pgBadger Duplicate Query Count Testing: Four Parser Runs
A 20-execution synthetic ledger produces 20 or 28 log records. Four actual parser runs compare pgBadger 13.2 and 13.3, with repeated SQL and a multiline control.
Cloudflare Workers request.signal: Test What Actually Stops
Eighteen local Node simulations separate a disconnect notification from downstream cancellation. Includes three charts, a lifecycle infographic, four CSVs and runnable code.
Safari Adaptive Dark Mode: Test the Rendered Page
Separate authored colours from rendered appearance. Includes a six-component fixture, 18-observation worksheet, real authored-theme screenshots and clear untested Safari states.
CI Checks Pass Without Testing Files: Test the Checker
Eight executed local comparisons expose masked errors and empty file selections. Includes raw receipts, code, four charts and an evidence-contract infographic.
GitHub Actions Brownout Nigeria: A WAT Readiness Audit
Check the announced 9 and 16 November brownouts in WAT. Includes six fictional repository cases, a blank audit ledger and a decision tree that keeps policy blocks separate from code failure.
Turning Real Business Problems Into AI Projects
Identifying Recurring Business Problems Begin by observing problems that appear repeatedly during normal business activities. Recurring problems often…
Learning to Code Without a Computer Science Background
Why Formal Computer Science Study Is Not Required Learning to code begins with learning how to express instructions…
Firefox DevTools MCP Testing: Separate Session Failures From App Bugs
A connection failure does not establish an app bug. Use our five-scenario Firefox DevTools MCP protocol, unexecuted 15-run ledger, diagnostic labels and issue-report worksheets to record the evidence before assigning blame.
pgAdmin 9.19 Export Testing: A UTF-8 Fixture Pack for Nigerian Names
Test whether your downstream reader preserves text, identifiers and null distinctions. Use 12 fictional fixtures, 36 local Python comparisons and an unfilled pgAdmin export ledger, with their evidence boundaries clearly labelled.
Product Verification Hackathon Projects: Test Record Matches, Not Authenticity
A database match does not authenticate a physical item. Explore our 20-case fictional record-matching exercise, executed Python results, five research figures and reusable claim-evidence tools.