Summary

  • AFRINIC’s published Resolution 201110.128 records an instruction: the Board asked Dr Viv Padayatchy, whom it described as former chair and registered director, to take necessary actions to place seven directors before the Registrar of Companies according to their mandates. It proves that AFRINIC published those words, not who the seven were, whether their selection was valid, whether a filing occurred, or what the Registrar did.
  • The institutional chain has distinct links: internal election or appointment, assumption of office, written consent and certification, company notice, Registrar receipt, and public-register visibility. Collapsing these links either turns an administrative record into a source of office it may not be, or treats an evidentiary gap as automatic invalidity when the historical statute itself warns against that conclusion.
  • The proportionate answer is a dated, privacy-conscious authority packet connecting each mandate to consent, filing and receipt, plus an explicit mismatch notice and cure path when records diverge. Routine registry operations must continue; number resources, reverse DNS, RPKI, contact records and transfers must never be used as leverage in a corporate-record dispute.
  • AFRINIC remains a private bookkeeper and coordinator. Neither a member vote, a Board resolution nor a Mauritian corporate filing gives it sovereign, regulatory, police, punitive, confiscatory or adjudicative power over operators, assets or the Internet.

The former chair at the hinge

The most revealing detail in Resolution 201110.128 is not the number seven. It is the choice of messenger. In the wording AFRINIC published in its October 2011 group of Board resolutions, the Board resolved to request Dr Viv Padayatchy—described on the page as a former chair and registered director—to take the necessary actions to “appoint the 7 Directors to the Registrar of companies according to their mandates”. An incumbent Board was therefore asking a former chair who still appeared, by AFRINIC’s own description, in a registered capacity to help connect seven current mandates to the Mauritian corporate record.

That sentence is awkward, and its awkwardness is useful. It reveals a hinge between two institutional worlds. On one side sat whatever internal process had produced the seven mandates. On the other sat a public company register meant to tell outsiders whom the company represented as its directors. Between them were actions that could not safely be replaced by a title, a recollection or a slogan about community choice.

The resolution does not say who the seven people were. It does not supply their election certificates, appointment instruments, mandate dates, written consents or non-disqualification certificates. It does not record the filing date, a receipt, a query, an acceptance or a rejection from the Registrar. It says nothing about Padayatchy’s response, whether he had exclusive capacity to submit anything, or whether another person could act. Nor does the archived Board page establish the attendees, quorum, mover, seconder, vote or conflicts behind this particular resolution.

The evidence supports an instruction and the company’s published description of the people and task; it does not support a verdict on implementation or legality.

This restraint matters because corporate records invite two opposite mistakes. The first is to assume that an internal election is enough for every audience and purpose, so a public filing can be treated as decorative administration. The second is to treat the register as an altar from which directorship, legitimacy and authority all descend. Resolution 201110.128 supports neither position. It shows that the Board itself regarded some further action involving the Registrar as necessary, while saying nothing that would make the Registrar the source of the seven mandates.

Six links, six different questions

A sound reading begins by separating six questions.

First, what competent internal act selected or appointed each person? The archived 2007 AFRINIC bylaws describe members electing primary and alternate directors, a Board of up to seven directors—six primary directors plus the chief executive—and three-year terms for the elected seats. They also required at least one director ordinarily resident in Mauritius. Those bylaws are close historical context, not conclusive proof of the exact constitutional text operative in October 2011, and their numerical resemblance to the resolution does not identify any of the seven.

Second, when and on what terms did each mandate begin? A result, resolution or appointment instrument needs a date, a seat and a term. Without those particulars, an observer cannot tell whether two records genuinely conflict or merely describe different moments. The available resolution supplies none of them.

Third, did each person consent to office and provide the statutory certification? The historical Companies Act text says a person may not be appointed as a director without written consent and certification that the person is not disqualified. Consent is not a ballot result. It proves the named person’s acceptance and certification, not the validity of the process that chose that person.

Fourth, did the Board deliver or cause delivery of an approved notice of the change? Historical section 142 placed that responsibility on the Board. The notice was to identify the change date, list current directors and, for a new director, include the required consent and certificate. The text stated a 28-day period following an appointment or resignation change. Yet neither the change date nor any delivery date for these seven is known, so the number 28 cannot honestly be converted into an allegation of lateness.

Fifth, did the Registrar receive the notice? A submission copy and a receipt answer different questions. One shows what the company intended to send; the other establishes that the public office received something at a stated time. There is no receipt in the evidence considered here.

Sixth, what did the public register show, and from when? Visibility matters for reliance, but it is not the same as the preceding acts. A public extract can show the state of the record; it cannot by itself prove the election, consent or substantive validity behind every entry. No before-and-after extract is available for this resolution.

These distinctions prevent institutional wishful thinking. If the internal choice is known but the consent is not, the remedy is to verify consent. If the filing is known but the public display has not caught up, the remedy is to establish receipt and mark the visibility mismatch. If a register entry exists but the mandate instrument cannot be produced, the public entry should not be asked to prove an internal event it did not witness. Each document answers its own question.

Why a private coordinator still needs corporate precision

AFRINIC’s substantive function is narrow. It keeps unique number-resource records, coordinates allocations and transfers, maintains contact and security metadata, supports auditability and helps operators prove control. These are valuable services. They do not amount to legislation, policing, punishment, confiscation or adjudication. AFRINIC is a private company and bookkeeper, not a sovereign power.

Corporate compliance therefore cannot enlarge its public authority. A member election can choose an office-holder inside the applicable private constitution; it cannot bind absent operators as citizens, create ownership of the Internet or convert the electorate into a state. A Board resolution can arrange the company’s affairs; it cannot create powers the company itself does not possess. A notice to the Registrar can make a corporate change legible; it cannot transform a bookkeeping organisation into a regulator of networks. The public register records a company fact under corporate law. It is evidence and coordination, not a throne.

Yet narrow authority is not an excuse for loose authority records. Precisely because AFRINIC performs infrastructure-facing services, its staff, banks, auditors, insurers and counterparties need a reliable answer when someone gives an instruction, signs a contract or represents the company. Operators need assurance that an argument about office-holders will not spill into resource records, route-security services or ordinary support. The fewer substantive powers an institution has, the easier it should be to specify who may exercise those powers and on what documentary basis.

The affected principal is the member or operator who depends on predictable service, not the office-holder whose title happens to be disputed. A mismatch between the internal mandate list and the public corporate register can raise the cost of legal review, banking, insurance, audit and contracting even if no appointment is ultimately invalid. It can make staff ask which signature to follow. It can make a cautious counterparty demand redundant assurances. It can also tempt institutional actors to settle a corporate question using operational leverage. That last move is the one a resilient design must categorically prevent.

Nothing in Resolution 201110.128 proves that any such cost or disruption occurred. The point is architectural: the resolution reveals a seam at which those costs can arise. Governance is strongest when it makes the seam observable before a disputed transaction exposes it.

What the page proves—and what it cannot

AFRINIC’s Board page is an official institutional record, but “official” describes its provenance, not the reach of its proof. It establishes that AFRINIC published Resolution 201110.128 in an October 2011 block. That block ends with Resolution 201110.136, which says a teleconference held on 12 October 2011 adjourned at 17:20 UTC. The association helps place the material, but it does not give Resolution 201110.128 a separate adoption timestamp.

The page also establishes AFRINIC’s wording: the request to Padayatchy, its description of him, the number seven, the Registrar reference and the phrase “according to their mandates”. It does not independently verify that Padayatchy was on the public register at that moment; no register extract is present. It does not show that the seven mandates existed on legally sufficient terms, that the Board meeting was validly convened, or that the instruction was carried out. An organisation’s publication can prove what it said and purported to do. It cannot certify its own quorum, legality, completion or wider legitimacy merely by carrying its logo.

That evidentiary boundary is not hostility to official records. It is what makes them useful. Once the record is asked only to prove what it can prove, the missing links become questions rather than accusations. One can ask for an election or appointment instrument without asserting that none existed. One can seek a receipt without saying a filing failed. One can compare a public extract with an internal list without declaring either fraudulent. The result is stricter than institutional deference and fairer than institutional suspicion.

The operational firewall

Any cure must begin with a firewall between corporate-office questions and number-resource services. If an authority mismatch appears, AFRINIC may need to preserve contested documents, require dual review for unusually consequential corporate acts, seek independent company-law advice or ask the competent public office to clarify the record. It may not freeze an operator’s resources, revoke registration, impair reverse DNS, change RPKI status, obstruct an ordinary transfer or degrade service in order to force movement on a director filing.

Those tools would be both misdirected and institutionally dangerous. The filing duty concerns the company’s own corporate record. An operator did not create the mismatch merely by relying on registry service. Number resources are not hostages that the bookkeeper may seize to discipline its officers, members or critics. Nor is technical coordination a substitute courtroom in which staff can adjudicate who deserves corporate office.

Continuity does not mean pretending that authority uncertainty is harmless. It means containing the uncertainty at the level where it belongs. Existing operational controls should remain in force; routine and reversible registry actions should continue; unusually consequential company commitments may receive documented secondary approval; and every temporary safeguard should carry a narrow scope, an owner, a review date and an automatic stop condition. The institution protects operators by refusing to turn a corporate paperwork question into an infrastructure event.