Representation by Economy Is Not Representation by Network

Summary

• Counts of countries or economies represented at a meeting are useful measures of geographic contact. They do not establish how many independent organisations attended, which network operators carried the consequences, who held voting authority, or whose interventions changed a decision.
• A credible account needs separate measures for economies, unique organisations, Autonomous System Numbers and address holdings, operator classes, market concentration, formal decision authority and speaking influence. Each answers a different question, and none should silently substitute for another.
• Resource holdings and market size belong in an impact audit, not in a formula that buys votes. Small economies and small networks may face distinctive risks that aggregate traffic or customer numbers conceal, while large exposed networks should not disappear behind a diverse set of flags.
• LACNIC can publish an economy-to-network representation matrix, report concentrated and unknown affiliations, and attach an impact audit to consequential decisions. The purpose is to describe blind spots and seek missing evidence, not to declare any entity more entitled to speak.

The flag is evidence, but evidence of what?

The familiar line that entities came from a given number of countries or economies performs an honest and useful function. It shows that an event crossed borders. It can reveal whether travel support, remote access, interpretation and local hosting reached beyond the organiser's home base. Across repeated meetings, it can expose geographic stagnation or expansion. ICANN's meeting reports publish regional and attendance material for precisely this kind of institutional learning. A recent ICANN meeting retrospective, for example, stated that ICANN79 entities represented 108 countries and territories, alongside separate in-person and virtual totals. The distinction already hints that one headline cannot carry every meaning.

The problem begins when geographic reach is allowed to imply representative authority. An economy field normally describes a entity's declared location, nationality, residence, employer location or another registration choice. Even when the field is accurate, it does not say whether the entity operates a network, advises a ministry, sells equipment, studies policy, represents customers, speaks only for herself, or carries instructions from an organisation entitled to vote.

Nor does one person from an economy establish that its broadband market, public networks, universities, community networks, content providers or future applicants were heard.

Economy count is therefore a contact metric. It answers whether at least one observed entity was associated with each listed place. That is worth publishing. It should be described in that limited way. When an institution moves from “people from many economies attended” to “the region was represented,” it has crossed from observation to an unproven conclusion. The cure is not to abandon geography. It is to place geography beside the other dimensions required to understand who bore risk, who held authority and who shaped the outcome.

One economy can contain many incompatible network interests

An economy is not a network constituency. Within a single jurisdiction, the interests of a dominant mobile carrier, a rural wireless provider, a university, a government network, an Internet exchange, a bank, a hosting company and a new entrant can diverge sharply. They may all depend on Internet number resources, yet encounter policy through different operational surfaces. A transfer rule may matter to an address broker and an expanding provider in opposite ways. A routing-security requirement can impose different costs on a large engineering team and a small organisation with outsourced routing.

An election rule can concern a member's legal contact even when its engineers never attend the meeting.

Geographic aggregation erases those differences. If five people from one economy attend, the headline gives no indication whether they represent five independent networks, one corporate group, several institutions with no routing operations, or themselves. If one person attends from another economy, the count gives that place the same geographic presence while revealing nothing about the breadth of interests behind either observation. Equal treatment in the geography column is appropriate because the column measures reach. It becomes misleading only when readers are invited to infer equality of network exposure or organisational independence.

This matters in Latin America and the Caribbean because economic size, geography, connectivity patterns and institutional capacity vary greatly across the service region. A small island economy may have few visible networks but severe dependence on a limited set of international links. A large market may contain hundreds or thousands of operationally distinct networks, layered wholesale relationships and concentrated customer access. Neither deserves to be reduced to a flag.

The small economy's presence can reveal a risk invisible in regional totals; the large economy's single flag cannot demonstrate that its varied network classes participated. Respect for both begins by refusing to make geography answer questions it was never designed to answer.

Unique organisations are the first missing denominator

The next line beneath an economy count should be the number of unique organisations represented. People are not interchangeable with organisations, and badge totals are not independent institutional voices. A company may send public-policy staff, engineers, executives and counsel. A registry, sponsor or host may have many employees present. A consultant may list a firm while working for several clients. A person can hold an employer role, a committee seat and a community title at once.

Without careful organisational resolution, an event can appear plural because it contains many individuals while a small number of employers supply much of the repeated participation.

LACNIC's own prior-event entity description distinguishes average attendants from represented organisations and then describes categories including Internet service, security, cloud, data-centre, telecommunications and content providers, exchanges, government institutions, academia and civil society. That separation is analytically stronger than a country total alone. It recognises that headcount, organisation and activity type describe different features of an audience.

Publishing unique organisations requires a stated method. Subsidiaries under common control should be visible both as local operating entities and as a corporate group. Trade associations should not be merged with their members, but their relationship should be disclosed. Universities with autonomous network operations should not be automatically collapsed into a ministry. Consultants need a way to declare the interests they are presenting without exposing confidential engagements. Self-employed and unaffiliated entities remain legitimate entities rather than errors to be discarded.

The report should show a raw organisation count, a control-group count, and the share whose affiliation could not be resolved. It should never pretend that uncertain names have been definitively matched. The resulting denominator will still be imperfect, but it tells readers whether apparent diversity comes from independent institutions or from multiple badges attached to the same centres of control.

ASNs and address holdings describe exposure, not entitlement

Internet number-resource governance has an additional observable surface: Autonomous System Numbers and delegated address space. The NRO's RIR statistics page explains that the joint delegated-extended files report distributions of IPv4 ranges, IPv6 ranges and ASNs, with daily updates and documentation for interpretation. These records can help an institution ask whether organisations with direct resource responsibilities were present, absent or concentrated in a discussion whose consequences fall on routing and addressing operations.

The appropriate unit is not simply “number of ASNs in the room.” An organisation may control several ASNs for sound operational reasons. One ASN may serve an enormous customer base; another may support a small research or community network. Address holdings can reflect historical allocation, transfers, business structure and technical design rather than current traffic, revenue or social importance. IPv6 and IPv4 measurements are not interchangeable. Published delegation records also do not reveal every downstream dependency, customer, reseller or network that uses resources under another organisation's account.

Used carefully, resource data answers an exposure question: what portion and variety of directly registered network operations had an identifiable path into the discussion? It can reveal, for example, that many economies were touched while resource-holding organisations were concentrated among a few markets or operator classes. It can also reveal the reverse: a technically concentrated session may include operators responsible for substantial parts of the region's routed infrastructure while omitting governments, users, entrants or small economies with distinctive policy concerns.

What resource data must never do is create a property-weighted franchise. A larger address holding should not purchase a louder voice, and multiple ASNs should not become multiple civic identities. Holdings belong in the evidence used to anticipate consequences and seek missing operational testimony. Formal voting rights must remain governed by published membership rules, while open policy participation must remain open. Measurement should improve the institution's view of affected networks, not transform public identifiers into political shares.

Operator class reveals consequences hidden by totals

A representation report needs an operator-class column because organisations with similar geography and even similar resource holdings can experience policy very differently. Useful high-level classes include access provider, mobile operator, fixed carrier, community network, enterprise network, educational or research network, government network, content or cloud provider, data centre, Internet exchange, domain or registry organisation, vendor, consultancy, civil society body and unaffiliated individual.

The categories should be published, stable enough for comparison and flexible enough to allow a entity to select more than one relevant role.

Classification must follow the capacity in which a entity contributes, not merely the broad business description of an employer. An equipment vendor's engineer may offer valuable routing evidence without claiming to operate a customer access network. A university may run an autonomous network and also contribute academic research. A government official may be responsible for a public-sector network or may attend solely in a regulatory capacity. A civil society entity may bring evidence from users who have no direct registry relationship. These roles should be represented honestly rather than forced into a single prestige hierarchy.

The class distribution exposes blind spots that an economy total cannot. A meeting might span most of the region while access providers dominate the microphone. Another might include a balanced mix of institution types but lack the small operators that face the highest per-person compliance burden. A third might draw many technical experts while decision authority resides with membership contacts who do not attend the policy sessions. The proper response is targeted outreach, not exclusion.

If a class is missing, the organiser can seek written evidence, commission a structured consultation, arrange remote testimony or delay a claim of broad support.

No class should be treated as the authentic voice of an economy. Networks, states, companies and users have different legitimate interests. Operator-class reporting simply prevents a broad list of places from concealing a narrow list of functions.

Market concentration changes the meaning of presence

Organisation counts treat independent institutions as units, which is necessary but incomplete. In a concentrated access market, one or two firms may serve a large share of customers. In a fragmented hosting or enterprise market, many organisations may each carry a small portion of the observable activity. A meeting report that lists ten operators without context cannot show whether it heard from marginal entrants, dominant providers or a balanced range. Conversely, a room containing a dominant carrier does not thereby represent all its customers, competitors, resellers or interconnecting networks.

Market concentration belongs in an impact layer built from authoritative, jurisdiction-specific evidence. Telecom regulators may publish subscriber or revenue shares. Competition authorities may describe market structure. Operators may report customers, coverage or traffic indicators, though each measure has limits. Routing data can illuminate reachability relationships but cannot reliably stand in for retail market share.

Where no comparable official evidence exists, the report should mark concentration as unknown and specify what would have to be collected: current provider shares by service, corporate-control links, wholesale dependencies, or verified customer counts.

The analytical question is not whether large companies deserve more votes. They do not. It is whether a decision likely to affect a substantial customer or routing surface was made without evidence from the organisations carrying that exposure, or with evidence supplied almost entirely by those organisations and too little from smaller competitors and dependent users. Both patterns can produce risk. Absence of a major operator may leave implementation consequences unexplored. Dominance by major operators may suppress costs borne by entrants, community networks or customers with few alternatives.

A concentration indicator should therefore be paired with pluralism indicators. Report the share of represented organisations accounted for by the largest control groups, the distribution across operator classes, and the known market exposure of entities where reliable. Then state the unknowns. The objective is a better map of consequence, not a corporate census that converts economic power into constitutional authority.

Decision authority is not visible on the attendance list

Attendance and authority are separate institutional facts. A public policy forum may be open to any interested entity, while a member assembly exercises powers reserved to members. A chair may assess consensus in one setting; the Board may ratify a proposal later; a registered membership contact may vote in an election; staff may implement an adopted rule. Reporting that combines these people into a single entity total blurs who could speak, who could decide and who carried the organisation's formal instruction.

LACNIC's Policy Development Process describes an open public list and forum, consensus functions and a later Board role. Its membership FAQ explains that membership includes voting rights according to category and assigned votes, and that the membership contact exercises those rights for the organisation. The LACNIC Bylaws on assemblies separately identify the General Assembly as a governing body and establish notice, agenda, quorum and voting provisions. These are not variations of the same generic attendance relationship. They are distinct chains of authority.

A representation audit should mark which observed people were ordinary entities, proposal authors, chairs, staff, Board members, candidates, eligible organisational voters, designated voting contacts, proxies where permitted, and actual voters. The categories can overlap, but the overlap must remain visible. A person from an underrepresented economy who attends a technical session may improve geographic reach without gaining any election authority. A membership contact who votes remotely may exercise formal authority without appearing in the conference hall.

A Board member may speak little in public while retaining decisive institutional power later.

This separation protects open participation. It avoids dismissing non-voters as irrelevant, and it avoids presenting their presence as consent to decisions they could not formally make. Legitimacy becomes traceable: access is evidenced by participation, authority by rules, and acceptance by the specific acts through which each body reaches a conclusion.

Speaking influence is a further, observable layer

Even a room with varied economies, organisations and operator classes can produce a narrow record if a small group supplies most interventions, receives most chair responses or frames the text that survives. Speaking influence is not identical to speaking time, but the public record provides observable signals: number and duration of interventions, agenda placement, proposal authorship, requested clarifications, chair summaries, accepted edits, action assignments and references in final decisions. Those signals can be reported without pretending to read minds.

The first task is to distinguish access to the microphone from influence on the outcome. A entity may speak often and fail to change the proposal. Another may offer one operational example that causes a revision. A chair's synthesis can amplify, narrow or omit an intervention. Written list contributions may matter more than remarks in the hall. Informal preparation can establish the range of acceptable options before the recorded session begins. A complete account therefore connects public speech to subsequent text while acknowledging unrecorded influence as a limit.

Basic concentration measures are still valuable. Reports can state what share of interventions came from the most active speakers and organisations; how many economies and operator classes spoke, not merely attended; how much affiliation remained unknown; and whether authors, chairs or staff dominated responses. They can code whether a contribution introduced evidence, supported a position, objected, requested information or proposed language. The point is not to grade eloquence. It is to test whether geographic diversity reached the deliberative record.

Publishing this layer also disciplines celebratory claims. If forty economies appear on registration but entities from six organisations produce most substantive interventions, the institution can truthfully celebrate reach while acknowledging concentrated influence. That candour creates a concrete next step: ask why attendance did not become voice, and whether language, scheduling, expertise, hierarchy or agenda design prevented conversion.

The economy-to-network representation matrix

The central reporting instrument should be an economy-to-network representation matrix. Each row represents an economy observed in the event or relevant to the decision. The columns describe separate dimensions: registered individuals, checked-in individuals, active entities, unique organisations, control groups, identifiable resource-holding organisations, ASNs associated with represented organisations, IPv4 and IPv6 exposure bands, operator classes, known market-concentration bands, eligible member voters, actual voters, substantive speakers and accepted contributions.

Every cell carries a coverage status: observed, estimated, unavailable, not applicable or withheld for privacy.

The public version should use bands and aggregates where small counts could identify people. The internal analytical copy can preserve the evidence needed to reproduce totals under access controls appropriate to personal and corporate information. Definitions must travel with the table. “Active entity,” for example, might mean a person who spoke, posted to the relevant list, submitted text, voted in an open poll or held a defined session role during the reporting period. It must not mean merely opening an event page.

Rows should not be ranked from best represented to worst. The matrix is diagnostic, not a medal table. A small economy with one entity and one critical infrastructure operator may have strong evidence for one issue and none for another. A large economy with many badges may show weak organisational independence. An economy with no visible attendee may still be affected through cross-border services or a corporate parent. Regional and global organisations may require a separate row so they are not assigned arbitrarily to headquarters locations.

The matrix allows precise language. Instead of saying “thirty economies were represented,” an organiser might say that thirty economies had at least one checked-in entity, twenty-two had an active contributor, sixteen had an identifiable network operator, and affiliation coverage was limited public evidence to assess the rest. The exact figures would come from collected evidence, not assumption. The important innovation is grammatical: every representation claim names the unit and the activity that justify it.

Unknown affiliation is a result, not an inconvenience

Affiliation data will never be complete. People change jobs, use personal addresses, advise clients, participate in several capacities, or reasonably decline to disclose sensitive relationships. Corporate structures can be opaque. Some networks operate under brands different from legal entities. Event registrations may contain free text, abbreviations and stale information. Any audit that forces these uncertainties into clean categories will create false confidence.

The matrix should therefore include unknown affiliation as a first-class reporting category. It should distinguish missing disclosure, ambiguous organisation identity, unresolved corporate control, uncertain operator class and unavailable resource linkage. Those forms of unknownness carry different implications. A high missing-disclosure rate suggests that registration design or trust needs improvement. A high unresolved-control rate suggests that apparent organisational diversity cannot be confirmed. A high unknown operator-class rate limits claims about operational breadth even if the economy count is accurate.

Publishing unknown shares creates an incentive to improve evidence without punishing entities. Registration can explain why optional fields matter and permit later correction. Organisations can maintain public descriptions of corporate control and network roles. The organiser can invite entities to verify aggregate classification before publication. Sensitive advisory relationships can remain undisclosed while their absence is reflected in the confidence statement.

Unknown should never be silently redistributed across known categories. Nor should uncertain entities be excluded from the attendance total. The report can say: these people attended; their geographic association was observed; their organisational independence or network role could not be established. That sentence is more credible than a polished chart whose apparent precision depends on guesses. Missing information is itself evidence about the limits of institutional visibility.

Voting rights must remain explicit and bounded

LACNIC does not need an invented voting principle to recognise network exposure. Its published governance already specifies voting rights. The membership FAQ ties voting to membership category and assigned votes, while the table of votes publishes how votes apply in assemblies and elections. The bylaws establish the Assembly's authority and decision rules. Any analytical report must respect those legal boundaries rather than suggesting that an attendance analysis can alter them informally.

That means no vote multiplier for ASNs, address holdings, subscribers, revenue, traffic or number of economies served. Such multipliers would reward scale, entrench incumbency and confuse operational consequence with political entitlement. They would also encourage technical restructuring for electoral advantage and reduce people dependent on networks to assets on a corporate balance sheet. The matrix is designed to expose whose evidence is missing, not to auction authority to the largest holder.

At the same time, formal voting totals should not be described without their organisational and geographic context. Election legitimacy benefits from knowing turnout among eligible organisations, the distribution of votes across membership tiers, corporate-control concentration where lawfully observable, the economies in which voting organisations operate, and the share of eligible authority that remained unused. These measures do not invalidate a result reached under the rules. They help members assess whether the franchise is functioning broadly or resting on a narrow active core.

The distinction is constitutional: rules determine who may decide; evidence reveals the environment in which they decided. A lawful result may still disclose a participation weakness that deserves outreach, review or a future rule change through the proper process. An analytically broad meeting may still lack authority to override the Assembly. Keeping those propositions together protects both participation and institutional order.

Small economies must not become statistical decoration

Criticism of economy counts can easily be misunderstood as criticism of small-economy participation. The opposite is intended. Small economies are most at risk when a flag is used as sufficient evidence and their distinctive dependencies remain unexplored. A single attendee may permit the organiser to colour a map while having too little speaking access, preparation support or mandate to explain the country's network conditions. Geographic inclusion should trigger curiosity, not close the inquiry.

The matrix should therefore preserve qualitative significance alongside scale. A small market may rely on few international routes, face high transit costs, operate under unusual licensing rules, or depend heavily on one provider. A community network may serve a remote population not visible in national subscriber totals. A government or university network may carry essential services despite modest address holdings. These cases can change the assessment of a rule even when they contribute little to regional aggregate volume.

Impact review should ask whether a proposal creates fixed costs, expertise requirements, deadlines or documentation burdens that fall disproportionately on smaller organisations and economies. It should invite direct evidence rather than assume vulnerability. Where evidence is missing, the report should name the needed material: implementation cost estimates, network staffing levels, dependency maps, local legal constraints, outage history or testimony from affected operators. It should not fabricate a hardship narrative on behalf of people who were not consulted.

Equal respect does not require pretending all networks are operationally identical. It requires allowing both scale and specificity into the record without allowing either to purchase authority. A dominant operator's implementation evidence may be indispensable; a small economy's resilience constraint may be decisive. The institution earns trust by showing how both were considered.

The impact audit connects participation to the decision

Not every meeting requires the same depth of representation analysis. A training session, operational update, policy proposal, member resolution and Board election carry different consequences. An impact audit should scale evidence to the decision. Its first page identifies the action, legal authority, affected resource or service, implementation date, reversal path and bodies entitled to decide. It then maps likely effects across economies, network classes, resource relationships, market positions and downstream dependencies.

The audit asks six practical questions. Which organisations must change conduct? Which networks carry technical or compliance risk? Which customers or public services depend on them? Which member rights or election outcomes are implicated? What evidence came from each materially affected class? Which effects remain unknown? The economy-to-network matrix supplies the participation side; policy text, implementation plans and authoritative market or resource records supply the consequence side.

Where participation and impact do not align, the audit proposes safeguards. A consequential rule supported mainly by one operator class may receive a staged implementation, targeted consultation and scheduled review. A decision affecting small networks may require a burden estimate and accessible assistance. An election with lawful but concentrated turnout may trigger member engagement before the next cycle, not retroactive invalidation. A meeting with broad reach but weak speaking diversity may revise agenda and chair practices.

The audit should not claim certainty beyond evidence. If no verified customer, traffic or market data exist, it should identify the missing records rather than estimate them from address holdings. If corporate affiliation is unclear, it should publish the unresolved share. If affected future entrants cannot be counted, it should test scenarios and preserve a review route. The strength of the audit lies in making ignorance actionable.

Chairs need a representation note at closure

When a chair declares consensus or closes a consultation, the public record should include a concise representation note. It states who was eligible to participate, which economies and organisation classes contributed, where speech or authorship was concentrated, which materially affected groups were not observed, how affiliations were treated, and what safeguards address the gaps. The note does not replace substantive evaluation of arguments. It explains the sample from which that evaluation emerged.

The chair should avoid two opposite errors. One is numerical majoritarianism: treating the largest class, market or resource holder as correct. The other is numerical indifference: asserting that evidence about participation never matters because consensus concerns argument quality. Argument quality and participation breadth answer different legitimacy needs. A technically sound proposal can still miss implementation evidence held by absent operators. A broadly supported proposal can still be wrong. The note keeps both inquiries visible.

If important evidence is missing, closure need not always stop. Proportionate responses include a last call directed to a missing class, an implementation test, a narrower initial scope, an explicit review date, or monitoring criteria. For urgent operational risks, the institution may act while recording why delay would be worse and how the decision can be revisited. The note makes that trade-off auditable.

Chairs should receive analytical support rather than being asked to resolve affiliations personally. Staff can prepare aggregate tables under published definitions, while entities can challenge classification and correct errors. The chair remains responsible for explaining how the participation record affected confidence in the outcome.

Boards and members should review the impact of narrow participation

Board and member oversight begins after the meeting report, not before it. The Board can ask whether policy conclusions were supported by evidence from materially affected network classes, whether implementation plans address documented gaps, and whether recurring concentration warrants investment in access or notice. Members can examine whether electoral and assembly participation extends across the organisations entitled to exercise authority. Neither body should use the audit to substitute its preference for an open policy conclusion.

LACNIC's governance documents make the distinction possible. The public policy process identifies roles for discussion, consensus and Board action. The bylaws define Assembly powers and member voting rules. A representation audit connects these stages by showing when the population changes. It can reveal that broad public discussion narrowed to a small group of sustained contributors, that Board ratification occurred after limited operational evidence, or that a member vote drew authority from a concentrated subset of eligible organisations.

Oversight should focus on repeat patterns and consequences. A single missing operator class may be benign for one issue and serious for another. Persistent unknown affiliations, low member turnout, concentrated speaking or absence of small networks across several consequential decisions deserves a response. Budgets can support translation, remote participation, evidence collection and targeted engagement. Election administration can improve contact accuracy and participation information. Policy review can test whether predicted burdens occurred.

The audit does not hand the Board a veto based on demographics. It provides the factual basis for proportionate stewardship. Members and directors can then explain what they changed, what remained uncertain and when the effect will be reviewed.

What evidence still needs to be collected

Some elements of network representation cannot be derived from public records today. A defensible audit should name them. First, event records need stable organisation identifiers and voluntary role declarations that distinguish employer, represented interest and personal participation. Second, corporate-control relationships need verified, contestable sources. Third, direct resource links need careful matching to RIR delegation records, with separate treatment for IPv4, IPv6 and ASNs and no inference that holdings equal use.

Fourth, network exposure requires evidence beyond resources. Depending on the issue, this may include regulator-published subscriber shares, verified service coverage, interconnection dependencies, critical public-service roles, implementation staffing and downstream customer relationships. Fifth, influence analysis requires timestamped transcripts, list archives, proposal versions, chair summaries and decision records that permit a contribution to be followed into or out of final text.

Sixth, authority analysis needs eligible-voter totals, valid participation, membership tier and applicable corporate-control disclosure consistent with election rules.

Where these records are unavailable, the institution should publish a collection plan rather than fill gaps with convenient proxies. It can state who maintains the evidence, what legal or privacy limits apply, how entities can correct it, and which future report will test the improved coverage. It can seek regulator or member cooperation without making unsupported claims in the meantime.

The list will change by decision. A routing-security proposal needs different exposure evidence from an election-rule amendment. An operational outage review needs different affected classes from a training programme. The matrix supplies a common spine, while the impact audit chooses the columns relevant to the actual consequence.

A minimum publication standard

A credible meeting or decision report can adopt a practical minimum without waiting for perfect information. It should publish eight separate panels. The first gives registrations, check-ins and active entities by mode. The second gives economy reach with definitions and unknowns. The third gives unique organisations and control groups. The fourth gives operator classes and direct resource-link coverage. The fifth gives formal roles, eligible authority and actual decision participation. The sixth gives speaking and accepted-contribution concentration. The seventh describes known market or dependency exposure for the issue.

The eighth lists missing evidence, safeguards and the review date.

Every panel should include numerator, relevant denominator, coverage rate, privacy treatment and methodological change from the prior comparable report. No panel may borrow the conclusion of another. Many economies do not prove many networks. Many organisations do not prove many independent control groups. Large holdings do not prove large customer impact. Many speakers do not prove influence. A lawful vote does not prove broad participation, and broad participation does not confer a vote.

The public narrative should then use disciplined verbs: attended, registered, checked in, spoke, authored, voted, held resources, operated, was eligible, was observed, or could not be verified. The word “represented” should always identify the represented unit and the mandate or evidence connecting speaker to it. If that connection is absent, “entities from” is usually the accurate phrase.

This standard is demanding because legitimacy claims are consequential. It is also achievable because much of the underlying material already exists across meeting systems, public records, membership administration and decision documents. The remaining task is to connect it responsibly and publish its limits.

Conclusion: count economies, then keep counting

Countries and economies represented remain worth counting. Geographic contact matters in a region whose networks, languages, markets and institutional capacities differ sharply. A meeting that repeatedly misses entire parts of its service region has a problem. A meeting that expands that reach has accomplished something real. The mistake is to stop the inquiry at the map.

Network representation requires several independent views. Unique organisations test whether badges reflect separate institutions. Corporate-control groups test whether those institutions can act independently. ASNs and address holdings illuminate direct resource exposure while remaining unsuitable as voting weights. Operator classes reveal functional diversity. Market and dependency evidence show where consequences may concentrate. Formal roles identify who could decide. Speaking and text records show whose contribution entered the outcome. Unknown affiliation reporting marks the boundaries of knowledge.

The economy-to-network representation matrix brings these views together without collapsing them. Its rows respect geographic breadth; its columns prevent geography from claiming more than it proves. An impact audit then asks whether participation matched the decision's likely consequences and attaches safeguards where it did not. Confidence statements keep every conclusion tied to the quality of its evidence.

This approach neither privileges large networks nor diminishes small economies. Scale can reveal implementation exposure, while specificity can reveal risks hidden by scale. Neither purchases authority. Open entities retain equal standing to offer evidence and argument; members retain the voting rights established by published rules; boards and chairs retain responsibilities that must be exercised transparently.

Sources

• ICANN, Meetings By the Numbers — official meeting reports, attendance and regional reporting, public-session count material, and stated limits in historical collection.
• ICANN, A Look Back at ICANN79 — an official example separating total in-person and virtual participation from countries and territories reached and regional distribution.
• NRO, RIR Statistics — joint delegated-extended reporting for IPv4, IPv6 and ASNs, plus documentation links and individual RIR statistical sources.
• LACNIC, Prior Event Entities — official separation of average attendees, represented organisations, activity types, government institutions, academia and civil society in LACNIC event participation material.
• LACNIC, Terms and Conditions for Hosting a LACNIC Event — the composition of LACNIC's two annual events, including the Member Assembly, Public Policy Forum, technical activities and co-located regional bodies.
• LACNIC, Policy Development Process — open policy-list participation, consensus responsibilities, public forum administration and the Board's place in the published process.
• LACNIC, Membership FAQ — membership voting rights, assigned votes, the membership contact's authority and nomination responsibilities.
• LACNIC, Chapter IV: Assemblies — the General Assembly's governing role, notice, agenda, quorum, voting and voters' registry provisions.
• LACNIC, Table of Votes — published treatment of votes for the General Assembly and statutory election processes.