Technology means seeing is no longer believing
By Jay Fidell
| |||
On July 9, in more saber rattling, Iran said it had test-fired missiles capable of carrying nuclear warheads that could reach Israel. Iran's news release was an exaggeration, as is frequently the case.
To bolster the story, Iran posted a photo on its news site showing four missiles being fired. This was shown around the world. The next day, The New York Times, tipped off by the Little Green Footballs blog, reported that one of the missiles was painted in. Where was the CIA?
Caught in the act, Iran replaced the photo with another photo showing only three missiles in the air and an unfired one on the ground, a dud, still in a launch position. This confirmed the fraud. The blogs were flooded with thousands of comments, understandably merciless.
We've seen fraudulent photos from the Middle East before. In 2006, for instance, Reuters published doctored and staged photos of the war in Lebanon. Photos like this predictably get on the front page, and millions of people get taken in. These photos can affect public opinion and distort history, effects not necessarily reversible by subsequent retraction.
PHOTOSHOP EXPERTISE IS PROLIFERATING
The mother of photo editing programs is Adobe Photoshop. Adobe also has the lesser-known Fireworks. In open source, there's Gimp. But the fraud here was so clumsy we can't be sure which program was used.
Digital editing software has come a long way since Photoshop, the Internet and digital photography all converged in the mid-'90s. It's hard to trust images these days. So many photos that look "real" are in fact manipulated. Take a look at any fashion magazine.
Making a model look good is one thing, but you can also influence public opinion with subtle changes in light and color, e.g., Time magazine's darkened O.J. Simpson photo. Where's the ethical line? However you look at it, cloning a nuclear missile is not an innocent change; it's intended to create global mischief, and is neither cute nor forgivable.
PBS recently profiled Dartmouth's Hany Farid, an expert in "digital forensics." (See www.pbs.org/wgbh/nova/sciencenow/0301/03.html.) He created software that will recognize digital cloning and verify lighting sources by reflections, including those in the human eye. Using this, he uncovered the fake Jane Fonda photo used against John Kerry in 2004.
Farid is in discussions with Adobe, and perhaps someday there will be an "Identify Cloning" button in the Photoshop toolkit. That would certainly help us spot fake photos, but at the same time it would also help the fraudsters make their work less detectible, so it's a mixed bag.
IS THE WRITTEN WORD ANY MORE RELIABLE?
Eye-popping images are everywhere these days. We are wired to believe what we see, but any photo can be a thousand lies. Many are digital and on the Net. This is double-edged — just as the Net shows photos that are false, it also allows us to examine them to evaluate their authenticity.
As they say in the news business, "if your mother says she loves you, check it out!" The news junkie should know what Photoshop is capable of doing, and undoing. And how about an "Honest Photo Association" that would independently verify photos and certify them to the public?
Hopefully, we won't be duped too often and wars won't start on the basis of fraudulent photos. But is the written word any more reliable? Even The New York Times has had reporters who lied. Can we be sure a given story is true? How about this column? Now, that's really chilling.
Jay Fidell is a business lawyer practicing in Honolulu. He has followed tech and tech policy closely and is a founder of ThinkTech Hawaii. Check out his blog at www.HonoluluAdvertiser.com/Blogs