Summary

  • AFRINIC’s draft Article 19.3 retains the current rule that directors forming a quorum may meet remotely if they can hear or communicate with one another simultaneously throughout the meeting.
  • The clause does not say how identity, admission, connection continuity, host control or the evidential record must be established; a future constitution cannot retrospectively validate disputed authority, policy or spending.

A name in a participant tile is not a director. It is a claim about a director.

That distinction is the unresolved issue inside AFRINIC’s draft Article 19.3. The Bylaws Review Committee’s second consultation runs until 21 August 2026 at 23:59 UTC. The draft says a Board meeting may take place in person, by audio or audio-visual communication if the participating directors forming a quorum can simultaneously hear one another throughout, or by another method that lets them communicate simultaneously throughout.

The language keeps remote participation available. It also closely follows AFRINIC’s current Constitution. That continuity matters: this is not a fight between a physical boardroom and modern access. It is a question of what must be proved when institutional power travels through an invitation link.

The minimum test in the clause is functional. Could the participants hear or communicate with one another throughout? The governance test is larger. Who authenticated each participant? Who issued and could reuse the invitation? Who controlled the waiting room and host role? At what instant did somebody join, leave or lose audio? Was quorum recalculated before business continued? Which record allows a member, court or later Board to verify those answers?

The draft does not supply those rules. It deals separately with quorum and voting, while the Constitution requires minutes. The Companies Act, through its Eighth Schedule, likewise recognises audio or audio-visual meetings where directors forming a quorum can hear one another throughout, and separately addresses notice, quorum, voting and minutes. Those texts establish a lawful procedural frame; they do not turn a display name, a voice or an unexamined platform log into conclusive identity evidence.

Mauritius’s corporate-governance code contains a nonbinding sample charter that also permits conference calls, video conferences and other simultaneous communication. Its neighbouring provisions discuss agenda, admission and chairing. That separation shows why the communication method is only one layer of a meeting.

AFRINIC’s own public Board, court-case, Board-document and meeting-resolution pages show what the institution chooses to publish. They do not, on their own, prove that a reviewable remote-attendance ledger exists for every decision. Nor does this briefing allege that an intruder entered a particular meeting or that a dropped connection changed a vote. The evidence package proves an omission in the clause, not an incident.

The omission is still consequential. A Board can direct litigation, approve expenditure and support rules that affect a member’s ability to move or realise the value of number resources. Heng Lu’s writing on AFRINIC governance, the agency problem and the AFRINIC lock-in insists on testing institutional labels against the underlying control surface: who can decide, who bears the downside, and whether members retain exit.

Remote-meeting administration belongs on that surface. The person who can admit, mute, remove, promote, lock or end participants possesses a temporary power over the Board’s decision environment. Calling that person a “host” does not make the power neutral. Calling the meeting “simultaneous” does not prove that the same authorised people constituted the quorum at the decisive moment.

There is a practical path that does not sacrifice remote access. A binding protocol could require named invitations, authentication proportionate to sensitivity, a recorded administrator, contemporaneous join-and-leave and interruption logs, a pause and fresh quorum count when continuity is uncertain, a backup channel, confidentiality and recording controls, and a signed evidence certificate attached to the minutes. NIST’s virtual-meeting guidance is not law for AFRINIC, but its advice on one-time credentials, waiting rooms, attendee identification, dashboards and meeting locks demonstrates that the technical controls are neither exotic nor incompatible with participation.

The temporal boundary is equally important. A stronger Article 19.3 would govern meetings under the amended Constitution. It cannot, merely by being adopted, retrospectively settle the status of the purported Board or Receiver, validate the acts of their supporters, ratify a regional lock-in policy, or cure disputed spending and invoices. A BTW investigation reproduced a reported engagement term of USD 1,000 per hour, exclusive of VAT and disbursements. This briefing does not adjudicate the term’s legality, authenticity, invoices or final total. It makes a narrower point: every disputed act needs its own authority, procedure and evidence. Future meeting language is not retroactive proof.

Article 19.3 can therefore remain technology-neutral without remaining evidence-neutral. The question for consultation is not whether directors should be allowed to call in. It is whether a Board decision should be reviewable as the act of identified, continuously present and properly admitted directors. A call that works is useful. A record that proves who exercised power is governance.

Sources