docs: editorial pass on README, write-up, and About bio

[savvides]

Summary

Editorial polish across the prose-heavy recent additions. Three commits, no code changes.

Commit 1 — first editorial pass (ba4b664)

• README — Dropped the duplicate "An interactive data story" lead. Replaced a "fresh eyes" cliché in the 2026 Retrospective.
• HOW-IT-WAS-MADE — Same "fresh eyes" cliché in the Realization section. Tightened "What I realized was that I didn't need to know..." → "I didn't need to know...".
• index.html — Switched the About-Philippos footer bio from third person to first person to match the coda's voice.

Commit 2 — humanizer pass (b00a979)

• README lead — Replaced the performative reveal "The answer was surprising:" with the answer itself, and trimmed "uncovered an ecological fallacy hiding in" to a plainer "the presence-learning correlation is an ecological fallacy."
• README retrospective — Dropped "rock-solid" and the passive "could be made much stronger." Removed the "Key issues:" announcer.
• HOW-IT-WAS-MADE realization — Removed the duplicated "hiding in" pattern and replaced "a genuine discovery, buried in an eight-year-old dataset" with "a real finding, eight years late."
• About bio — Broke a rule-of-three list and replaced "My work looks at how" with a direct verb plus a doublet.

Commit 3 — bio removal (1f67113)

• index.html — Replaced the inaccurate "About Philippos" bio paragraph with the author's name as the column heading. The LinkedIn link remains. Better an empty column than a wrong claim.

Documentation health

• README.md — Updated (lead + retrospective)
• HOW-IT-WAS-MADE.md — Updated (Realization + Process tightening)
• index.html — Updated (footer About column)
• CLAUDE.md — Current
• CHANGELOG.md — Current (no new entry; matches prior convention for prose-only fixes)
• VERSION — N/A (no VERSION file)

Test plan

• Render the README on GitHub and read the lead from top to bottom — confirm the doubled "interactive data story" phrasing is gone and the answer ("They don't.") lands cleanly.
• Open the deployed site, scroll to the footer, confirm the About column now shows just "Philippos Savvides" + LinkedIn.
• Skim HOW-IT-WAS-MADE for any remaining clichés or run-on phrases.

🤖 Generated with Claude Code

[gemini-code-assist]

Code Review

This pull request refines the copy across several documentation files and the main index page, improving clarity and shifting the 'About' section to a first-person perspective. A review comment correctly identifies a technical inaccuracy in the README, suggesting the use of 'similar' instead of 'identical' to describe a small effect size (Cohen's d = 0.23) for better scientific precision.

[gemini-code-assist]

The term 'identical' is technically inaccurate when describing a Cohen's d of 0.23, which represents a small effect size rather than zero difference. Using 'similar' or 'comparable' would be more precise and scientifically accurate.

Suggested change

The 2018 PhD study (n=108) asked whether VR headsets improve learning. The answer was surprising: the simulation matters, not the headset. VR and desktop produced identical outcomes (Cohen's d = 0.23) while both massively outperformed passive instruction (d > 2.4). A 2026 reanalysis uncovered an ecological fallacy hiding in the original results — a correlation that looked real at the group level vanished within each condition.

The 2018 PhD study (n=108) asked whether VR headsets improve learning. The answer was surprising: the simulation matters, not the headset. VR and desktop produced similar outcomes (Cohen's d = 0.23) while both massively outperformed passive instruction (d > 2.4). A 2026 reanalysis uncovered an ecological fallacy hiding in the original results — a correlation that looked real at the group level vanished within each condition.