|
Your "thin-layer" analogy sucks. I propose one instead that I think is about 10000+10 times better. How about instead we take your 10kloc of code, and replace every char with a randomly-chosen char with the same rough shape ("w" => "w" or "m", "l" => "1" or "l" or "I", etc). And we publish that. The new product has some interesting properties. Information-theoretically, the new product has less entropy than your code, so it cannot be converted into your code. It doesn't do what your code did (like let you recognize Miles Davis if you didn't already know it was him). On the other hand, it's clearly derivative, because it uses _exactly_ the same whitespace, and the same line-lengths. If you hold printouts of both of them side-by-side, from far enough away, you can't tell which is which... but if you get up close enough the difference is obvious. One is obviously a technically-detailed work that had some specific requirements and took a lot of skill, the other is obviously and evidently less so (but still took a LITTLE skill, or at least effort). If someone already knows the original, the derivative will call it to mind, but if they don't, the derivative is uninteresting. The derivative clearly is a comment on the original, because otherwise it wouldn't exist. Actually, I think I've got a pretty tight analogy here. Why on earth would you be pissed off? |