High friction filters out everything except serious intent. Low friction lets weird ideas breathe.
When cost of experimentation collapses toward zero, the bottleneck shifts from can I build this? to what’s worth building? Play emerges. A joke becomes a product (Fable: pet Renaissance portraits — nobody would greenlit it from a strategy meeting; internet had demand anyway).
Software is experiencing this now. Photography had it with smartphones: dark rooms and expertise gave way to a vast amateur ecosystem. Professional photography still exists. But now there’s room for everything else too.
Implication: Building bias toward serious, validated, de-risked projects is rational when friction is high. When friction is low, rationally biased toward experimentation. Same people, different optimal behavior.
For vibe coding specifically: The threshold that just dropped isn’t just model quality — it’s context length + agentic maturity + stable builder platforms. Those three together made it finally feel like play, not babysitting.
“Three things now true simultaneously: building is inherently satisfying + internet has infinite demand + cost of hobby-scale experimentation approaches zero. Something new emerges from that conjunction.”
See also: 2026-04-11-software-vision-is-the-new-bottleneck for the next constraint that matters.