Why I built it
Data center fights are decided town by town, and the terms that got projects approved are scattered across minutes, ordinances, agreements and agenda packets. I built GridFriction to put those decisions, and the documents behind them, in one place: what governments said no to, and what they required when they said yes.
How the records are built
- Decisions are found through systematic searches of local government records and local news, refreshed each month.
- For each approval I retrieve the government documents, code each condition as a value with a unit, cite the page, and quote the operative words. Quotes are checked automatically against the text of the cited page.
- Proposed terms are kept separate from adopted terms, and each record states how well it is verified.
- I use AI tools to search, extract and check at scale; the counting rules, the review and the decisions about what counts are mine.
Corrections
When a document shows a record is wrong, I fix the record and log the correction on the methodology page. If you see an error, write to contact@gridfriction.com.