I agree, a real world test is key. But as structured, the test was doomed to fail. It would need more groundwork & education, probably A/B testing of signs to

I agree, a real world test is key. But as structured, the test was doomed to fail. It would need more groundwork & education, probably A/B testing of signs to eliminate min/max confusion (e.g., would a sign saying "Speed only 65-75mph" be better than the "Speed 10mph minimum"), TV public relations campaign and even advertising about the test, etc. Maybe some of this happened, but it wasn't evident from the article