Methodology · version 1.0

HOW WE MEASURE
THE THREAT.

QDayWatch separates observations, estimates, scenarios, and editorial assessments. The distinction is the product.

We do not predict Q-Day.

There is insufficient public evidence to produce a scientifically defensible exact date. QDayWatch never turns a vendor roadmap, a qubit count, or a single research estimate into a countdown.

What we track instead

Logical error rates, code overhead, fault-tolerant operations, algorithmic improvements, cryptanalytic resource estimates, standards, deployments, and transition requirements. Each is useful on its own terms; none is silently converted into an arrival date.

X + Y > Z is a planning test, not a forecast.

X is how long vulnerable data must remain secret. Y is the time required to migrate. Z is a hypothetical time to a cryptographically relevant quantum computer.

The terms X and Y add because data may continue to be created throughout migration. Data protected on the final day before the transition completes can remain sensitive for another X years.

Result states

When X + Y exceeds Z, the modeled exposure window is open. Equality is shown as a zero-margin threshold—not as safety. Only X + Y below Z produces modeled migration margin. Z is always labeled as a scenario selected by the visitor.

Observation is not relevance.

Laboratory demonstration

A result performed under disclosed experimental conditions. It can be significant without being scalable, general-purpose, or cryptanalytically useful.

Resource estimate

A model built from assumptions about algorithms, error rates, cycle times, connectivity, decoding, and architecture. We show dates and assumptions because estimates can change sharply without hardware changing at all.

Cryptographically relevant

A capability sufficient to attack a real production key at meaningful scale and reliability. A toy factoring demonstration is not classified as cryptographically relevant.

Physical qubits are not a common currency.

Architectures have different error rates, connectivity, gate sets, cycle times, and logical encodings. QDayWatch does not rank machines by raw physical-qubit count or combine incompatible vendor metrics into one chart.

When a number is shown, its architecture, observation date, source, and relevant uncertainty travel with it. Qualitative status indicators are explicitly labeled and are not a hidden composite score.

Uncertainty remains visible.

  • Primary sources are preferred over commentary.
  • Publication date and event date are stored separately.
  • Vendor roadmaps are labeled as plans, never achievements.
  • Withdrawn and deprecated standards remain in the historical record.
  • Material corrections are reflected in the source record and review date.