In my first post about owning intellectual property I laid out some of the difficulties that arise when a person tries to assert property rights over information stored in a digital format.
Once something (say, a song) is stored digitally, it is represented by a long series of 1s and 0s -- the song is essentially translated into a really big number. Depending on the format used for storing the song (MP3, WAV, &c.), the number that represents the song will be different. So does the owner of that song also own the number that represents that song? And if so, the number generated by what storage scheme? All of them?That's a big problem because any number can be changed to any other number using the appropriate math. For instance, I could easily write a program that takes in the number that represents a particular song and then adds 1 to that number. Does the owner of the song own that number too? I could write a music program that takes this new number, subtracts 1, and then plays the song... so in a sense I simply created a new storage format for music. Adding and subtracting 1 is simple, but there are literally an infinite number of possible storage formats that haven't been invented yet, and they span every single number. Any number, when combined with the right algorithm, can be used to generate any song. So where is the copyright infringement? Is the song stored in the long binary number, or in the algorithm that decodes it? I could write another piece of software that no matter what file you put into it always plays the same song. Is every single file in existance now in violation of that song's copyright? ...
In the end, it's impossible to own numbers. Since numbers are used to represent everything stored digitally, it seems impossible to me that copyright as we now know it can continue to exist. Add to this other thorny issues such as digital child pornography ("that's not a picture, officer, that's just a really long number I got from my math equations") and decryption warrants and I think we're just on the tip of dealing with the changes that the digital revolution will force upon our civilization.
Rand Simburg delves into similar territory by questioning how anyone can verify that news photos are authentic, but his proposal doesn't look possible to me.
How, then, to know if a published photo is, in a paraphrase of the old commercial, real, or Memorex?There are no obvious easy solutions to this problem, other than the traditional ones for validating evidence -- chains of custody. Press photographers could be required to use certified cameras that time stamp pictures in an encrypted way that doesn't permit modifying the stamp. They could go to accredited image processors who would verify the validity of the original picture from the camera (perhaps even uploading it to a certified notary storage site), and describe any image processing they performed, at risk of loss of accreditation if they pull any funny business. This would, of course, come at a cost, in both dollars for the intermediary and (more importantly for the news business) timeliness. Unfortunately, in the wake of this and other news bias scandals, any news organization that doesn't pull in the reins on its stringers and freelancers, and implement a solution like this, is going to suffer in credibility as time goes on.
The part I bolded is the weak link in the proposal, because there's no way to generate encrypted timestamps that cannot be modified given the right knowledge (of the encryption algorithm) and tools (to modify/bypass whatever hardware restraints are built into the camera). For instance, how would one verify that a particular picture/timestamp pair was created by a "certified" camera? As soon as one such camera is sold, its encryption routine will be reverse engineered to allow anyone to generate timestamps on their own computers at any time -- e.g., after modifying an authentic photograph. Such a device might make forgery a little more difficult, but considering how easy commercial software is to crack I doubt it will take long for pirates to break into a camera. (Say about one day after the first camera is commercially available.)
Actually, I would say that the weak link is the accrediting agency. What that comes down to is reputation, and I don't think a seperate agency is required for that. We are seeing the effects of reputation being wreaked on Reuters right now, and it is having a ripple effect on the other agencies.
The right thing for us to do is express our indignation when it is exposed, and the market will police itself.
Plagiarism is just one of those “I know it when I see it” things and that does not change because of the digital form. Then there is “Figures don’t lie, but liars figure”. Playboy was using the airbrush on centerfolds before computers came along, so there is nothing new. I imagine the editing that comes from Dan Rather and Reuters today was done similarly in cuneiform.
Saying that a file, once stored in digital format, is just a number is factual, but dishonest.
A song is a song. When played, it sounds thus and so, and it is artwork performed by artists often for hire by some other party who ACTUALLY owns the work.
Boiling it down to 'just a really big number' and then saying (with a smirk, no doubt) that one can not own a number is intellectually dishonest.
Shouldn't we be talking about the concept of ideas as property, in the more general sense? I mean sheesh! Talking about big long numbers is geeky and all, but that song I just heard was 'Sting' and regardles of how it was encrypted, it was 'Sting' when I heard it, and if that track was copied without being paid for, then it was stolen. (fair use and backup caveats of course)