DNA sketch leads to suspect confession in Texas slaying | Hacker News Reader

DNA sketch leads to suspect confession in Texas slaying

seattletimes.com

30 pointssmaili24 comments

Open articleView on HN

This looks like something that is just begging to misinterpreted and abused by investigators and prosecutors.

>Opaque technology with an unknown error profile? Check.

>Broad search criteria that guarantee a high false positive rate? Check.

>Evidence that requires careful consideration of prior odds to properly assess? Check.

Imo, the main ingredients here were the low population density and the suspect's own sense of guilt/fear.

I'd be curious to see a comparison between the accuracy of this new technology and good old-fashioned eyewitnesses (who are working from memory) and hand sketches done by a human artist. I wouldn't assume that this is less accurate than the human approach.

Eyewitnesses are potentially quite unreliable and the reliability can be heavily influenced by the stress level of the situation, relative racial experience of the witness (It's harder to recognize/differentiate members of an ethnicity you don't spend much time around), interview techniques and line-up methodology.

Yet eyewitness testimony is generally treated as highly accurate and some places do not allow expert witnesses who can educate the jury on the factors that can influence the reliability of eye witness testimony.

Eyewitnesses can easily be confronted and cross examined. Questioning eyewitnesses is a relatively common skill among defense attorneys, and the arguments being made in such cross examinations are easily grasped by jurors.

The algorithm behind something like this might not even be made available to defense counsel. If it is, examination will require an expert witness (which costs $$$$), and it's doubtful that 12 average citizens would be able to assess the validity of or even comprehend the arguments.

If there's enough DNA for a DNA sketch, then there pretty much has to be enough DNA for a regular DNA test. Nobody is going to go to jail for looking like a DNA sketch. There is a risk here where innocent people with no connection at all to the crime might be compelled to give a DNA sample. If the DNA from the "suspect" is tested, excluded, and then all information about that DNA is then destroyed, maybe that's not a big deal. But I don't think we have that guarantee now for DNA samples.

As someone who is often mistaken for other people, I'm not looking forward to this.

How is it any different than being misidentified by a witness's description?

Because people can discredit a witness. But with this they say "DNA evidence" and the jury is already deciding.

It makes no sense for the jury to look at the sketch when you have the DNA from the crime scene and can directly compare it to the DNA from the accused person.

In other words, if you have what you need to produce this sketch, once it goes to trial you have everything you need to make a much more authoritative determination of either their guilt or innocence.

Don't worry, once we have an AI in charge of the legal system, it will know the stats of DNA identification. It's much less likely that you'll be liquidated without cause.

> It's much less likely that you'll be liquidated without cause.

Well I'm sure the AI would prefer behavior modification devices to liquidation as humans are too expensive to grow from scratch.

But what's their marginal value? Fewer could be better, overall!

We could implement a human screening program that keeps only the humans of value! This program needs a name, I say we name it after an informal grouping of preditory members of Telluraves

I don't get it, nobody needs to be discredited, peoples memories and descriptions are imperfect to begin with. All this is doing is providing a sketch that might vaguely look like the suspect, and since they already have that persons DNA, it should be easy to rule people out who get falsely identified, so no jury would be involved yet.

I'll believe it if there is an actual DNA match between the man and DNA found at the scene.

Am I the only one that doesn't see a strong correlation between the sketch and the actual suspect, other than "generic white guy"?

Does anyone know what the likelihood of is of developing algorithms to do this without relying so much on statical modeling?

Shouldn’t it be possible in principle? I realize there are other factors that contribute to hoe we look, even what we eat, the environment around us, etc. But even if theoretical best rendering of just DNA was achieved I’d imagine it would be very impressive.

So basically, the people who look like "generic white guy" or whatever are going to get hounded by the police. Great.

At least they can probably escape attention by gaining a lot of weight, growing a beard, or smoking and drinking lot to cause skin damage.

Super afraid for this technology to be applied to fetuses. Parents can't decide on an abortion? They can check if they think he or she will be ugly or not.

If the parents decide to abort based on a DNA prediction of ugliness, that kid would have a bad life either way. Those are terrible parents, and that would be such a rare scenario it seems basically irrelevant.

Imagine if you could pay a sum of money to have a certain number of eggs harvested and implanted, then measured to determine fitness over: - Likelyhood of diabetes/cancer - Height - Intelligence - Mental Illness

How strange would it be to add attractiveness to that list? Is it really being a bad parent to pick out the best attributes for your child?

There are a ton of embedded assumptions in this question, and I disagree with pretty much all of them. It wouldn't be that strange to add attractiveness to the list, but you're starting from a very already strange place. It's good parenting to pick the best for your kids. But that's different from picking your kids.

I disagree that it's "very strange" or even strange at all to "pick your kids", setting aside adoption.

Choosing which sperm is fittest is something that's done unconsciously (by the egg) and consciously (by choosing a partner), and selecting them through technology, (such as DNA sequencing or abortion) is just another method of accomplishing the goal that is fundamental to reproduction. So I disagree with the idea that it's strange to "pick your kids" - although that doesn't address moral issues.

As far as moral issues go, some people want to outlaw abortion of babies with Down Syndrome. But in any case, it's common to do that (in some places more than others), hence why people propose to stop it.

Eugenics is a bad word. My own thoughts are that while some people say "eugenics" and mean the selection of desirable genetic characteristics by anyone and any means, I think that is an untenable definition. I find it more useful to use "eugenics" to mean the selection of genetic characteristics by someone other than the parents, an external authority that "knows best". That is the sort of eugenics that I think ought to be guarded against, primarily.

Relax. They'll do this at the embryo stage.

Reply on news.ycombinator.com