Latest edition
Blog · 267 analyses
Context for better decisions.
Analysis on AI, cybersecurity, regulation, governance and digital systems. With sources, practical consequences and one clear next action.
All analyses
217–228 of 267
Current knowledge, organised by topic, date and practical consequence.
The second beta was published on 16 July with fixes across vacuumdb, temporal tables, logical decoding, SQL/PGQ, autovacuum, and other areas. The project advises against beta in production. The right use is a faithful, comparable, and reversible rehearsal.
Read analysis
Availability, latency, and cost show whether a service responds. They do not establish whether the function remains correct, people understand the output, misuse has emerged, an applicable obligation is met, or cumulative effects have changed.
Read analysis
A manufacturer will have to state how long it handles vulnerabilities, justify that horizon in technical documentation, and align product, components, updates, and support with the promise made before purchase.
Read analysis
A document can list the right contacts and steps yet fail at the first authority conflict, third-party dependency, unavailable channel, or decision made with half the facts.
Read analysis
The control is not a decorative second signature. It is the boundary preventing one identity from initiating, legitimising, executing, and hiding an irreversible action.
Read analysis
A uniform rate cuts cost and visibility together. A useful design preserves end-to-end context, errors, latency tails, and critical services while measuring what it loses.
Read analysis
Leaderboards are useful for comparing systems under known rules. They become dangerous when they replace the question that matters: does this system improve our process, with our data, without creating an error we cannot detect or afford?
Read analysis
The worst way to manage a regulatory deadline is to treat it as the opening of the project. On that day, processes, data, decisions, suppliers, and evidence should already work together.
Read analysis
Teaching people is necessary. Making them the primary control against convincing pages, adversary-in-the-middle MFA, and approval fatigue is an architecture decision disguised as awareness.
Read analysis
Two hundred pages of intent do not prove that access was revoked, an exception expired, or a supplier was reassessed. Governance improves when audit values sufficient, reliable, relevant, and useful evidence.
Read analysis
Neither always ship nor always freeze. A mature team decides how much reliability risk it may consume and changes behaviour when the service exceeds that limit.
Read analysis
A tool list is not enough. Classification depends on intended purpose, the function actually performed, the context of use and, where relevant, Article 6's two routes: regulated product or an Annex III use case.
Read analysisHow we publish
Every analysis links the official documentation it relies on.
We translate change into decisions, owners and evidence.
We update the analysis when facts or guidance change.
We disclose generated photography and credit documentary photography.
From context to execution
We map the problem, write the scope and build the solution with acceptance criteria.
