If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(Washington Post)   The FBI is quietly reviewing the forensic evidence used to convict thousands of defendants after coming to the conclusion that "hair and fiber analysis" is basically junk science   (washingtonpost.com) divider line 70
    More: Followup, FBI, Forensic identification, fiber analysis, FBI Laboratory, Innocence Project, mistakes were made  
•       •       •

4501 clicks; posted to Geek » on 11 Jul 2012 at 1:23 PM (2 years ago)   |  Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»



70 Comments   (+0 »)
   
View Voting Results: Smartest and Funniest

Archived thread

First | « | 1 | 2 | » | Last | Show all
 
2012-07-11 12:43:50 PM
Wasn't Wayne Williams convicted on mostly hair & fiber analyses?
 
2012-07-11 01:13:58 PM
Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups
 
2012-07-11 01:16:02 PM
If these fellas had been convicted in Tejas, they'd be dead by now.
 
2012-07-11 01:20:00 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


What are your feelings on prosecutors that hide or falsely submit evidence just to score a prosecution?
If the FBI is looking into 10,000 cases then there are numerous other cases where the wrong person has been prosecuted. They should make sure it doesn't happen again.
I'm glad to see this getting attention.
 
2012-07-11 01:27:16 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


Are you saying fingerprints can't be used as evidence?
 
2012-07-11 01:28:29 PM

AbbeySomeone: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

What are your feelings on prosecutors that hide or falsely submit evidence just to score a prosecution?
If the FBI is looking into 10,000 cases then there are numerous other cases where the wrong person has been prosecuted. They should make sure it doesn't happen again.
I'm glad to see this getting attention.


My feelings? Well having handled three death penalty appeals where it came out AFTER about 20-odd years that the defendants were either deliberately framed or convicted becasue the prosecuters intentionally overlooked and hid evidence of police misconduct, I might be biased on this score. However I think when cops and prosecutors get caught doing this they need to serve a MINIMUM of the time already served by the person they wrongly convicted. I'd charge anyone who pursued a capital case while knowing their case was bogus, with attempted murder.
 
2012-07-11 01:29:49 PM
The review comes as the National Academy of Sciences is urging the White House and Congress to remove crime labs from police and prosecutors' control, or at least to strengthen the science and standards underpinning the nation's forensic science system.

But the police and prosecutors are completely unbiased and just want to catch the bad guys.

Seriously, why the fark did anyone ever think this was a good idea?
 
2012-07-11 01:31:48 PM

SN1987a goes boom: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

Are you saying fingerprints can't be used as evidence?


I'm saying that they ain't science, andSHOULDN'T be able to be used. The Key to science is repeatablilty, any "scientific" result should be able to be repeatable. There are NO standards for the minimum points of comparison that must match up for a print to be declared "a match" to another. It is completely up to the examiner's discretion. Therefore two examiners looking at the same print can reach different conclusions, which means neither conclusion is "scientific"
 
2012-07-11 01:31:56 PM
Also amazing to me is how often eye-witnesses and confessions turn out to be wrong.

Prosecutors need to win cases to be considered 'good' prosecutors, so they do what they have to in order to win.
 
2012-07-11 01:34:40 PM

Magorn: AbbeySomeone: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

What are your feelings on prosecutors that hide or falsely submit evidence just to score a prosecution?
If the FBI is looking into 10,000 cases then there are numerous other cases where the wrong person has been prosecuted. They should make sure it doesn't happen again.
I'm glad to see this getting attention.

My feelings? Well having handled three death penalty appeals where it came out AFTER about 20-odd years that the defendants were either deliberately framed or convicted becasue the prosecuters intentionally overlooked and hid evidence of police misconduct, I might be biased on this score. However I think when cops and prosecutors get caught doing this they need to serve a MINIMUM of the time already served by the person they wrongly convicted. I'd charge anyone who pursued a capital case while knowing their case was bogus, with attempted murder.


I like the way you think. The penalties for destroying a person's life should be severe.
 
2012-07-11 01:37:04 PM

aaronx: Also amazing to me is how often eye-witnesses and confessions turn out to be wrong.

Prosecutors need to win cases to be considered 'good' prosecutors, so they do what they have to in order to win.


If I ran the world I'd make one basic change to our criminal justice system: I'd combine the State's Attorney and Public Defender's office into one, and lawyers assigned to cases wouldn't know which side they were on until after the indictment (maybe make it a coin-flip). Since they are judged on their W/L records, the attorneys would be highly motivated to ensure the investigation was as fair and by-the-book as possible
 
2012-07-11 01:38:15 PM

SN1987a goes boom: Are you saying fingerprints can't be used as evidence?


Fingerprint matching is actually a very subjective thing, more than you would think. It depends not on actual objective measurement, but on the experience of the technician examining the prints, the quality of the print itself, and a number of other factors.

I was actually surprised when I found this out.

Years ago, I applied for a programmer/analyst position at a government agency that does fingerprint analysis. When they showed me how it was done, I was kind of taken aback. I hope that if I ever become a suspect for whatever reason, that the fingerprint examiner who checks my prints isn't having a bad day, because that could royally fark you for life, and it would be very difficult to argue against in court. In essence, you'd have to put both the honesty and skill level of the examiner on trial, and also the "science" behind fingerprinting.

/Didn't get the job.
//Hiring freeze: The interview was originally scheduled for 1:00pm, September 11th, 2001.
 
2012-07-11 01:40:55 PM

dittybopper: SN1987a goes boom: Are you saying fingerprints can't be used as evidence?

Fingerprint matching is actually a very subjective thing, more than you would think. It depends not on actual objective measurement, but on the experience of the technician examining the prints, the quality of the print itself, and a number of other factors.

I was actually surprised when I found this out.

Years ago, I applied for a programmer/analyst position at a government agency that does fingerprint analysis. When they showed me how it was done, I was kind of taken aback. I hope that if I ever become a suspect for whatever reason, that the fingerprint examiner who checks my prints isn't having a bad day, because that could royally fark you for life, and it would be very difficult to argue against in court. In essence, you'd have to put both the honesty and skill level of the examiner on trial, and also the "science" behind fingerprinting.

/Didn't get the job.
//Hiring freeze: The interview was originally scheduled for 1:00pm, September 11th, 2001.


It does seem like there would have been some vacancies to fill right about then.

/sorry

/aisle seat
 
2012-07-11 01:46:55 PM
We should hurry up and execute all of these inmates.
 
2012-07-11 01:47:56 PM

AbbeySomeone: It does seem like there would have been some vacancies to fill right about then.

/sorry

/aisle seat


By the time the hiring freeze was ended, I had a different job. Not that I really wanted to work there anyway, because I also would have ended up supporting that 10 year abortion of a program called CoBIS.
 
2012-07-11 01:56:29 PM
maybe this will budge some of my family members from their stance that the death penalty is okay, even though innocent people have been/ar put to death, because it's only a few innocent people and mostly "bad" people.
 
2012-07-11 01:57:10 PM
hmmm, english. It's in there somewhere!
 
2012-07-11 02:00:45 PM
Prosecutors and police hiding and making up evidence is the just the natural extension of allowing the police to makeup and fabricate any "prop" evidence to get a criminal to confess. It isn't uprising the mentality leaked over into the court system as well.
 
2012-07-11 02:03:40 PM
In addition, prosecutors did not notify defendants or their attorneys even in many cases they knew were troubled.

Wouldn't that alone - witholding potentially exculpatory evidence - justify overturning the verdict? Aren't the required to tell defendants if the found something that even might prove them innocent?
 
2012-07-11 02:09:28 PM

Magorn: My feelings? Well having handled three death penalty appeals where it came out AFTER about 20-odd years that the defendants were either deliberately framed or convicted becasue the prosecuters intentionally overlooked and hid evidence of police misconduct, I might be biased on this score. However I think when cops and prosecutors get caught doing this they need to serve a MINIMUM of the time already served by the person they wrongly convicted. I'd charge anyone who pursued a capital case while knowing their case was bogus, with attempted murder.


How dare you hold a government agent/agency up to a similar negligence standard as common folk!
 
2012-07-11 02:13:04 PM
I gotta say I agree with the group at the bottom of page 1, a forensic lab should be associated with law enforcement as little as possible. Police should collect evidence, stuck a random ID number on the bag and have it couriered to the lab. The lab looks at it, and sends it back with a report saying it matched samples x, y, and did not match sample z. A fingerprint doesn't change depending on what the owner's name or race is, or whether he is suspected of jaywalking or killing a cop so why should the print technician know those details; at best it would do nothing, at worst it would influence his findings to get the bastard.
 
2012-07-11 02:14:01 PM

aaronx: Also amazing to me is how often eye-witnesses and confessions turn out to be wrong.

Prosecutors need to win cases to be considered 'good' prosecutors, so they do what they have to in order to win.


Criminal law makes up a significant part of my practice and I completely agree. Prosecutors care very much about putting people in jail and much less about whether the person actually did anything.
 
2012-07-11 02:21:30 PM
The FBI is quietly reviewing the forensic evidence used to convict thousands of defendants after coming to the conclusion that "hair and fiber analysis" is basically junk science

So, sort of like "Profiling" and "Drug detecting dogs" then.
 
2012-07-11 02:22:23 PM

Magorn: aaronx: Also amazing to me is how often eye-witnesses and confessions turn out to be wrong.

Prosecutors need to win cases to be considered 'good' prosecutors, so they do what they have to in order to win.

If I ran the world I'd make one basic change to our criminal justice system: I'd combine the State's Attorney and Public Defender's office into one, and lawyers assigned to cases wouldn't know which side they were on until after the indictment (maybe make it a coin-flip). Since they are judged on their W/L records, the attorneys would be highly motivated to ensure the investigation was as fair and by-the-book as possible


I nominate Magorn to run the world.

Do I hear a second?
 
2012-07-11 02:24:12 PM

dittybopper: AbbeySomeone: It does seem like there would have been some vacancies to fill right about then.

/sorry

/aisle seat

By the time the hiring freeze was ended, I had a different job. Not that I really wanted to work there anyway, because I also would have ended up supporting that 10 year abortion of a program called CoBIS.


CoBIS is a bad joke, but they never learn. Now it's the even more asinine "micro-stamping". It's hard to believe anyone with even a lick of sense thinks that would work.
 
2012-07-11 02:31:32 PM

Magorn: aaronx: Also amazing to me is how often eye-witnesses and confessions turn out to be wrong.

Prosecutors need to win cases to be considered 'good' prosecutors, so they do what they have to in order to win.

If I ran the world I'd make one basic change to our criminal justice system: I'd combine the State's Attorney and Public Defender's office into one, and lawyers assigned to cases wouldn't know which side they were on until after the indictment (maybe make it a coin-flip). Since they are judged on their W/L records, the attorneys would be highly motivated to ensure the investigation was as fair and by-the-book as possible


You might need to add some details to that plan; maybe have them pull six months terms as the DA or public defender. I could see a situation occuring where you might end up with one guy who defended some bank robber two years ago, gets some kind of knowledge of his gang's inner workings, and then uses it in a prosecution of a partner sometime down the road. It needs some checks and balances or oversight, but in principle I'd heartily agree with your plan.
 
2012-07-11 02:35:09 PM
1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

6) Solving the crime in an hour
 
2012-07-11 02:44:44 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


Except police departments do this. Typically chemists. Are chemists scientists?
 
2012-07-11 02:49:38 PM

Magorn: 1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*


Let's not forget to add polygraphs to the junk list.
 
2012-07-11 02:51:11 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"


I'm curious about this one. Can't it at least help determine an accidental fire vs arson? I would assume there's a observable difference between, "yup it started at the stove" and "wow, this person had an awful lot of gasoline on their couch."
 
2012-07-11 02:54:46 PM

cig-mkr: 6) Solving the crime in an hour


I never understood this gripe. It's a television show. If you're going to show a television show where someone does DNA matching or whatever, you can't have it take two weeks. It's only an hour show. You should afford the TV people at least a modicum of the suspension of disbelief.

No one ever biatches that Winterfell and King's Landing or so far apart no one could *possibly* ride there in 15 minutes...
 
2012-07-11 02:56:09 PM

The Homer Tax: cig-mkr: 6) Solving the crime in an hour

I never understood this gripe. It's a television show. If you're going to show a television show where someone does DNA matching or whatever, you can't have it take two weeks. It's only an hour show. You should afford the TV people at least a modicum of the suspension of disbelief.

No one ever biatches that Winterfell and King's Landing or so far apart no one could *possibly* ride there in 15 minutes...


Unless you're watching "24" and the whole premise is based on the fact it's "real time" and so much shiat happens that could never fit into 24 hours.
 
2012-07-11 03:02:53 PM
Magorn:
5) DNA*
*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


...not to mention that, while the test is theoretically that sensitive, the chances of a tech screwing up in the lab is about a million times more probable.

Then, of course, there's the flat-out lying that goes on.

A number of years ago, a guy was convicted of murder. He was a known peeping Tom, and they found his fingerprint on the outside of a locked sliding glass door at a murder scene (an apartment in the same complex where he lived). The local "expert" claimed that the fingerprint was put on the glass at the time the murder was committed (this is a flat out lie, which came out years later - you can't tell the time a fingerprint was placed on an object by looking at it.) The guy was convicted, and was on Death Row for several years before being released.
 
2012-07-11 03:03:08 PM
Screw evidence.

In Denton, TX, we had a man convicted of murder with NOTHING. No body, no crime scene, no blood stains. NOTHING. And the purported victim had a history of running away and manic depression.
 
2012-07-11 03:08:00 PM

ProfessorOhki: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"

I'm curious about this one. Can't it at least help determine an accidental fire vs arson? I would assume there's a observable difference between, "yup it started at the stove" and "wow, this person had an awful lot of gasoline on their couch."


You would think that but most fire investigation again comes down to the personal feelings of the investigator. It is true that there is some science that can actually help determine how a fire started/spread but most "Arson investigators" have utterly no clue about it, as it is pretty advanced physics.

The case of Todd Willingham in TX, a likely innocent man executed killing his three kids by deliberately setting a house fire, particularly the post conviction appeal effort is a good read, as it pit the findings of an actual professor of physics against the poorly trained "Arson investigator" the prosecution relied on as its key witness.
 
2012-07-11 03:09:37 PM

cirby: Magorn:
5) DNA*
*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

...not to mention that, while the test is theoretically that sensitive, the chances of a tech screwing up in the lab is about a million times more probable.

Then, of course, there's the flat-out lying that goes on.

A number of years ago, a guy was convicted of murder. He was a known peeping Tom, and they found his fingerprint on the outside of a locked sliding glass door at a murder scene (an apartment in the same complex where he lived). The local "expert" claimed that the fingerprint was put on the glass at the time the murder was committed (this is a flat out lie, which came out years later - you can't tell the time a fingerprint was placed on an object by looking at it.) The guy was convicted, and was on Death Row for several years before being released.


I seem to recall that case, wasn;t the guy convicted of severly dimished mental capacity as well?
 
2012-07-11 03:14:13 PM
7) Polygraph
 
2012-07-11 03:21:50 PM
Department officials had known for years that flawed forensic work might have led to the convictions of potentially innocent people but had not performed a thorough review of the cases. In addition, prosecutors did not notify defendants or their attorneys even in many cases they knew were troubled.

Shouldn't this alone vacate their convictions?
 
2012-07-11 03:24:30 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


Next thing you will tell me is that blood splatter analysis isn't a hard science.
 
2012-07-11 04:21:39 PM
and they don't even have a semen database either.
 
2012-07-11 04:32:58 PM

Pants_Optional: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

Except police departments do this. Typically chemists. Are chemists scientists?


First, police departments often do this stuff. The NAS report was pretty unhappy about that. Many analysts are not, in fact chemists, but ex-cops. this is a problem.

The DNA work is done by biologists, who are generally trained pretty well (not always). It is chemists doing the controlled substance analysis, but that is pretty simple and routine. fingerprint analysis is usually done by non-scientists. It's not necessarily a bad field, subjective though it can be, but those other fields are 100% crap. I would say that most or all of the pattern evidence stuff is crap.
 
2012-07-11 04:56:57 PM

Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups


My understanding was that ballistic analysis could be pretty sketchy too, like matching a recovered and deformed round to a particular barrel. Any forensics people here?
 
2012-07-11 05:09:58 PM

Magorn: ProfessorOhki: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"

I'm curious about this one. Can't it at least help determine an accidental fire vs arson? I would assume there's a observable difference between, "yup it started at the stove" and "wow, this person had an awful lot of gasoline on their couch."

You would think that but most fire investigation again comes down to the personal feelings of the investigator. It is true that there is some science that can actually help determine how a fire started/spread but most "Arson investigators" have utterly no clue about it, as it is pretty advanced physics.

The case of Todd Willingham in TX, a likely innocent man executed killing his three kids by deliberately setting a house fire, particularly the post conviction appeal effort is a good read, as it pit the findings of an actual professor of physics against the poorly trained "Arson investigator" the prosecution relied on as its key witness.


See I always wondered that about fires, but not being a scientist I'd be talking out of my rear end if I declared it BS. Since the act of burning alters everything at a molecular level, how would you definitively know so many things about how it happened? How do you even tell where it started if the whole building is now just a smoldering pit?
 
2012-07-11 05:17:40 PM

ProfessorOhki: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"

I'm curious about this one. Can't it at least help determine an accidental fire vs arson? I would assume there's a observable difference between, "yup it started at the stove" and "wow, this person had an awful lot of gasoline on their couch."


I'm assuming he's talking about things like depth of char, alligator char, and spalling. For a long time, it was held that certain types of damage were key components in determining the origin and cause of a fire. They can't.

I will humbly disagree that fire reconstruction isn't possible. Organizations like the NIST can successfully create models that accurately replicate fires, but the capability is limited to a very few organizations.

That isn't to say that most arson investigations are bad - they aren't. Many arsonists are dumb and do a bad job of setting fires, so the FD is able to suppress the fire and preserve the evidence. A prime example is the guy in AZ that killed himself prior to sentencing.
 
2012-07-11 05:59:22 PM

pyrotek85: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

My understanding was that ballistic analysis could be pretty sketchy too, like matching a recovered and deformed round to a particular barrel. Any forensics people here?


Like fire reconstruction, and all pattern evidence, it isn't that a good scientific analysis isn't possible, it's that it's never (or seldom) been done. There's a lot of 'good common sense' type stuff in these fields. Which means that they're mostly crap. Some science is starting to creep into these fields at the academic and federal levels. Not so much at the state or local level. Emphasis on 'starting'.
 
2012-07-11 06:37:35 PM

MugzyBrown:
Unless you're watching "24" and the whole premise is based on the fact it's "real time" and so much shiat happens that could never fit into 24 hours.


Except shiat never happens on 24. Not once was there ever even a 10 minute slow part because Jack is in the john. They dont eat, they dont sleep, and nobody craps. for 24 hours.
 
2012-07-11 07:13:07 PM
FTA: "The review comes after The Washington Post reported in April that Justice Department officials had known for years that flawed forensic work might have led to the convictions of potentially innocent people but had not performed a thorough review of the cases."

So, there would be no review, even tho the JD knew that innocent people were rotting in jail, if the Post hadn't published their story.

Time for Obama to invoke the War Powers Resolution and spend the next 60 days bombing Washington DC and its surrounding suburbs into rubble. After all, the purpose of the military is to defend this country.
 
2012-07-11 09:28:48 PM

pyrotek85: Magorn: Things people (epecially CSI watchers) think are real science

1) Hair and fiber analysis
2) "paint matching"
3) Fire reconstruction and "burn pattern analysis"
4) Fingerprint Matching
5) DNA matching

Things that are actually science

5) DNA*

*and then it is far more defintive in excluding people than when including them, the often used line in testimony that there is only a "1 in 600 billion chance" of two people having the same DNA is pure bullshiat because a) They've never tested a sample size that large for obvious reasons b) The test only a limitied number of alleles and the science isn't all that clear yet how many of them are purely random and how many are closely tied to certain racial groups

My understanding was that ballistic analysis could be pretty sketchy too, like matching a recovered and deformed round to a particular barrel. Any forensics people here?


I'm not in forensics, but I've been in crime labs and have looked at bullets through comparison microscopes. You put the recovered bullet from the crime and the one from a test shot under the scope and view them superimposed on each other. You only need a small part of the tail end of the bullet to be in good shape to match the grooves to a very high certainty. Besides the main grooves caused by the original rifling, there are smaller grooves caused by scratches in the barrel. Those scratches change over time when a gun is used frequently or if you run some coarse steel wool through the barrel. Then you only have the rifling grooves to go by, and that would only narrow the match to the same rifling tool used to make the gun.
 
2012-07-11 09:48:20 PM

Karac: I gotta say I agree with the group at the bottom of page 1, a forensic lab should be associated with law enforcement as little as possible. Police should collect evidence, stuck a random ID number on the bag and have it couriered to the lab. The lab looks at it, and sends it back with a report saying it matched samples x, y, and did not match sample z. A fingerprint doesn't change depending on what the owner's name or race is, or whether he is suspected of jaywalking or killing a cop so why should the print technician know those details; at best it would do nothing, at worst it would influence his findings to get the bastard.


I've been toying around with a similar idea for a while now. Some way to make the entire legal system actually, factually blind. Find some way to completely mask the identity of the defendant for the entire trial. Never use their name, etc.

It would go a long way towards preventing juries from deciding based on a slew of non-relevant criteria (sex, race, celebrity, etc). Imagine if the rich, powerful, and famous weren't treated differently in court?

/Very tough to implement though...
//Justice should be blind
 
2012-07-11 10:02:48 PM
Honest Bender:

I've been toying around with a similar idea for a while now. Some way to make the entire legal system actually, factually blind. Find some way to completely mask the identity of the defendant for the entire trial. Never use their name, etc.

It would go a long way towards preventing juries from deciding based on a slew of non-relevant criteria (sex, race, celebrity, etc). Imagine if the rich, powerful, and famous weren't treated differently in court?

/Very tough to implement though...
//Justice should be blind


Too bad our country won't work like that.

It never has, and it never could.
 
Displayed 50 of 70 comments

First | « | 1 | 2 | » | Last | Show all

View Voting Results: Smartest and Funniest


This thread is archived, and closed to new comments.

Continue Farking
Submit a Link »






Report