Applause is available immediately. Evidence takes years. Choose carefully which one you're optimising for.
I want to make an argument that will sound obvious and is, in my experience, rarely practiced: public credibility should be earned in the order that work actually happens, not in the order that makes for a good announcement.
Applause is available immediately. Evidence takes years. Choose carefully which one you're optimising for.
Everything I've built since this speech has tried to follow that order — pilot before press release, evaluation before expansion, documentation before description. This archive itself is an attempt to hold that standard in public, permanently.