Quote Originally Posted by hacker View Post
We should be. I think it's dumb to dismiss a powerful tool at our disposal just because some mistakes have been made. I'm pretty sure the mistake in the article you linked (and similar mistakes) represents a fraction of a percent of overall tests, which means we could contact trace 99.9% or more cases correctly. There's always going to be errors when the scale of an operation is as large as this one, but it's not statistically relevant.
What good does that do you when the information is wrong?