A bit OT - Ear ID gone wrong.

[steve B]

Not a poke at anyone, most methods can be good or not depending on circumstances.

But an interesting article.

http://usnews.nbcnews.com/_news/2013...-identity?lite

Steve B

[bmarlowe1]

"FBI agents investigating the kidnapping of a one-day-old baby boy from a Chicago hospital decided young Scott and the missing son of Chester and Dora Fronczak had the same ears."

What we can very effectively do is say that two images of ball-playing age men do not depict the same person when the ears are different.

That is not the same as saying two images depict the same person because the ears are very similar with no additional evidence (like the face in question is in a team photo and we know the guy was on that team). Worse yet is comparing a newborn to a toddler several years later.

[thecatspajamas]

Also note that the story says nothing of the FBI agents' actual proficiency in ear and facial feature comparisons. Maybe they took other factors into account, maybe not. Maybe they were good at their job, or maybe they were just phoning it in that day and "decided" the ears matched to be able to close the case quickly.

As Steve said, "most methods can be good or not depending on circumstances."

[steve B]

I mostly thought it was interesting that they used that method for ID in 1964.
I thought they might take fingerprints early, I think my parents have a set of mine from very early on. And I wasn't exactly from a big place like NYC.


Steve B

[bmarlowe1]

Steve - it's older than you think. The image below is from "The Bertillon System of Identification: Signeletic Instructions: The Theory and Practice of Anthropometrical Identification, by A. Bertillon, 1896.

The reproduction here is not good so you can't really see the necessary detail, but the idea is not new. The caption reads, "Identity revealed despite beard by the spiral lobe of the ear and the facial angle....same man at ten years interval." Of course there has been a mountain of research and practice since then.