Summary

  • Two ICANN Board decisions and one IANA evaluation carried the transfer of African number-resource registry functions; read in sequence, they show an approval that was conditional 190 days before it was final.
  • The provisional grant of 30 September 2004 was paired with an expectation — a completed transition plan and a revised application — and it did not certify conformity with the criteria that full recognition later rested on.
  • The endorsement that supported full recognition on 8 April 2005 carried the signature of the chief executive of the registry handing over the service region, acting as chair of the body assessing readiness.
  • AFRINIC began operating as a fully functional registry on 21 February 2005, 46 days before recognition. No ICANN instrument fixes that date; the service relationship moved before the instrument usually treated as the transfer.
  • On the face of the located record, the only body positioned to revise either decision was the Board that made it, and no challenge, dissent or reconsideration appears in the instruments located for this article.

The question the recognition language does not answer

A proclamation of recognition is not the same as a record of what was assessed. The two are easy to conflate when a handover is described after the fact, because the closing instrument speaks in the present tense: from 8 April 2005 there was a fifth regional internet registry for Africa. That sentence is true and it is not very informative. It does not say which earlier decision made the arrangement provisional, what condition attached, who signed the assessment that resolved the condition, or what recourse existed if a resource holder in the region disagreed with the outcome.

This article works through the instruments themselves for those answers. The chain is short: a special meeting of the ICANN Board on 30 September 2004, a joint registry communication and an operational start on 21 February 2005, a letter of support dated 28 March 2005, an IANA evaluation report, and a regular Board meeting in Mar del Plata on 8 April 2005.

The reading below relies on the operative language as it appears in the located record, including provider-retained excerpts of the resolution pages rather than a full verbatim reproduction of every underlying PDF. Where a quotation is presented as published wording, it is excerpt-reported; where the record itself is inconsistent, that inconsistency is described rather than smoothed over.

30 September 2004: approval with a condition attached

At its special meeting of 30 September 2004, the ICANN Board addressed the AFRINIC application for recognition as item three on its agenda. The operative resolution reported in the located record, numbered [04.82], gave provisional approval to the application and to the transition plan, "with the expectation that the transition plan will be completed and an amended or revised application for recognition will be submitted" (ICANN approved board resolutions, special meeting of 30 September 2004).

Two features of that sentence matter more than the word "approval". First, the object of the approval was compound: the application and the transition plan together. Second, the resolution recorded an expectation rather than a finding. Nothing in the located text of [04.82] states that the criteria for regional internet registry recognition had been met; it states that a plan existed and that a further application was expected.

The companion resolution, [04.83], gives the mechanism. It directed the ICANN President to work with AFRINIC, and with ARIN, the RIPE NCC and APNIC through the Number Resource Organization, to assist a smooth transition and the preparation of a revised application for full recognition conforming to the then-current ICP-2 criteria and the ASO memorandum of understanding (ICANN approved board resolutions, special meeting of 30 September 2004).

Read as a pair, the two resolutions describe a gate rather than a destination. The Board approved progress and delegated the preparation of the instrument that would later be judged. That structure is ordinary in institutional practice, and it is precisely why the later instrument has to be read on its own terms: the thing that was approved in September 2004 is not the thing that was certified in April 2005.

The evaluator problem: one signature, two roles

Full recognition depended on an assessment of readiness. The located record places that assessment inside the Number Resource Organization, and the NRO's written support to ICANN's President is signed by Axel Pawlik in his capacity as NRO chair, alongside AFRINIC chief executive Adiel A. Akplogan. The letter, dated 28 March 2005, reports AFRINIC's progress since operations began, states that the NRO reviewed AFRINIC against the ASO memorandum and ICP-2 and found the requirements met, and concludes that the NRO "strongly supports an ICANN Board resolution to recognize AfriNIC as the fifth RIR" (NRO letter to the ICANN President, 28 March 2005).

The IANA evaluation report records an earlier and equivalent step: on 21 February 2005 the chairman of the NRO and the chief executive of AFRINIC jointly communicated the NRO's favourable assessment of AFRINIC's readiness for final approval and recognition (IANA report on recognition of AfriNIC, 8 April 2005).

The NRO chair at the time was also the chief executive of the RIPE NCC — the registry whose African service relationships were being handed over. That is not an allegation of impropriety, and the record does not support one. It is a structural observation with consequences for how the evidence should be weighed. The body assessing whether the receiving registry was ready to take over the service region included, in its chair, the outgoing provider of that service.

A favourable finding by that body resolved the region's interest in an orderly handover and the outgoing registry's interest in not leaving a vacuum it would have to keep filling. Both interests are legitimate. Neither is independent of the question being assessed.

The IANA report itself is signed by neither registry, and its stated role was to evaluate the application against the ten ICP-2 principles and to conclude whether the requirements were met (IANA report on recognition of AfriNIC, 8 April 2005). That is the closest the located chain comes to an assessment separated from the parties to the transfer, and it is a technical evaluation of criteria rather than an independent audit of competing claims.

21 February 2005: the service moved before the instrument

The most consequential date in the whole sequence is the one that no ICANN resolution establishes. AFRINIC began operating as a fully functional regional internet registry on 21 February 2005, in anticipation of final recognition the following April; the same date carried an NRO progress update submitted to ICANN (NRO announcement archive, February 2005). The IANA report confirms the date independently in its account of the joint communication (IANA report on recognition of AfriNIC, 8 April 2005).

That leaves a 46-day interval — from 21 February to 8 April 2005 — in which the region's registry operations were being performed by an entity that had not yet been recognised as a regional internet registry. During that interval the evaluation was still being completed and the Board had not yet acted. Resource holders were transacting with AFRINIC, and the NRO was reporting on that transacting as evidence of readiness.

The sequence is not secret; it appears in the NRO's own announcement. It matters because the recognition instrument is often treated as the moment the registry changed hands. On the dates, the service changed hands in February. The instrument confirmed it in April.

8 April 2005: what (05.25) actually certified

At its regular meeting in Mar del Plata on 8 April 2005, the Board acted on the completed record. The resolution reported in the located record, (05.25), proclaimed AfriNIC "a fully approved and recognized Regional Internet Registry, to provide IP address registration and other services for the Africa service region", and (05.26) recorded the Board's congratulations and thanks to the NRO, AFRINIC and related parties (ICANN adopted board resolutions, Mar del Plata, 8 April 2005; ICANN historical resolution tracking, recognition of AfriNIC as regional internet registry).

The recited basis is worth extracting, because it is where the certification actually lives. The whereas clauses record the 30 September 2004 provisional recognition; a joint statement from APNIC, ARIN, LACNIC and the RIPE NCC conveyed through the NRO, praising AFRINIC and recommending a favourable response; the NRO chair's collective favourable assessment and recommendation for final approval and recognition; and the President's determination that the application was in full conformance with ICP-2 criteria and that the transition plan was complete (ICANN adopted board resolutions, Mar del Plata, 8 April 2005).

So the operative proclamation sits on top of a single-executive determination: the President reviewed the application and the IANA report and determined conformance. The Board then proclaimed recognition as an act of the whole body. The distinction between the two is not a technicality. It tells a resource holder where the reasoning was done and which part of the record can be checked against criteria — the report and the determination recorded in the recitals — versus which part is the granting act itself.

The evaluation's own dates do not line up

The IANA report exists in more than one located form, and they disagree with each other. The HTML version carries the date 8 April 2005 (IANA report on recognition of AfriNIC, 8 April 2005); a PDF variant carries a header date of 6 April 2005 and a filename suggesting an August 2005 posting (IANA report on recognition of AfriNIC, PDF variant). The PDF also contains the typographical error "September 20054" in its body text (IANA report on recognition of AfriNIC, PDF variant).

These are version and posting artefacts rather than substantive conflicts. Their content on the ICP-2 evaluation is parallel: both conclude that AFRINIC met all the conditions specified in ICP-2 and all the necessary requirements for approval and recognition (IANA report on recognition of AfriNIC, 8 April 2005; IANA report on recognition of AfriNIC, PDF variant).

They nevertheless matter for citation discipline. A chain of dated instruments is only as strong as the dates attached to it, and here the technical evaluation supporting a recognition decision circulated in copies whose internal dates differ by two days and whose file metadata points to a different month. An article that asserts the transfer "rested on a report of 8 April 2005" is doing a small act of tidying that the record does not quite perform on its own.

The safe formulation is the one this article uses: the report supporting the recognition decision is dated 8 April 2005 in its HTML form, with a PDF variant dated 6 April 2005.

Who could challenge each step

The located instruments are silent on recourse. [04.82] and [04.83] impose a process and delegate preparation of a revised application; (05.25) proclaims a status. None of them names a forum in which a resource holder in the Africa service region could contest a registry decision, or in which a party could challenge the recognition itself.

On the face of the record, the body positioned to revise either decision was the Board that made it. That is a reading of the instruments, not a quotation from them, and it should be labelled as such. What can be said of the record itself is narrower and still useful: in the instruments located for this article, no reconsideration, dissent or recorded challenge appears at any step between 30 September 2004 and 8 April 2005.

The peer endorsement sequence ran from the joint communication of 21 February 2005 through the letter of 28 March 2005 to the Board's April decision without a documented objection among the registries or from any other located source.

Two gaps limit how far this silence should be pushed. The standalone instrument of the 21 February 2005 joint communication was not located; the record shows it as a progress update submission rather than as a retrieved document of its own (NRO announcement archive, February 2005). And the ICANN historical tracking page for the recognition resolution carries a footer note stating that the resolution did not address funding for the items identified in it — a qualification that has not been confirmed against the primary document and is therefore flagged rather than relied on (ICANN historical resolution tracking, recognition of AfriNIC as regional internet registry).

A third dating discrepancy belongs in the same list, because it shows how quickly secondary retellings drift from the primary instrument: NRO announcement material describes the provisional recognition as granted "in October 2004", while the operative resolution is dated 30 September 2004 (NRO announcement archive, February 2005; ICANN approved board resolutions, special meeting of 30 September 2004). The primary instrument should govern; the drift should be visible.

What the instruments did not address

Two absences follow from reading the chain closely. The first is the one already described: no instrument in the chain creates a review route for the people who would live under the decisions. The second is continuity. Nothing in [04.82], [04.83] or (05.25) conditions recognition on what would happen to registry objects, reverse delegations and member records if the recognised registry later ceased to be able to hold them. The instruments move authority and confirm the qualities of the receiving institution at closing. They do not escrow the object of the transfer.

That absence is not a criticism of any signatory at the time, and it is not a finding of fact about the present. It is a description of scope: the 2004–2005 instruments were written to move a function between institutions, not to specify what the function's beneficiaries could rely on if the receiving institution failed. Whether that mattered, and when, is a question for the documents that came afterwards.

For anyone who transacts with a regional internet registry today, the practical upshot of this reading is modest and specific. When a handover is described as an act of recognition, ask which instrument made it provisional, what condition attached, whose signature carried the assessment, and on which date the service actually moved. In this chain the answers are: [04.82] and [04.83]; a completed transition plan and a revised application; the NRO chair who was also the outgoing registry's chief executive; and 21 February 2005.