A ten-year moat of barcodes

Julia’s Shipaton candidate comes from her own adolescence: Desrotulando, the Brazilian app she used at the start of her vegetarianism ▸ 9:09. Scan a product’s barcode and the indecipherable ingredient list becomes a plain-language verdict, additives flagged, sugar flagged, a 0-100 score, alternatives suggested ▸ 10:26. Missing product? You photograph the label and the community pipeline verifies it within days ▸ 12:18. She wants to build it for Colombia.

Before opining, they research, ChatGPT voice mode on speaker again, and the infrastructure lesson lands: barcodes are governed by GS1, which assigns each company a unique prefix ▸ 15:05, but, the crucial clause, there is no central repository of products; each company keeps its own records ▸ 15:37. The identifier is global; the knowledge behind it is nowhere.

el código es público; la base de datos que lo explica es el negocio →

Which reframes the competitor analysis exactly: Desrotulando’s value was never the scanner, a weekend of code, it’s the decade of community-verified labels behind it, “tendrías que entrar a competir con una base de datos de 10 años” ▸ 17:27. The same shape as Anki’s decks, as Reddit’s threads, as every product in this diary where the software is fungible and the accumulated contributions aren’t. The encouraging corollary goes unstated but is real: in Colombia that ten-year database doesn’t exist yet, which means the moat is unclaimed, and whoever starts filling it first gets the decade…