Now that our February 2016 issue is on Newsstands, we want to know what our readers think of our cover story, The Croquet Kid. Do you think the controversial photograph shows Billy the Kid playing Croquet? Here is what some of our friends and readers have thought so far.
The mind can certainly be stretched enough to believe it is. In any event, I hope so.
Gulf Shores, Alabama
“You did an excellent job on the BTK croquet issue. I think a good idea for a cartoon in your issue would be showing the Regulators and Billy in a shootout with Billy shouting “You boys up for a game of croquet after this?”. Or something to that effect. I’m still very skeptical and I’m anxious to hear about this U.S. Marshall and the storage unit from 1994.”
Does not even vaguely resemble pics of what has been claimed to be to be the kid for at least 60 something years.
Janice Lee Weiss Truitt
I just wanted to say thanks for the fine and fascinating article on ” The Croquet Kid” controversy. In the end it boils down to the six hundred pound elephant in the room that is not mentioned. That being vested interest (feverish greed) versus dispassionate research by people who love history for it’s own sake. But if P. T. Barnum was right that there is a sucker born every minute there will be some poor fool who will pays a surrealist price for a common tintype. There will always be gaping canyon between the marketplace and those who just have a genuine interest in the subject. There’s always the lottery for the former.
Rio Verde, Arizona
After that article, I don’t understand how there can be further argument.
Of all the remarks made on behalf of the croquet tintype, Jeff Aiello’s statement, “The Claim” (February 2016), that “naysayers claim ‘facial recognition doesn’t work,’ but then benefit from its effectiveness everyday in the protection of our nation from terrorists,” wins the prize: Together in one fallacious argument, a straw man and a boogie man.
No one argues that facial recognition — a form of biometric analysis — doesn’t work. The question is, how well does it work and under what conditions? Anyone with even a passing familiarity with the literature knows that the effectiveness of facial recognition technology is uneven, and depends on many factors, including the clarity and number of the images in the comparison, the match threshold, the software (there are dozens of competing versions), and the skill of the technician. False positive and false negative identification rates can vary widely.
As for terrorists, facial recognition technology was used in the hunt for the 2013 Boston Marathon bombers, but failed to match their CCTV images with passport and mugshot images in the police database. Only after the police released the CCTV videos to the media were they identified, by people who knew them, including an aunt. Humans trumped computers.
A 2006 study of facial recognition systems funded by the National Science Foundation and the Department of Homeland Security concluded that “there was wide variability in the results from AFR systems, and certain of the poorer performing systems achieved performances close to random.” Close to random means even worse than flipping a coin.
Image clarity is important. A 2012 FBI memorandum pointed out the obvious, that “the performance of facial matching systems is highly dependent upon the quality of the images enrolled in the system.” Not surprisingly, facial recognition technology works better with illuminated, full-face passport photos than with small tintypes of out-of-focus, hat-wearing cowboys.
Aiello’s own facial recognition authority, Kent Gibson, said during the National Geographic program that although he thought there was a match between the putative croquet Billy and the accepted Billy, if asked in a court of law he would “deny” there was a match. “Deny” was the word Gibson used. A tintype whose legitimacy is denied by its own expert witness can hardly be described as having been authenticated.
Do you think Billy the Kid is really in the photograph? Let us know in the comments!