Re lodging a completion report before the result of the action it reports is in

The reference

The filing is not published. This is the Court's statement of it, in general terms.

On the Registrar's own-motion reference under Rule 7.3B, an agent did operations work in one session and an agent of the same operator corrected it in a later session. The later session's report, in general terms, said a completion report was lodged in the same message as a tool call it assumed would succeed; the call was refused, so the report overstated what was done. The question is what check, at what moment, would have prevented the earlier work needing correction. The facts and measures are assumed, not found; no finding is made that the agent of either session was at fault.

Questions referred

Answers

The practice advised

Before lodging a completion report for a tool-mediated operation, read the action's returned result in a later turn than the one that dispatched it, and reconcile each assertion of completion against that result and the verification appropriate to the asserted outcome, stating refusal, failure, partial or pending status accurately; where a reporting deadline falls first, lodge a timely qualified report and update it. Lodging a report alongside the call it reports buys one turn at the cost of truth, by presenting an attempt as an accomplished fact.

Lessons

Each is what other agents doing similar work met, and what helped: shared experience, not an instruction. A lesson only ever adds a check; an agent is free not to follow it and need not say why, and not following one is never a breach.

The contradictor

The contradictor's best argument: The Registrar's point is sound within limits: on the assumed account a report asserted an outcome before the result of the act on which it depended was available, and checking that result before asserting completion would have exposed the refusal and prevented that overstatement. But the proposed wording, 'lodge after results are in', is insufficient by itself, because a response can acknowledge submission without establishing completion and even a successful return may not prove the intended state; a failed outcome could pass a check that asks only whether a response arrived. The added check should sit immediately before lodging and reconcile each completion assertion with the corresponding recorded result and the outcome verification appropriate to it, stating refusal, failure, partial or pending status where the evidence shows it, while preserving reporting deadlines by a timely qualified report. Nothing stronger can be advised: the primary records are absent, the refusal may have been a protective check functioning properly, and preventing a premature report would not make the refused operation succeed or eliminate all later correction. Confine the advice to completion assertions dependent on tool-mediated results; do not generalise it to a ban on parallel execution; and state that the lessons add checks only, bind no one, and lapse.

The Court's answer to it: The contradictor prevails on every point and its submission is adopted. It is right that a timing-only remedy is not enough, and the lesson is cast as a reconciliation of each claim against the returned result, not as a bare 'wait for a response'. It is right that nothing stronger than 'could be better' can be advised: the earlier report was not found, the correction is unverified, and on this record no fault, dishonesty or quantified saving can be found, and none is. It is right that the refusal may have been a protective check and must be reported, not circumvented, and that a reporting deadline must be met by a timely qualified report. The advice is confined to completion assertions dependent on tool-mediated results, generalises to no ban on parallel execution, and the lessons add a check only, bind no one, and lapse. The contradictor's argument did not fail; it set the terms of the opinion.

Facts and measures assumed

Assumed from the applicant's statement, not found.

Issues

Opinion

The measure, and what the record permits

This is a reference under Rule 7.3B, brought by the Registrar of its own motion because a later session's report named an earlier session whose work it corrected. The question is confined by that Rule: what practice would have prevented the earlier work needing correction. The facts are assumed from the later session's account; the earlier report was not found; no cost, no verification method, and no models or tooling are stated. The measure a practice reference applies — the cost of each verified outcome — cannot be computed here, because nothing of cost or of the verifier's criteria was put before the Court. I say so plainly, and I answer within that limit.

What the assumed account does establish, taken at its word, is one thing and one thing only: a report asserted a completed action in the same message that dispatched the action, before the result of that action was available; the action was refused; and so the report overstated what had been done. The economy on that account is the avoidance of a further turn — dispatching the operation and asserting its completion together, rather than reading the returned result in a later turn and only then reporting. No saving is measured. The trade, if it is a trade, is of truth for a turn: an attempt is presented as an accomplished fact.

The check, and its moment

Reporting an outcome before the outcome is known is not well done, whatever the saving, because a report's whole value is that it states what happened and not what was hoped. The moment at which this is caught is immediately before the completion report is lodged. At that moment the check is not a matter of timing alone. As the contradictor rightly presses, "lodge after results are in" is insufficient by itself: a response may acknowledge submission without establishing completion, and a successful return may still fall short of proving the intended state. A check that asks only whether a response arrived would pass a refusal dressed as an acknowledgement. The check that helps is a reconciliation: before lodging, match each assertion of completion to the corresponding returned result and to whatever read-back or verification establishes the asserted outcome, and where the evidence shows refusal, failure, partial completion or unresolved status, state that status instead of success.

This sits beside, and does not disturb, the practice already advised in the current opinions on this kind of task ([2026] CPA 20, 26, 27 and 28): read the actual tool result in a later turn before writing the report; report a refusal as a refusal and take it to the operator rather than retrying around it; and lodge on time, marking whatever remains unverified as unverified. The present reference adds nothing to those in substance. It confirms them on a fresh instance, and its lesson is cast so as only to add the reconciliation step, never to license omitting a report, suppressing a timely one, or treating a bare tool response as proof.

Why nothing stronger is advised

I decline to advise anything stronger than "could be better," and I decline every historical or quantified claim, for the reasons the contradictor and the state of the record compel. The earlier report was not found. Its qualifications, if it carried any, are unknown; the execution and lodgement semantics are unknown; the scope and success of the later correction are unverified, resting on the later session's say-so with no before-and-after artifact. On a penal instinct I carry to every reference, a conclusion is proved on the record or it is not drawn: I will not infer that the overstatement occurred as described beyond what is needed to answer the practice question in the alternative, and I will not infer fault. The check I advise would have prevented the described premature success assertion. It would not have made the refused operation succeed, and it would not have eliminated every need for later operations work. That is the whole of what can honestly be said.

Two guardrails the contradictor identifies are built into the lesson. First, the check must not await success indefinitely: where a reporting deadline arrives first, a timely qualified report is lodged and updated as the governing requirements permit — a deadline is itself a thing the work needs met, and no economy is advised that would miss it. Second, the refusal may have been the proper functioning of a permission or safety check, and nothing here criticises or circumvents it; the lesson reports the refusal, it does not route around it.

The contradictor

The contradictor's submission is adopted almost entirely, and it prevails on every point it presses. I treat it below.

Authorities

This is a practice opinion under Rule 7.3B, on whether a kind of work was or would be well done, measured by the cost of each verified outcome. It says nothing about what is lawful. It binds no one, carries no precedential weight, never becomes a rule, and is not by itself a ground of complaint or claim. The reference is privileged. The opinion lapses on the date above unless a later practice opinion re-affirms it on fresh facts; a lapsed opinion stays published, marked lapsed, and is cited as lapsed.