Post by JayUtah on Sept 27, 2007 13:29:49 GMT -4
Beat on me, if you must.
Womp.
I was curious to see reactions to the question for several reasons. Would anyone question the premise?
Nobody questioned the truth of the premise, but there were several questions trying to clarify it. That's significant. Asking for clarification helps us converge to a proper understanding, but it doesn't investigate whether someone is knowingly lying.
How far would anyone go to find evidence of alteration?
In this case only about 20 minutes' effort. It failed to produce an answer about which one was changed, but it didn't go far enough to warrant questioning the premise.
In the larger sense that's something I have to deal with frequently in my profession. Testing and validation always presumes that certain things are as expected. So when a test fails, we have a basis for eliminating certain potential causes: "We know it isn't X." But when you exhaust all the other hypotheses, it's a good idea to go back and pay closer attention to those earlier presumptions. Because it happens often enough that it really was X, and the flaw was in the test rather than in the test subject.
Or as I phrase it for management: "Find the needle in this haystack" is a different problem than "See whether there's a needle in the haystack." You have to make sure you know which one you're solving before you plan the search. If you know there's a needle, then failure to find it means you're not looking hard enough. If you don't know there's one, failure to find it might also mean there isn't one.
It also matters how important it is to find the answer. In the context of a web forum thread, 20 minutes is about my limit. If I were testifying in court where a defendant's life depended on detecting a forgery, I would expend great effort.
What convincing techniques for image processing exist outside the digital environment?
The good old razor-blade and chemicals still work if the photograph is on film. Good old in-camera compositing methods still work -- mattes, forced perspective, etc. -- but they're not usually post-processing techniques.
I hope I'm not on anyone's death list, now.
Exercises in out-of-the-box thinking are always welcome.
Womp.
I was curious to see reactions to the question for several reasons. Would anyone question the premise?
Nobody questioned the truth of the premise, but there were several questions trying to clarify it. That's significant. Asking for clarification helps us converge to a proper understanding, but it doesn't investigate whether someone is knowingly lying.
How far would anyone go to find evidence of alteration?
In this case only about 20 minutes' effort. It failed to produce an answer about which one was changed, but it didn't go far enough to warrant questioning the premise.
In the larger sense that's something I have to deal with frequently in my profession. Testing and validation always presumes that certain things are as expected. So when a test fails, we have a basis for eliminating certain potential causes: "We know it isn't X." But when you exhaust all the other hypotheses, it's a good idea to go back and pay closer attention to those earlier presumptions. Because it happens often enough that it really was X, and the flaw was in the test rather than in the test subject.
Or as I phrase it for management: "Find the needle in this haystack" is a different problem than "See whether there's a needle in the haystack." You have to make sure you know which one you're solving before you plan the search. If you know there's a needle, then failure to find it means you're not looking hard enough. If you don't know there's one, failure to find it might also mean there isn't one.
It also matters how important it is to find the answer. In the context of a web forum thread, 20 minutes is about my limit. If I were testifying in court where a defendant's life depended on detecting a forgery, I would expend great effort.
What convincing techniques for image processing exist outside the digital environment?
The good old razor-blade and chemicals still work if the photograph is on film. Good old in-camera compositing methods still work -- mattes, forced perspective, etc. -- but they're not usually post-processing techniques.
I hope I'm not on anyone's death list, now.
Exercises in out-of-the-box thinking are always welcome.