I built YearToBeat, a lightweight daily browser game to test chronological memory. You watch 5 short music video clips and guess the exact release year.
I expected people to be fairly accurate, but after collecting ~18,600 data points from players, I found a bizarre statistical anomaly that completely breaks our assumptions about memory.
The Data: People’s chronological memory of the 1980s is razor-sharp (guesses form a tight bell curve with a standard deviation of ~1.5 years). But the moment we cross into the late 90s and 2000s, our collective memory collapses. The standard deviation doubles to 3.5+ years. I graphed the anomaly this morning, and Reddit is currently having a massive debate about why "The 2000s Blur" exists.
The Theories: The death of the TV/Radio monoculture (Napster and the iPod destroyed our shared chronological anchors). The "HD Plateau" (Video and audio production homogenized around 2005, removing visual cues). We use the 80s as a rigid cultural reference point, but treat the 2000s as "just a few years ago."
The Gauntlet: The Reddit crowd is currently getting destroyed by this temporal blindspot. I’m curious if the HN analytical brain is actually any better at pattern-matching recent history. I bet most of you will confidently nail the older tracks in today's puzzle, but miss the post-1999 tracks by at least 3 years.
Tech stack: Go backend with a "old school approach" index.html with CSS and JS. No ads, no logins.
Would love your technical feedback on the UI, or your sociological theories on why our memory broke after the year 2000.