They fail with a 403 Client Error, perhaps due to our once-per-day request being classified as scraping.