Summary

  • On 3 December 2021, Cloud Innovation obtained a wide ex parte interim order protecting its Resource Member membership from specified termination measures and similar acts by AFRINIC.
  • On 14 February 2022, the appellate court concluded that this later protection meant that allowing Cloud Innovation’s appeal from a 7 July 2021 judgment would serve no practical purpose.
  • The appeal was set aside with costs, and its continued pursuit was held to be an abuse of process; the court did not thereby affirm the 7 July reasoning or decide the underlying membership, contract or number-resource merits.
  • The sound institutional response is to track each application, order, unresolved issue and permitted registry action in a legal-state receipt, keeping technical services stable and every provisional intervention reversible.

When an appeal loses its practical job

On 3 December 2021, Cloud Innovation Ltd obtained something that changed the purpose of another proceeding already under way. In SN 1947/2021, a Judge in Chambers granted a wide ex parte interim order restraining the African Network Information Centre, or AFRINIC, from giving effect to its 8 July Board resolution, its 1 December letter, or similar acts that would terminate Cloud Innovation’s membership as an AFRINIC Resource Member. The order also restrained conduct in breach of an undertaking dated 15 July in another application.

It protected, in broad terms, the same practical interest that Cloud Innovation’s earlier court application and the appeal arising from it had sought to preserve.

That earlier application had met a different fate. On 7 July 2021, a Judge in Chambers upheld AFRINIC’s preliminary objection and set the application aside with costs. Cloud Innovation appealed. It asked the appellate court to quash the July judgment and send the first application back for consideration by a different Judge. Had that been the only route to protection still open, the appeal would have had an obvious practical function: reversing the July outcome could restore an opportunity to seek restraint against termination of Resource Member membership.

But it was no longer the only route. By the time the appeal was heard, the 3 December order was already standing between AFRINIC and the termination measures identified in it. On 14 February 2022, Judges D. Chan Kan Cheong and R. Teelock, sitting in the Supreme Court of Mauritius in Record No. 121865, treated that intervening fact as decisive. Their judgment, 2022 SCJ 51, reasoned that remitting the first application would waste time and resources because Cloud Innovation had already secured wide interim protection for the same practical interest. Even if the appeal were allowed, it would add no useful protection.

The appeal had therefore become academic.

Here, “academic” is a conclusion about present utility. It does not mean scholarly, abstract in the casual sense, or intellectually unimportant. Nor does it mean that the court reviewed the 7 July reasons and found them correct. It means that, in the legal state confronting the court on 14 February, victory in the appeal would not produce a practical result that the later order had not already supplied. The later remedy had overtaken the earlier appellate vehicle.

That distinction matters beyond the courtroom. Number-registry records sit close to operational dependencies: membership status, registration information, contactability and related technical services can shape how operators understand and maintain their position. If an academic appeal is reported as a merits victory for the registry, a procedural closing can be mistaken for authority to make lasting changes. The registry may then appear entitled to act as though the underlying disagreement has been adjudicated when it has not.

A sentence about present judicial economy can, through careless retelling, harden into a false mandate affecting a live network.

The court did impose consequences. It concluded that pursuing the appeal would constitute an abuse of the process of the court, and it set the appeal aside with costs. It criticised the failure of Cloud Innovation’s legal advisers to disclose the existence and particulars of the later application, and it expressed concern about successive applications seeking in effect the same remedy. These are real findings about the use of the court’s processes. They should not be softened into mere administrative housekeeping.

They should not be enlarged either. An abuse-of-process conclusion about continuing an appeal after overlapping protection was obtained does not decide whether the Resource Member Services Agreement was breached. Costs do not decide final entitlement to number resources. Criticism of non-disclosure does not resolve ownership, leasing, routing, damages or policy validity. The literal disposition is that the appeal was set aside with costs. It was not a judgment affirming the 7 July reasoning, and it was not a merits award to AFRINIC on the underlying dispute.

What the later protection did—and did not do

The 3 December order must be held in equally precise focus. It was ex parte, meaning it was granted at that stage without the ordinary full contest between the parties. It was interim, not a final determination. It was wide in the restraints it recorded. And it was returnable to show cause why it should not be made interlocutory pending determination of the disputes between the parties. Those features explain both why it could remove the immediate practical need for the appeal and why it could not settle the underlying merits.

Provisional relief can be practically sufficient without being legally final. If a party seeks to prevent a step tomorrow, an interim restraint in force today may protect the relevant interest even though the deeper contractual questions remain open. That is the causal core of 2022 SCJ 51. The appellate court did not need the 3 December order to be permanent in order to ask whether remittal of the first application would do useful work at that moment. Its practical breadth was enough for the court’s utility analysis.

The judgment does not establish what ultimately happened to SN 1947/2021. It does not prove that the order later became interlocutory, permanent or final. It does not supply a final determination of the disputes that the return hearing was meant to address. Any account that silently upgrades “returnable” to “continued,” or “interim” to “won,” fills an evidentiary gap with an invented outcome.

The same care applies to the main case. The appellate court stated that, as far as it had been able to ascertain on 14 February 2022, Cloud Innovation had not yet lodged a main case. That is a dated observation about what the court could ascertain, not a claim that no main case was ever brought. Later events are not proved merely because the judgment describes the position on the day it was delivered.

Cloud Innovation had originally advanced seven grounds of appeal. At the hearing, it dropped grounds 1(v), 2, 3 and 5. The remaining grounds were not decided on their merits because the later procedural state removed the appeal’s practical purpose. This non-decision is important only to keep the account honest: dropped grounds are not the same as rejected grounds, and unexamined grounds are not judicially disproved propositions. The centre of this judgment is the effect of later relief on present utility, not a catalogue of what each unresolved ground might have meant.

The result can be stated without paradox. The later order mattered enough to make the earlier appeal academic, but it remained provisional. The appellate disposition mattered enough to close that appeal and impose costs, but it did not validate the decision under appeal. Both effects were consequential. Neither should be asked to carry a legal conclusion it did not contain.

Repetition, disclosure and the court’s finite attention

The annexed court record identified applications lodged on 12 July, 13 July, 3 September, 6 September, 26 November, 1 December and 3 December 2021 after the first application. They sought, in effect, the same protection from termination of Resource Member membership. The appellate court recorded that all those later applications had been set aside except those lodged on 6 September and 3 December. The dates matter here not as an invitation to reconstruct every turn of the litigation, but as evidence of procedural overlap.

Overlap changes what candour requires. A court considering whether to restore an earlier application needs to know if another order is already protecting the same interest. That information goes directly to whether its decision can still change the parties’ position. Without it, the court may spend time addressing a vehicle whose requested destination has already been reached by another route. The judgment’s regret that the later application and its particulars had not been disclosed follows from this practical problem.

There is a strong and legitimate case for the court’s response. Courts exist to resolve live disputes, not to serve as advisory bureaux on questions that no longer alter the parties’ practical position. Their capacity is finite. An opposing party should not have to respond indefinitely to parallel routes seeking materially the same remedy, and judicial attention should not be consumed by a remittal that duplicates protection already in place. Litigants and their advisers owe candour about later developments capable of changing the usefulness of pending proceedings.

On that account, repeated applications and missing disclosure were not trivial. Once the 3 December order provided wide protection, continuing to press the earlier appeal without squarely presenting the newer state could burden the court and AFRINIC with a proceeding that no longer offered practical value. The abuse-of-process conclusion and the costs order express the court’s control over that burden. Accepting that conclusion is compatible with insisting on its boundaries.

Case-management finality and merits finality answer different questions. The first asks whether this proceeding should continue now. The second asks who is substantively right about the disputed rights and obligations. A court may close a proceeding because it has been overtaken, duplicated or rendered useless while leaving the disputed contract, entitlement and authority questions untouched. Efficiency would be undermined, not protected, if every procedural disposal were later inflated into a substantive judgment the court never made.

This is also why “AFRINIC won” is a misleading shorthand. AFRINIC received the benefit of the appeal being set aside with costs, and the court accepted that its process should not be used to pursue an academic result. But the judgment did not award AFRINIC victory on RSA breach, membership termination, resource entitlement, ownership, leasing, routing, damages or policy authority. A careful account names the precise proceeding, remedy and consequence. It does not transform a bounded appellate disposition into a general litigation result.

A present-tense ruling, not a rewritten past

Mootness is temporal. On 7 July, the first application had been set aside. On 3 December, another application produced a wide interim restraint. On 14 February, the appellate court evaluated what useful work remained for the appeal in light of that restraint. The answer depended on the state as it then stood. Change the later order, and the utility inquiry might look different; preserve it, and remittal might remain duplicative. The judgment’s logic therefore runs forward from the operative protection, not backward to certify the earlier reasoning.

That temporal direction is easy to lose in ordinary summaries. People often read a closed appeal as confirmation of whatever lies below it. But an appellate court can decline to decide because there is no longer a live practical question. Its refusal to spend judicial resources is not secretly an answer to the substantive question. In 2022 SCJ 51, the later remedy is the bridge between the procedural history and the disposition. Remove that bridge from the account, and the result appears to say much more about the 7 July judgment than it did.

The official judgment is the controlling source for that bridge. It establishes the case identity, the first application, the later applications recorded by the court, the breadth and provisional character of the 3 December order, the court’s practical-purpose reasoning, its criticism of disclosure, its abuse-of-process conclusion, and the set-aside with costs. It does not establish the truth of every allegation made by either party. It does not reveal every service, disclosure or hearing event associated with the later applications. Nor does it quantify any operational harm.

The uncertainty is not a weakness to be concealed. It is part of the legal state. We know why the appellate court considered the appeal academic. We do not know from this judgment the later final fate of the protective order, the full contents of all written submissions, every one of the seven grounds in its complete form, or any eventual registry implementation. Accurate institutional reporting keeps these absences visible so that a reader can distinguish the court’s holding from inference and later history.