Live data from Hacker News

Forensics Gone Wrong: When DNA Snares the Innocent (2016)

science.org

91–100 of 119 posts

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#91
post #56

DNA evidence is scary because we have decades of experiences and milestones saying that it isn't as simple and clear cut as we think it is, and still we treat it is hard, inarguable truth. Even though history teaches us we should know better. It is one of those things that reminds me that most people with a science degree do not actually practice science.

The alternative being?...

Just because we don't currently have an alternative doesn't mean we should keep using a flawed methodology.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#92
post #2

Evidence handling and testing should not be under the control of police, or even prosecution. Incentives need to be aligned toward justice, not convictions.

Well, if you don't think it should be under control of police or prosecution, who do you suggest needs to do it? The defense? A third party private contractor like Theranos? Who's going to be the ones validating whoever is decided can do the testing?

everything you listed is still better than giving it to the cops. "these random examples sound bad" isn't really a good argument for doing nothing.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#93

Earlier quoted context omitted.

Historically, forced religious confession has been used to get the subject to confess to heracy, and then the subject is executed as punishment. All in hopes that the subject will become a better person, I assume.

This is 2024, not 1492.

And? Human psychology hasn't changed.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#94
post #24

A lot of what the article is saying makes sense, but I really wish the reporter had pushed a bit harder on Hampikian. He conducted a study to show that mixed DNA analysis can falsely exclude people, but are there similar studies that confirm the complex strategies he's exploring in the Tapp case can't falsely exclude people? "Civil rights advocates object to [familial DNA] because it means that simply being related t…

It doesn't have to be 1-to-1. You can have different standards for offensive vs defensive use of the information. Showing the DNA matches some relative of a different family could be enough to generate reasonable doubt while still being too weak to generate probable cause to arrest or even search others. Frankly, they had no probable cause to compell a DNA sample from that guys son. Probable cause is supposed to be t…

I guess that's kinda my point. The Innocence Project should be one of the forefront groups opposing such fishing expeditions. Instead, the article says, the head of the Idaho Innocence Project suggested testing Michael Usry based only on a familial match, because a positive result might have helped get their client Christopher Tapp out of jail. I can't make sense of that, and it certainly doesn't seem consistent with principled caution about DNA forensics.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#95

Earlier quoted context omitted.

The inquisition was effectively a judicial body. They were religious representatives and burned witches for questionable reasons, but their… questioning did force quite a few false confessions if I recall correctly. This has little to do with your Sunday Priest.

See above. What people did 400 years ago is not relevant to my point about how things are today.

It is. Because the point GP / OP are making isn't that we are doing inquisitions today, but rather that the human psychology behind confessions has long been wielded as a tool.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#96
post #24

Earlier quoted context omitted.

It doesn't have to be 1-to-1. You can have different standards for offensive vs defensive use of the information. Showing the DNA matches some relative of a different family could be enough to generate reasonable doubt while still being too weak to generate probable cause to arrest or even search others. Frankly, they had no probable cause to compell a DNA sample from that guys son. Probable cause is supposed to be t…

I guess that's kinda my point. The Innocence Project should be one of the forefront groups opposing such fishing expeditions. Instead, the article says, the head of the Idaho Innocence Project suggested testing Michael Usry based only on a familial match, because a positive result might have helped get their client Christopher Tapp out of jail. I can't make sense of that, and it certainly doesn't seem consistent with…

And my point is that it's different if a defendant subpoenas you to create reasonable doubt vs the government issuing a warrant without true probable cause.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#97

Earlier quoted context omitted.

Well, if you don't think it should be under control of police or prosecution, who do you suggest needs to do it? The defense? A third party private contractor like Theranos? Who's going to be the ones validating whoever is decided can do the testing?

everything you listed is still better than giving it to the cops. "these random examples sound bad" isn't really a good argument for doing nothing.

random examples are necessary because we were walked to the edge of the cliff with no actual solution provided, so hyperbole felt like an appropriate response to encourage moving the train further down the tracks.

it's not a bad idea, but it's obviously ripe for just an additional layer of bureaucratic waste. clearly, there is a lack of trust that the police could handle evidence properly. so the natural follow up is who is trust worthy? how would it work? cops are called to the scene, but are then only there to "secure it" until some 3rd party comes along to collect evidence? what's the purpose of a detective at that point? we're not just shifting that role from a police agency to some 3rd party. are we going to re-establish the Pinkertons? if the detectives are no longer a role in the police, then more than likely those that would be attracted to that role would just join the 3rd party instead. so you haven't solved anything other than these people are no longer part of the police but are still involved.

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#98
post #56

DNA evidence is scary because we have decades of experiences and milestones saying that it isn't as simple and clear cut as we think it is, and still we treat it is hard, inarguable truth. Even though history teaches us we should know better. It is one of those things that reminds me that most people with a science degree do not actually practice science.

No, the scientists behind DNA testing absolutely understand the limitations.

The people who don't understand the limitations of science and possible errors in certain methods are clueless jurors and the cops who pay a random person $800 to sit on the stand, say they are "and expert" and claim there's a trillion to one odds that two people could have "the same" DNA, even though that isn't even remotely what was tested!

Courts allow basically anything as long as you can pay a guy to say they are an expert and parrot whatever you want. But somehow that's treated as if it's the fault of biologists and others who do DNA analysis?

Maybe our courts shouldn't be based on "Trust anything a cop says, period"

Re: Forensics Gone Wrong: When DNA Snares the Innocent (2016)

#100
post #42

> [...] he confessed after a series of lengthy interrogations that several experts have described as coercive. Police found plenty of male DNA at the scene, and it did not match Tapp's. But the prosecutor and jury believed his confession. This story isn't even about DNA evidence.

It's true. People reliably confess to crimes they did not commit. It's just another dead canary in the cages of our justice system.

[dead]
Post reply on HN