Two in five notifiable breaches involve nobody attacking anything, and sit outside everything this database has to say.
Eight per cent, in a scheme with a stable definition and eight years of continuity, is the most credible year-on-year change in this database.
A patient chooses a practice. The practice chooses the software. The patient has no relationship with the supplier and cannot change it.
A country that strengthens its disclosure law appears to get worse. One that has none appears clean.
Absolute counts describe where the most people are online. That is not the question these tables are read as answering.
Its precision is an artefact of counting, not evidence of accuracy. The buried sentence is the reliable part.
A dark-web listing is read by a professional audience. A Telegram channel is read by neighbours and employers.
Half the compromised accounts in Asia belong to people about whom this corpus can say nothing further.
A defaced page and a five-week production halt appear identically in an incident tally.
Without it, this corpus would have no basis for saying anything about Africa at all.
If this corpus’s assumptions came from markets with a desktop phase, they may describe the minority condition.
The closest thing in this database to a measured downstream consequence — and it still is not the proof.
This database has 350 files and none of them cover a stolen phone — because it produces no notification and no defendant.
Espionage succeeds by producing no artefact. The ones in this corpus are the ones that failed at the last step.