Nine Dimensions, Zero Facts: A Report Complete in Form, Empty in Substance

I opened the file at 2 a.m. Nine analytical dimensions, each with its own table, a risk matrix, confidence levels, and a glossary at the end. On first read it looks like a full post-match dossier. Every cell returns the same sentence: insufficient information, cannot assess. No title, no source, no list of information points. The scoresheet is immaculate. No match was ever played.

Eleven years in marine insurance compliance in Liverpool taught me to notice when a clause bites. That habit asked three questions before I read a single table: which clause, whose jurisdiction, and who carries the burden of proof.

The document is an information-integrity event. Two stages of a pipeline produced it. Stage one was meant to break the source article into title, outlet, article type, numbered information points and named entities. Stage two was meant to build a deep analysis from those points. This time stage one returned an empty template: title N/A, source N/A, information points blank. What stage two did next is the subject here — it declined to lie.

The two blank fields are the loudest evidence. Title and outlet are the two things any parser can lift from any document. When both read N/A, the failure sits at ingestion, not at analysis. There is a second fingerprint. The 'Additional Notes' field repeats the extraction instructions instead of reporting findings. A machine returned a default mould, not a completed reading.

Then the burden of proof. On 16 June 2026 in Kazan I watched Andrés Cunha wave away an Antoine Griezmann tumble against Australia, walk to the monitor, and reverse himself — the first VAR-awarded penalty in World Cup history. In that moment the on-field decision stopped being a decision and became a hypothesis. Here nothing of the sort was available, because nothing happened on the pitch. A referee who did not see an incident does not award a penalty; he says he did not see it. This document did exactly that, nine times over.

Traceability is the clause that held. Every conclusion had to be returned to a numbered stage-one point. That sounds bureaucratic; it is the same rule courts call the chain of evidence. With zero points, zero conclusions. The risk matrix then rated one row High — not injuries, not wage load, not financial rules, not the loss of a key player, all of which returned insufficient information. The High row read: analysis produced on an empty payload would be fabricated, likelihood certain, already occurred. The document indicted itself in the clearest available language.

The conventional reading is comfortable: the tool failed, run it again. That points the finger in the wrong place. Returning insufficient information on an empty payload is the pipeline's best work. The failure that would actually hurt is a confident 1,300-word piece with a transfer fee, a sell-on percentage, a weekly wage and a release clause arranged in smooth prose. In a transfer window a fabricated number and a real number wear the same clothes. Neither reader nor editor can tell them apart at a glance.

A second error sits further upstream. The brief attached to this document asked for a blockchain news report to be built from a football analysis that itself contained no football facts. That is asking a referee to apply the handball law to a cricket match — wrong rulebook, wrong jurisdiction. Two layers, two kinds of confusion: no data underneath, no subject above.

What a reader can actually use from this is a filter. Grade a transfer rumour by outlet tier first, then by the named journalist, then by the agent's motive, then by whether any contract detail is quoted — length, fee structure, add-on, sell-on. If a claim carries none of those, it is not a rumour, it is atmosphere.

The next step needs basic facts, not deeper analysis. Before any re-run: title and outlet; publication date and the date of the event described; article type — match report, transfer news, tactical analysis, governance, finance or opinion; at least three to five numbered information points; and the names of the clubs, players, coaches and competitions involved. Depth should match article type — a one-line transfer brief deserves a short factual note, not a nine-dimension audit.

And the null output should not be deleted. Keep it as a regression test, so the next time empty data arrives you can confirm the pipeline really does go quiet.

How many pages published this week are complete in form and empty in substance? In a transfer window, that question is not theoretical.