Comment by jimrandomh

Seen in light of http://www.slate.com/articles/news_and_politics/jurisprudenc... this looks like a systemic problem and not an innocent mistake. If crime labs were doing science, they would have run enough control samples to quickly discover the mistake. But the error "may have affected thousands of cases going back to 1999". I don't think forensic science, as currently practiced in the United States, deserves the name "science". And the errors are all in the same direction: towards conviction.