Newest first, grouped by year. Search or filter by topic.
The series retrospective - what the AI-era evidence actually says with every figure dated and attributed, the nine characteristics of ISO 25010 as revised in 2023, and what I now gate differently when much of the diff was machine-written.
Eleven numbers that still haunt engineering dashboards - what killed each one, from Goodhart to queueing theory to changing fashion, and the one-line replacement carved on every headstone.
The team, the team lead and the executives need different slices of the same truth - what each audience should actually see, the numbers I put in front of a boardroom, and dashboard principles that skip the red-amber-green pantomime.
What Azure DevOps will measure for you and where it stops - the two clock definitions, the reactivated-item rule, how to read a cumulative flow diagram, and why Microsoft's own guidance points past its widgets to OData.
Why no single metric captures developer productivity - the SPACE framework's five dimensions in plain terms, the measure-at-least-three rule, and what I do about the two dimensions no dashboard can hold.
The queue mathematics underneath delivery - one equation from 1961, Scaled Agile's six flow metrics rewritten in plain terms, and why the biggest delivery gains hide in the waiting rather than the working.
Measuring the lines between the boxes - afferent and efferent coupling, the instability ratio, Robert C. Martin's main sequence, and the cohesion metric that catches a class doing two jobs, plus where every one of them needs a human to overrule it.
The composite score that hides more than it shows and the percentage that translates code quality into money - where the maintainability index formula came from, why its logarithms flatten real problems, and how the technical debt ratio earns its place as a trend.
Two complexity metrics that look interchangeable and answer different questions - McCabe's 1976 path count is a testing instrument, SonarSource's cognitive score is a readability instrument, and only one of them earns a place in the merge gate.
Dashboards go green while codebases rot - the opener of a fourteen-part series on measuring code quality and software delivery without staging the theatre that ruins both.
DShop wires Serilog, App.Metrics and Jaeger through one shared kernel, propagates a trace span on the message envelope, and hangs its Consul health check on an endpoint no controller defines. A tour of pre-OpenTelemetry observability.
