What Will We Do With Innocent People's DNA?

NevDull writes "As creepy as it may be to deal with identity theft from corporate databases, imagine being swabbed for DNA samples as a suspect in a crime, being vindicated by that sample, and never even being told why you were suspected. This article discusses a man, Roger Valadez, who's fighting both to have his DNA sample and its profile purged from government records, and to find out why he and his DNA were searched in the BTK case. DA Nola Foulston said, 'I think some people are overwrought about their concerns.' -- convenient as she wasn't the one probed without explanation. The article then mentions that 'In California, police will be able in 2008 to take DNA samples from anyone arrested for a felony, whether the person is convicted or not, under a law approved by voters in November.' What will be the disposition of the DNA of the innocent?"

Comments

Cluster and Classify ...

[foobsr]

We will end up with two categories of samples:

• convicted
• not convicted (obviously)

Do some analyses to enable you to categorize from an unlabeled sample.

<cyn> Imagine how useful that could be!</cyn>

I think some people are overwrought about their concerns.

Yes, I am.

CC.

It's just data...

[zecg]

I don't think it's really about samples - the man hardly needs his skinflakes or his hair bits back and he sheds it all around anyway. As for the data it represents? Why, "we" keep it forever, of course. He is just the first in line, I'm willing to bet that within 20 years "we" will have a database of DNA samples from all "our" citizens - or whoever accepts my bet wins my slightly weathered tinfoil hat.

Re:Nothing to Fear

[Mistlefoot]

everyone is always a potential suspect.

What of the poor sap who has an affair with someone who happens to get raped/murdered on her way home.

That his sperm has been found in her body and definitly matches means he's guilty?

How do you prove you had consentual sex with a now dead women. There are many such instances were the DNA found at the scene does not mean guilt. It seems to be the rule of thumb these days. If the DNA is there you are deemed guilty.

innocent?

[tverbeek]

What will be the disposition of the DNA of the innocent?

"No one is innocent!" --Agent Rogersz, Repo Man

DNA is largely similar for close relatives

[aralin]

Here is a big difference. While your kids are going to have totally different fingerprints and even pictures, their DNA to you will be largely similar. So by taking your DNA, you are putting your kids and your relatives in the database as well. If there is a partial match with someone in the database, they will just go after all his relatives and eventually find the right one. They just got a recent mass murder case solved when a daughter of a suspect volunteered to give a DNA sample, when he refused.

Armed Forces Members Probably In Same Boat

[Goo.cc]

While I was in the US Navy, they started a DNA cataloging program, which they claimed was only intended to help identify people in the case of death. They claimed that the information would never be shared and would be discarded after discharge.

It has been 8 years since I was discharged. Want to bet that my information is available to law enforcement, even though I have never been convicted or accused of a crime?

Same law in the UK

[UpnAtom]

Blair is also trying to compulsorily fingerprint everyone and tie together ALL the computerised data held on people through a unique National Identity Number.

Oh, he's also going to track our daily movements through automatic CCTV facial recognition & the ID Card audit trail.

This law has been passed by House of Commons and is currently being debated in the House of Lords. Unless the Lords block it, I'm emigrating somewhere less Orwellian. Anyone want to swap citizenship? I'm serious...

Data never goes away

[billstewart]

Data never goes away once it's collected. (That doesn't count Murphy's Law of course - data you really want goes away quite easily.) Computer storage is cheap, and keeps becoming radically cheaper. Software and system administration / management costs aren't cheap, and don't get cheaper, and systems that weren't explicitly designed to get rid of data mean that expunging data is typically an expensive unreliable manual process. And that's just the costs of expunging the data in the active database - that doesn't count hunting through backup tapes, etc. New software and applications, on the other hand, can often import data from existing systems (again, minus the Murphy's Law issues), and when they do so, they usually aren't very good about maintaining any constraints on usage of the data, and usually aren't very good about backtracking if you want to find out who's had access to the data or get them to erase it.

All of this means that any law or policy that increases data collection is not only dangerous, but the data usually gets used for other things beyond the original purpose - information *does* want to be free. Anything that hangs an unique identifier on data, such as a National ID Card Number (or SSN, or SIN, or driver's license number), makes it easy for data to be imported into other systems and aggregated together. Anything that hangs a non-unique ID onto something, like a firstname+lastname, increases the chances that data will be imported into other systems incorrectly, combining your data with known criminal SameFirstInitial+DifferentMiddleInitial+SimilarLas tname who lives in a different city. In both cases, you'll never get the data expunged.

On the other hand, Moore's Law also means that applications that used to be unthinkable are now routine. When mainframes costs tens of millions of dollars and needed to be fed punchcards and stored stuff on magtape, writing database applications took a couple of years and a large budget, so only critical applications that could be used by lots of people got written. These days, a cheap desktop computer can hold lots more data, and any random civil servant can run a Spreadsheet query or simple fill-out-the-form database application for anything they feel like, such as tracking their ex-girlfriend's new boyfriend's phone bills. And most of that data could really fit in a pocket-computer as well, so next year that same civil servant or telemarketer can take a picture of your face or license plate using their camera-phone and look it up for some arbitrary reason (currently it takes a laptop for the license-plate lookup, and it's being done to nail parking ticket non-payers.)

I'm just wondering if criminals will use a DNA

[multiplexo]

bomb in the future to obscure evidence. You get some blood plasma, or some other fluid that contains a lot of DNA, get samples from different people, mix it up, put it in a bottle with an M-80 taped to it and set it off at the crime scene. Voila, the police end up with so many DNA fragments that there's no way they can tell who did the crime.

If you don't want to use an M80 just get a spritzer bottle full of some DNA containing fluid and spray it everywhere all over a crime scene. I wonder if you could extract DNA sequences from barber shop cuttings and do this?

DNA matching accuracy

[harlows_monkeys]

Maybe this has changed in the last few years, but the last time I looked into this, there was a significant risk in having DNA on file. The problem was that DNA tests consist of comparing the sample against the sample from the presumed criminal at a small number of positions. There will actually be many people in the world who will match the criminal's DNA.

When properly used, this is not a problem. "Properly used" means that you find your suspects using traditional methods, and afterwards run a DNA test on them. Get a match there, and you've got your criminal. It's a Bayesian thing, basically.

What is not proper is to start with DNA, and test the criminal's against all the people you happen to have DNA on file for, looking for a match.

Using DNA to find suspects is only good when you either have a comprehensive database that has DNA from everyone, or your tests are so accurate that they really do uniquely identify people. Like I said, I don't know if they are accurate enough for that. A few years ago, they were not.

Re:DNA instead of passwords.

[sconeu]

What about the guy from the blood bank, who needs money? He just took my blood, but steals the pack instead of putting with everyone else's. Now he can withdraw from my account, because after all, it's *my* blood, and only I would be using it to withdraw cash!

Re:Nothing to Fear

[plague3106]

So you'd be for cameras in every house too because it might help solve just one crime right? Why not..thats what your logic leads to.

If you had bothered to RTFA, you'd know that out of 18 DNA-drag nets, only ONE actually helped collar someone...and it was limited to 25 people that had access to the victim. The rest (where thousands of samples were collected) DID NOT HELP AT ALL.

So, whats the point?

Re:Nothing to Fear

[mankey wanker]

Oh, I laugh...

While I understand where you are coming from, I think you may be naive to the point of stupidity.

You would come forward as a good person, as a good citizen, as someone who seeks the truth.

The police have no such agenda. Their agenda is to provide society with a sense of law and order. That they regularly pin crimes on the most likely suspect is proof of the fact. No one knows the truth, the truth is rarely "outed" during a trial. We solve crimes by pinning them on the most likely person. It gives the appearance of law and order. The most likely suspect is often merely the last person to a see a victim alive, a close family member, a husband, a good friend. God help you if you fail to have an alibi, if you were sitting at home alone watching TV.

Now I could give a shit about Scott Peterson, but take the example anyhow. Scott Peterson had a pregnant wife, Laci. Many couples become estranged during a wife's pregnancy. Scott took a lover, Amber Frey. Juries don't like cheaters. Cheaters are liars and untrustworthy. FWIW, a very large number of people cheat all the time and we all know that fact. Laci was eventually found in the San Francisco bay, a place where Scott Peterson went boating on Xmas eve. Now, I don't know about you - but many people made much of the fact that Scott went boating on that day - as if it were beyond belief that someone would do such a thing just because he needed to get away or to think for a while. To wrap up, Scott Peterson was convicted because he was cheating on his wife, and was seen boating on the vast body of water in which Laci's body was later discovered. I am not saying that there aren't more details, but those are the main details.

They pinned it on him and gave him the thumbs down routine. The man will die largely based on mere speculation. There's not a single piece of incontrovertible evidence in the whole case. There's an alternate defense explanation for everything the prosecution claims.

A culture of spectacle and sacrifice doesn't care about the truth, it cares about appearances. We pin crimes on the most likely to be guilty, not those that are truly guilty. And there may be a universe of difference between those two categories. Is Scott Peterson "guilty beyond a reasonable doubt"? Hell no, but society hates to imagine that there is a murderer loose somewhere. Society likes to nail someone so that the collective can rest easy at night.

Now it turns out that one of my old professors at law school is one of the point men for the whole DNA as evidence movement, his name is Larry Marshall. His big break for DNA evidence came in the Rolando Cruz case. Read about it here: http://www.innocenceproject.org/case/display_profi le.php?id=07

The salient bits are: "...under pressure from the community and in the midst of an election year..." and "...a sheriff's department lieutenant recanted testimony he had given in previous trials." Wow, do I mean it's often just politics and cops that lie? Hell yeah...

I quit law school because Larry Marshall gave a speech in which he informed all of us idiot law school students that the most important thing about the practice of law was how the judge was feeling, what kind of day the judge was having. Did the judge just have a fight with his wife? Is he feeling poorly? Does his stomach roil because of the steak sandwich he had at lunch? That's the guy that will decide all of your motions. He probably won't even read your motions except for during the five minutes before he must render a decision while he trembles on the toilet seat before entering the courtroom. If anyone read anything, it was some poor fuck law clerk that rendered an opinion via post-it note on how the judge should decide the issue.

You know how you play 3D shooters instead of doing your homework? Judges are just like you...

So, would I come forward and admit I was the last person to see some now dead chick alive? I would like to say yes, but the real answer is no. In the adversarial process, you are a suspect until you are excluded as a suspect by the evidence. Does that sound like "guilty until proven innocent"? Yes, it does sound a lot like that.

Re:Nothing to Fear

[bleckywelcky]

But the larger problem is what minimal amount of evidence the public (jury) is willing to base their opinion upon. I was in California temporarily and talked to some permanent residents of the state who were typical juror candidates (towards the end of last year, in the midst of the trial, before a verdict had been reached). And the 4 or 5 people I talked to all agreed that they would convict Peterson based upon the following facts: 1) An affair while his wife was pregnant. 2) He went out on a boat on Christmas day in the same area that the body washed ashore. Scary eh? I tried to probe them as to why they felt this was sufficient evidence and they claimed things such as "intuition"; that he just seemed like such a shifty and immoral character that the only explanation for those two facts must have been that he murdered his wife.