatic and programmatic

The review did not emerge from one December meeting. In September 2024, the General Assembly adopted the Pact for the Future and its Global Digital Compact annex. In April 2024, NETmundial+10 endorsed guidelines for multistakeholder collaboration, which the final WSIS+20 text later noted. During 2024 and 2025, CSTD, ITU, UNESCO, the IGF and other processes gathered evidence and proposals.

In March 2025, the General Assembly adopted the modalities for the high-level review. A CSTD twenty-year report followed in April. The IGF met in Norway in June and the WSIS+20 Forum High-Level Event met in Geneva in July. Draft outcome texts then moved through consultations and intergovernmental negotiation before the General Assembly adopted the final document on 17 December 2025.

This sequence produced political convergence. It preserved a broad multistakeholder formulation, recognized the technical community, strengthened the IGF and set implementation and reporting tasks. It also reflected advocacy by institutions protecting their roles. ICANN, the Internet Society, RIRs, governments and civil-society groups each entered the process with interests and preferred language.

In 2026, follow-up moved through UN reporting, the proposed WSIS-Global Digital Compact roadmap and the first permanent-status IGF cycle. Meanwhile, ICANN, the ASO and RIR communities continued their own number-policy and RIR-governance work. That parallel movement is the key institutional fact.

The diplomatic sequence and the operational sequence touched but did not merge. A chronology that places them beside each other should not be misread as a single chain of command.

The final text preserved "respective roles"

Paragraph 88 of Resolution 80/173 repeats the Tunis working definition: Internet governance is the development and application by governments, the private sector and civil society, in their respective roles, of shared principles, norms, rules, decision-making procedures and programmes that shape the Internet's evolution and use. Paragraph 91 expands the named community to include international organisations and technical and academic communities, again in accordance with respective roles and responsibilities.

The phrase is politically useful because it permits agreement among institutions that do not surrender authority to one another. Governments retain public responsibilities. Private companies operate networks and services. Technical bodies coordinate identifiers and standards. Civil society and academia supply scrutiny, evidence and public-interest claims. International organisations convene, coordinate and sometimes make law within their instruments.

The phrase is also dangerously vague if treated as an answer rather than a prompt. It does not tell a resource holder which registry can update a record. It does not identify who signs a resource certificate, who may approve an inter-RIR transfer or who changes a router configuration. "All stakeholders" cannot press one collective button.

The correct use of respective roles is functional. For every proposed action, identify the actor with the relevant legal or technical lever. Other stakeholders may inform, regulate, finance, challenge or review that actor. They do not become interchangeable.

WSIS+20 preserved that differentiated model. It called for cooperation among roles, not a merger of them into the General Assembly or the IGF.

A functional map reveals what diplomatic language hides

Internet governance becomes clearer when divided into acts rather than stakeholder categories.

Function Principal execution point Evidence of execution
Global principles and diplomatic commitments States and international institutions acting through their instruments Adopted resolutions, treaties, national measures and programme decisions
Top-level IP and ASN registry coordination ICANN's IANA functions under current policy and agreements IANA registry and allocation records
Regional distribution and registration RIRs, LIRs and other customers under regional policy and service relationships Approved allocations, assignments and registration records
IPv4 transfer recognition Source and recipient RIR processes, or one RIR for an in-region transfer Authenticated approval and updated registry data
Resource certification and ROAs Certification authorities and authorized resource holders Valid certificates and published signed entities
RPKI validation Relying-party software and validated caches Validated ROA payloads and cache state
Route acceptance and preference Autonomous network operators Router configuration, BGP policy and observed forwarding state
Peering and transit Networks and counterparties, subject to law and contract Interconnection, routing sessions and commercial agreements
Public regulation and remedy Governments, regulators and courts within jurisdiction Laws, orders, decisions and enforceable remedies

No row says that public policy is irrelevant to operations. Regulation can alter duties. A court can bind a registry or operator within jurisdiction. Diplomatic coordination can lead to later law or agreement. Public investment can change deployment. The map says only that the effective act must be located.

WSIS+20 mostly operated in the first row and in coordination among rows. Claims that it directly changed another row need a later instrument and evidence.

Number registration is an executable administrative act

Public IP addresses and autonomous system numbers must be unique to support global operation. RFC 7020 describes an Internet Numbers Registry System rooted in the IANA address allocation function, with IANA serving RIRs, RIRs serving LIRs and other customers, and LIRs serving resource users. Registration accuracy is one of the system's core goals.

The hierarchy is not simply an organisational chart. Each level performs transactions and maintains state. A registry authenticates an account, determines whether a request satisfies policy, allocates or assigns a resource, records an organisation and publishes data needed for coordination. It manages corrections, disputes and security. Some services connect registration to reverse DNS and RPKI.

A diplomatic outcome can recommend that the system become more inclusive or accountable. It cannot create a valid entry without reaching the authorised system. If a resolution and a registry record conflict about which organisation is recognized for a block, operators and relying parties still encounter the registry and certificate state used by current infrastructure.

This does not make the registry fact legally supreme in every dispute. A court may order correction. A contract may establish rights between parties. A public authority may impose obligations. But those legal acts need execution at the registry if the public record is to change reliably.

Number governance therefore has an unavoidable administrative core. WSIS+20 addressed principles around it. It did not acquire the accounts, databases and duty structure that constitute it.

IANA coordinates the top, not every downstream decision

The ICANN Bylaws define a mission that includes coordinating allocation and assignment at the top-most level of Internet Protocol and autonomous system numbers, providing registration services for global number registries, and facilitating global number policy with the affected community and RIRs. RFC 7020 similarly places IANA at the root of the number-allocation hierarchy.

This role is narrower than "control of all Internet numbers." IANA allocates large blocks and maintains top-level registries under global policies. RIRs apply regional policy and serve members and other customers. LIRs and operators make downstream assignments and routing choices. A top-level coordination role does not turn ICANN into the approver of every transfer or route.

The distinction is relevant to WSIS+20 because diplomatic debate often speaks of ICANN as if it were one control room for the Internet. That simplification makes institutional transfer sound easier than it is. Moving or influencing ICANN would not automatically configure networks, settle every regional policy or update every holder relationship.

Resolution 80/173 did not amend ICANN's Bylaws, the ASO Memorandum of Understanding or IANA's number service arrangements. It did not direct ICANN to accept IGF outputs as global number policies. Paragraph 93 instead calls for cooperation between international and intergovernmental organisations and other stakeholders.

Cooperation can be meaningful. It can expose diplomatic concerns to ICANN processes and technical limits to governments. It remains cooperation between distinct authorities.

Regional registries carry the middle of the chain

Five RIRs administer number resources across large service regions: AFRINIC, APNIC, ARIN, LACNIC and the RIPE NCC. They are not UN organs. They are nongovernmental institutions formed under the laws of the countries in which they are incorporated, with different membership, board and policy arrangements. Their communities develop regional number policies, while staff implement services and maintain records.

This structure gives the operational layer a private and transnational character. A decision affecting networks in many countries may be executed by an association or nonprofit incorporated in one country. Government agencies may be members, customers or entities, but do not hold a reserved global veto merely because they are governments.

Independent research published in 2026 found strong average legitimacy beliefs for APNIC and LACNIC among interviewed constituencies and more mixed views of AFRINIC. The result is a useful warning against two shortcuts. Nongovernmental status does not prove illegitimacy, and multistakeholder vocabulary does not prove equal trust.

During 2024-2026, the RIR system was also working on updated criteria for recognition, maintenance and derecognition of RIRs. That work addressed continuity, compliance and handoff within the ICANN-ASO-NRO relationship. Its existence shows that operational governance is not frozen.

WSIS+20 could influence the expectations applied to this work. It did not itself adopt the governance document, designate a successor RIR or move regional services to a UN office. Those actions require the number system's own procedures and, where relevant, law.

Private execution does not mean public consequences disappear

The fact that RIRs and operators execute key functions should not be converted into a doctrine of private sovereignty. Registry decisions affect market entry, continuity, security and the value associated with scarce IPv4 space. RPKI errors can affect route acceptance. Interconnection decisions can determine whether users reach services. These are public consequences.

Governments can regulate corporate conduct within jurisdiction, enforce competition and consumer law, protect rights, respond to national security risks, fund connectivity and provide courts. International cooperation can address cross-border conflict. Civil society, researchers and affected networks can scrutinize private institutions and demand reasons.

The important distinction is between legitimate public influence and false execution claims. A legislature can create a legal duty. A regulator can enforce it. A court can order a party. Those are real levers because law connects the public act to an obligated actor. A statement at a high-level meeting has a different status unless an applicable instrument gives it effect.

The strongest public strategy is therefore not to pretend that private infrastructure has become intergovernmental. It is to make private power visible, subject it to appropriate law and accountability, and define how public decisions reach systems without destabilizing continuity.

WSIS+20's broad language about rights, access, cooperation and enabling environments can support that agenda. It cannot complete the institutional design for every operational function in one outcome document.

Transfers expose the gap between policy endorsement and execution

IPv4 scarcity has made transfers a central part of number-resource governance. Organisations acquire or dispose of blocks through commercial transactions, mergers, restructurings and other arrangements. Registry recognition is necessary if the new holder is to control records and related services predictably.

Regional rules differ. APNIC processes in-region, inter-RIR and historical-resource transfers under stated conditions and records approved changes. RIPE NCC authenticates authorised signatories, applies transfer restrictions and coordinates compatible inter-RIR cases. ARIN applies its Number Resource Policy Manual and service procedures. Other regions have their own arrangements and constraints.

Suppose every government delegation at WSIS+20 supported faster transfers. The statement could create political pressure and inspire proposals. It would not authenticate a seller, establish a buyer's eligibility, clear a transfer lock or update two regional databases. Those acts require registry decisions.

Suppose instead that governments wanted stronger conservation rules. The same limit applies. A diplomatic preference must enter a competent regional or global policy process, or become law binding relevant parties, before it changes execution. Governmental voice is important evidence and may carry legal potential. It is not a substitute for the adoption step.

This gap explains why broad review texts often have limited immediate effect on transfer markets. They speak at the level of cooperation and principle while transactions depend on specific records, contracts, timelines and remedies.

RPKI divides authority even more finely

RPKI is often described as a registry security service, but its effective chain has several actors. Certification authorities issue resource certificates following the allocation hierarchy. Resource holders publish Route Origin Authorizations that identify an ASN permitted to originate a prefix. Repositories distribute signed entities. Relying-party software validates them. Routers receive validated payloads. Operators decide how validation states affect route selection.

RFC 6811 describes origin validation and says that considering invalid routes in the BGP decision process is a local policy matter. RFC 8210 specifies delivery of validated data from caches to routers. RIRs such as the RIPE NCC provide hosted and delegated certification services to eligible resource holders.

No diplomatic actor can replace all these steps with one declaration. A UN resolution cannot sign a ROA for a holder whose key it does not control. An RIR cannot force every network to apply one rejection policy. An operator cannot create valid resource authority merely by announcing a route. Each lever is bounded.

WSIS+20 can improve the politics around deployment. Governments can learn why procurement or regulation should account for route security. Smaller operators can describe capacity barriers. Registries can be pressed on outages, revocation and recovery. Researchers can expose measurement gaps.

The operational outcome still appears only when signed entities, validated data and operator configuration align. That is why participation in a governance review and authority over route security are different categories.

Routing remains a federation of local decisions

The global Internet is formed by autonomous networks exchanging reachability. Operators choose neighbours, negotiate peering and transit, set preferences, filter announcements and respond to incidents. Standards make the exchange interoperable, but they do not make one global routing administrator.

RFC 7020 expressly distinguishes registration from whether and how addresses are announced. An accurate registry can coexist with a route that is not advertised. A registered holder may authorize one origin while an operator still applies additional policy. Commercial and security considerations influence which path is preferred.

Governments affect this environment. They license telecommunications providers, regulate competition, procure connectivity, impose security duties and sometimes order shutdowns or filtering. WSIS+20 itself urges states to refrain from Internet shutdowns and protect open, interoperable, safe and secure access. These are meaningful public commitments.

Yet governmental influence remains mediated by jurisdiction and implementation. A minister's WSIS speech does not establish a BGP session. A consensus resolution does not choose LOCAL_PREF inside a private network. An IGF report does not purchase transit. The operator or bound service provider must act.

The separation is valuable for accountability. When a route policy causes harm, the responsible network should explain its configuration and constraints. When law compelled the action, the government should own the legal decision. Neither should hide behind the general language of global Internet governance.

More government voice is not the same as more operator control

Developing-country governments have long argued that unequal resources limit their influence in global Internet governance. Resolution 80/173 recognizes barriers to participation and asks for stronger government and stakeholder engagement. This can improve policy because states understand public services, development priorities, rights duties and local market conditions that technical institutions may overlook.

The gain should be measured at the correct level. Better participation may change agenda priorities, expose regional harm, improve national capacity and produce more legitimate public policy. It may help governments enter ICANN, RIR and standards processes with clearer positions. It may support cooperation when cross-border problems exceed one jurisdiction.

It does not follow that governments should collectively operate private networks or number registries from the IGF. Voice and execution solve different problems. Voice affects whose evidence and interests enter a decision. Execution affects who can make the decision effective. A legitimate system needs both, connected by transparent procedure.

Conflating them can harm the governments whose inclusion is sought. If an official believes that a UN statement changed a registry rule, the state may neglect the regional proceeding where the rule is actually decided. If operators assume diplomacy is technically uninformed, they may withdraw rather than engage. The result is more theatre and less influence.

The practical objective after WSIS+20 is not to exchange government voice for operator credentials. It is to ensure that voice reaches the institution controlling the credential and receives a reasoned response.

The technical community's seat is not a veto or a surrender

ICANN and technical organisations campaigned during WSIS+20 for recognition of a distinct technical community and preservation of multistakeholder governance. Their concern was that intergovernmental language could marginalize institutions that maintain the Internet's unique identifiers and standards. The final resolution names technical and academic communities and preserves respective roles.

This recognition does not grant technical actors immunity from law or public criticism. Expertise is not a popular mandate. A registry may understand its systems while making a poor governance decision. An operator may be technically competent while imposing unfair contract terms. A standards process may exclude affected communities.

Nor does participation mean surrender. Technical actors who brief diplomats or contribute to a draft do not transfer their systems to the UN. Their statements are stakeholder advocacy and technical evidence. The General Assembly decides its own outcome; the technical institutions retain the authorities established by their constituting documents and operating relationships unless separately changed.

The balanced reading rejects both extremes. WSIS+20 did not hand the Internet to engineers, and it did not hand the registries to diplomats. It reaffirmed a cooperation model in which institutions remain differentiated.

The hard work begins after recognition. Technical bodies must show how public concerns can enter their procedures. Governments must state which lawful instruments they use. Both must avoid invoking "the multistakeholder model" as a substitute for reasons.

The IGF became a stronger bridge, not an operations centre

The most significant institutional change in the final outcome was permanent status for the IGF. The resolution also strengthened reporting, intersessional activity and participation. This creates a more durable bridge between diplomatic, social, commercial and technical communities.

A bridge can reduce the distance between a public problem and the institution able to act. A government concerned about address access can hear how regional policy works. A small operator can challenge an RPKI service design. A registry can explain why a proposed rule would damage uniqueness or routing. Researchers can compare claims with measurements.

The bridge metaphor preserves an essential fact: crossing is required. An IGF conclusion about a regional policy must cross into the RIR's proposal and adoption process. A route-security recommendation must cross into resource-holder and operator action. A rights concern may cross into legislation, litigation, contract reform or technical design.

If no one crosses, the forum becomes a waiting room. If organisers claim that discussion itself operated the system, it becomes misleading. If an operational institution cites participation but never answers, the bridge is decorative.

Permanent status gives the IGF repeated opportunities to test the handoff. It should publish who received mature recommendations, what response was given and whether evidence changed. That is a stronger contribution than claiming powers the forum does not have.

WSIS action lines operate through named implementers

The WSIS framework contains action lines covering infrastructure, access to information, capacity, confidence and security, enabling environments, applications, cultural and linguistic diversity, media, ethics and international cooperation. Resolution 80/173 calls them the main framework for translating the WSIS vision into outcomes and asks facilitators to develop targeted roadmaps and indicators.

This language is more implementation-oriented than a generic declaration. It still operates through named facilitators, UN entities, governments, development institutions and stakeholders within existing mandates. Paragraph 111 expressly refers to existing mandates and resources. Paragraphs 113 and 122 request roadmaps for later consideration and reporting.

Number registries interact with several action lines. Accurate registration and secure routing support infrastructure and confidence. Allocation policy affects access and development. Yet the framework does not turn an action-line facilitator into an RIR. The facilitator can convene, coordinate, measure and recommend; the registry or operator must execute the number-specific act.

This layered implementation is not necessarily a defect. A global development framework should not casually administer every technical database. Its value lies in making cross-cutting priorities visible and directing responsible institutions to account for them.

The weakness appears when responsibility becomes too diffuse. If every stakeholder is asked to act, no one may be answerable. Follow-up should therefore map each commitment to a concrete decision point rather than assuming the broad framework is self-executing.

Consultation evidence cannot prove a change in control

The WSIS+20 process included stakeholder consultations, surveys, written inputs, side events and a sounding board. These mechanisms can improve a diplomatic text by introducing expertise and affected perspectives. They also generate a large record that institutions may later cite as evidence of inclusion.

Inclusion in drafting does not prove a transfer of control. A registry may propose language recognizing the technical community. A government may accept it. The resulting paragraph demonstrates agreement on recognition, not a change to administrative credentials. A civil-society intervention may shape rights language without becoming an enforcement order.

Consultation also does not establish representative consent. Entities enter through different routes. Some speak for states, some for employers, some for member organisations and some for themselves. Time, language, accreditation and travel affect who appears. A consultation can be open while remaining socially unequal.

The public record should therefore be used for what it can prove: a view was submitted, a concern recurred, language changed, or an institution participated. Claims that the process represented every operator or user require additional evidence. Claims that it altered operational authority require an operative instrument.

This standard does not devalue consultation. It prevents the process from being credited with an act it was not designed to perform.

Operational separation can conceal accountability gaps

The phrase "the operational layer is separate" can be abused by the institutions that control it. A registry may say a dispute is technical when it concerns fairness or legal rights. An operator may treat routing policy as confidential even when a decision disrupts public services. A private organisation may invoke community process while repeat entities dominate it.

Separation should therefore trigger scrutiny, not deference. Who can initiate a policy change? Who authenticates and decides an individual request? Are reasons published? Is there an independent appeal? How are conflicts disclosed? What happens during institutional failure? Which court has jurisdiction? Can a smaller network participate without sustained funding?

The 2024-2026 work on RIR governance criteria reflects some of these concerns. Recognition, ongoing compliance, remediation, derecognition and continuity all need rules. The process itself must be assessed for whose interests and legal risks it includes.

WSIS+20 provides a venue and political vocabulary for asking these questions. Governments can demand that private power be accountable. Technical bodies can explain continuity constraints. Civil society and operators can expose consequences.

What WSIS+20 cannot do is make the accountability gap disappear by declaring the operation multistakeholder. The institutions holding the levers must supply the remedy, or an applicable public authority must lawfully require one.

Governmental authority should be traced through law

Governments possess powers that private technical bodies do not. They can enact law, levy tax, fund infrastructure, protect competition, regulate licensed providers, investigate crime and provide courts. At the international level, they can negotiate treaties and create organisations. These powers can reach the operational layer.

The reach is not automatic or unlimited. Jurisdiction, constitutional rights, administrative procedure, international obligations and technical feasibility matter. An order directed at a local operator differs from an attempt to rewrite a global registry. A court binding one corporation may not bind every RIR. A national policy can affect routes observed globally without becoming global law.

WSIS+20 increases the visibility and coordination of government positions. It does not replace the instrument through which a position becomes enforceable. A reader should be able to trace a chain from diplomatic statement to legislation, regulation, order or contract, then to the obligated operator or registry, and finally to the system change.

This chain protects democratic accountability. Officials cannot attribute coercive action to an undefined global consensus. Private actors cannot dismiss a lawful duty as mere forum speech. Courts can review the actual instrument.

Government voice is strongest when it states the authority it is using and the remedy available. It is weakest when it borrows the language of global governance to conceal an unexamined assertion of control.

A workable handoff needs six recorded elements

WSIS+20 and the permanent IGF can have greater operational impact without becoming operators. The mechanism is a disciplined handoff.

First, define the problem in operational terms. "Improve routing security" is too broad; identify the affected prefixes, validation gap, service barrier or policy defect. Second, identify the institution controlling the relevant lever. It may be a resource holder, RIR, ICANN process, standards body, operator, regulator or court.

Third, state the requested act. Ask for a policy proposal, record correction, service improvement, deployment commitment, legal review or publication of evidence. Fourth, identify the recipient procedure and deadline. A recommendation with no entry point is an aspiration.

Fifth, preserve evidence and disagreement. The receiving institution needs to know what supports the request and who may be harmed. Sixth, publish the response and review path. Acceptance, modification, rejection and silence should all be visible.

This structure respects institutional competence without insulating it. A registry cannot say the forum lacks authority and stop there; it should answer a properly directed concern under its own accountable process. A forum cannot say the issue was discussed and claim completion; it should follow the handoff.

The result is practical multistakeholderism: not everyone executing every function, but affected actors able to influence and review the institution that does.

Metrics should follow the lever

The success of post-WSIS+20 cooperation should not be measured by the number of declarations using the same vocabulary. It should be measured by changes at identifiable decision points.

For number registration, measure accuracy, correction time, authentication failures, service continuity and reasoned decisions. For transfers, measure processing distributions, refusal categories, appeal outcomes, inter-RIR coordination and record cutover errors. For RPKI, measure certificate availability, ROA coverage, publication incidents, recovery and operator adoption without overstating what origin validation proves.

For routing, measure observed invalid acceptance, leak containment, resilience and the effects of local policy. For participation, measure whether developing-country governments and smaller operators can initiate proposals, draft text and obtain responses, not merely attend meetings. For diplomatic follow-up, trace recommendations into law, budgets, institutional policies and documented refusals.

Metrics also need boundaries. Registration accuracy is not proof of legal ownership. ROA coverage is not complete route security. Attendance is not consent. A published roadmap is not implementation. A government statement is not an operator configuration.

This discipline prevents one layer from claiming another layer's achievement. It also shows where failures actually sit. If a mature recommendation receives no registry response, the registry's accountability is at issue. If a policy exists but operators do not deploy it, implementation is elsewhere.

The operational layer becomes governable when its acts are visible, not when its name is absorbed into diplomatic prose.

A number dispute demonstrates the translation gap

Consider a cross-border operator whose acquisition includes an IPv4 block registered in another region. The buyer has a commercial agreement, a network deployment deadline and government support for expanding connectivity. The source and recipient records are inconsistent, one authorised signatory has changed, and the two relevant RIRs apply different eligibility and documentation rules.

WSIS+20 principles can illuminate the stakes. Meaningful connectivity may depend on the capacity. Transparent and predictable enabling environments matter. Cooperation across institutions is preferable to fragmentation. Developing-country participation may reveal that the operator faces costs not visible to the registry communities that wrote the rules.

None of those principles completes the transfer. The parties must establish authority. Each registry must apply the policy that governs its side, exchange the necessary confirmations and update its data in a coordinated sequence. RPKI and reverse-DNS authority may need a planned cutover. If a registry refuses, it should give reasons and identify review. If a court order governs a party, the order must reach an entity it lawfully binds.

The example shows why diplomacy and operations need translation rather than substitution. A broad outcome identifies values and creates political attention. The affected operator supplies transaction evidence. Registry institutions convert applicable rules into an administrative decision. Technical services implement the changed state. Operators then announce and accept routes under their own policies.

A credible follow-up record would preserve every link: which WSIS principle was invoked, which registry procedure received the concern, which rule controlled, what decision was made, how the records changed and what remedy remained. If the transaction succeeds, no single layer should claim the whole result. If it fails, the record should reveal whether the obstacle was law, policy, evidence, registry administration or network execution.

This is the practical meaning of governance across respective roles. The public problem can be shared while the acts remain differentiated and accountable.

What the 2035 review should be able to see

Resolution 80/173 schedules another high-level review for 2035. Ten years is long enough for institutional memory to become selective. The next review should not rely on retrospective success narratives from institutions evaluating themselves.

It should have a public map of which WSIS+20 commitments entered which procedures. For Internet numbers, that means a record of proposals received by ICANN and RIR communities, registry-service reforms, continuity tests, RPKI incidents, transfer-policy changes and operator deployment. Government actions should be linked to the laws and orders that implemented diplomatic commitments.

The record should also include rejection and conflict. A technical body may reasonably reject a politically attractive proposal because it threatens uniqueness or stability. A government may reject private self-regulation because remedies are inadequate. These disagreements are evidence, not noise to remove from a consensus summary.

By 2035, a reviewer should be able to distinguish three outcomes: discussion that improved understanding, recommendations that changed a competent institution, and declarations that never reached an execution point. That distinction will reveal whether the separation between rooms became a working bridge or an excuse for inaction.

The permanent IGF can help maintain this memory. It should not rewrite the record to make itself the author of every downstream change.

Conclusion: voice needs a route to execution

WSIS+20 achieved a significant diplomatic settlement. It reaffirmed global and multistakeholder Internet governance, recognized differentiated roles, strengthened follow-up and gave the IGF permanent status. It addressed development, rights, security and new technology in a framework broad enough to sustain international cooperation.

The settlement did not place the Internet's number operations inside the General Assembly. IANA still coordinates the top-level registries. RIRs and LIRs still distribute and record resources. Regional procedures still recognize transfers. Resource holders and certification authorities still create RPKI objects. Relying parties validate them. Autonomous networks still choose routing and interconnection policy. Governments reach these actors through law, regulation, procurement, diplomacy and participation - not through an undefined conversion of voice into credentials.

This distribution has weaknesses. Private institutions can be opaque or captured. Regional arrangements can produce unequal rights. Operators can externalize harm. Technical language can hide policy choices. Acknowledging where execution occurs makes these problems more, not less, open to challenge.

The central post-WSIS+20 task is to connect public voice to the correct lever. Every mature recommendation should name a recipient, procedure, requested act, evidence, deadline and remedy. Every operational institution should explain what it accepted, rejected or changed. Every government should identify the lawful instrument behind coercive action.

Diplomacy can define values and create pressure. Forums can discover problems and build relationships. Neither should claim that discussion updated a registry or configured a route. The operational layer was left outside the room because it resides in a distributed set of institutions and systems. Good governance does not erase that fact. It builds an accountable path across it.

Sources