Essays

AI Ethics in the Supply Chain: Training Data Provenance Problem

Tracing training data lineage for 3 models revealed none could document full provenance. One model included 12 sources with no consent chain. AI has a data supply chain problem.

Jira Has a Philosophy Problem, Not a UX Problem

Jira attempts to be both a coordination tool and a surveillance instrument. Separating these functions reduced cycle time by 28%.

Estimation Is Epistemology: Confidence Intervals

Calibrated teams delivered within 15% of estimates 74% of the time versus 31% for gut-feel. The difference was self-knowledge, not technical skill.

Cognitive offloading and the changing shape of human expertise

Data Literacy Is a Leadership Competency, Not a Technical Skill

A survey of 40 executives found 28 could not explain how their key revenue metric was calculated. Data literacy is a competency for every decision-maker, not just analysts.

The SOPs That Survive: What Makes Them Standard

Studying 34 SOPs across 5 organizations found only 26% were followed. Surviving SOPs were practitioner-authored, under 2 pages, included decision points, and updated within 90 days.

Ethics of Attention in the Age of Notifications

Attention is ethical because the quality of your thinking determines the quality of systems people depend on. Distracted architecture is negligence.

Data Mesh Is an Org Design Problem in a Tech Costume

Of 14 data mesh implementations studied, 11 failed by treating it as a technology pattern. The technology was never the bottleneck; the organization was.

Semantic Layers Are the Missing Piece in Most Data Architectures

A semantic layer reduced duplicate metric definitions from 34 to 1 per metric and decreased discrepancy tickets from 12 to 1 per month. Most organizations need one.

The psychology of intellectual honesty: Why we resist updating our beliefs