A useful weekly brief is mostly an exercise in restraint. Far more crosses the desk than is worth a reader’s attention, and the discipline is deciding what not to publish. This note sets out how that decision is made, so readers can hold the publication to it.
What qualifies as a signal
- It touches critical or communications infrastructure, or the capability development that precedes an infrastructure threat.
- It shows movement — across sectors, across regions, or from a proving ground toward a strategic target.
- It can be grounded in open, attributable evidence and, where relevant, cross-referenced against The Observatory.
Verification before publication
Nothing is published on a single unverified claim. A development needs corroboration from independent open sources, and contested attribution is presented as contested. We do not scrape or republish third-party headlines, and we do not launder an unreviewed summary into an assessment. Where a primary source exists — a court filing, an advisory, a dataset — we link to it directly rather than to coverage of it.
The most important patterns sit between the silos. The adversary works there on purpose — so the reading has to live there too.
Confidence tiers
Every material claim carries an explicit confidence level, and the boundary between confirmed, assessed and inferred is kept visible on the page. A reader should never have to guess whether a sentence is a fact or a judgement. When the honest answer is that we do not yet know, that is the sentence we write.