A few days ago I posted a link to Roger Koppl's column in Forbes. Roger has read the Comments you posted and has sent me this response:

Hi Todd,

I've been enjoying the comments on your post quoting my Forbes article. "JCCamp" warns against "ignoring the investigative assistance an in-house lab can provide for law enforcement." You bet. Good principles will give you bad results if you don't recognize the relatively intricate design problem involved. The example JCCamp raises is relatively easy to handle by clever task separation: The experts who do blind testing should be different from the "case manager" who has all case information and shields domain irrelevant information from the forensic scientists doing the tests. A group of us has worked out a template for doing that in DNA profiling. Our template will be published in the July Journal of Forensic Sciences as a letter to the editor ("Sequential Unmasking: A means of Minimizing Observer Effects in Forensic DNA Interpretation"). I would put J. F. Thomas's concerns about chain-of-custody in the same category. It's a huge issue, but it is not really that tricky to handle. My institute's website addresses lots of these issues, especially on the research page.

My Forbes article doesn't address sequencing. If you want a set of improvements that sticks, then you need to need to put them in place in the right order. An improvement may lose its effect over time if the affected parties learn compensating behaviors. A substantive right of defendants to forensic science expertise, however, would be self-sticking because the defense experts would work to maintain their place in the system. The simultaneous existence of defense and prosecution experts would apply pressure needed to make other changes stick.

Cheers,

Roger