Now you are starting to see. The central issue can never be just uncertainty. The issue is, exactly WHAT is uncertain? What exactly is SIGNIFICANT to the sender of a message? What is significant to the intended receiver of the message? What is significant to an unintended observer of a message?
What is significant about the fact that I capitalized some of the words above? Does "What" signify the same meaning as "WHAT" or "what" or "wHaT"? Are you sure I am using the same alphabet that you are, or that you think I am? Perhaps, to me "W" and "w" are entirely different letters of a very large alphabet. Perhaps what you perceive to be as merely a different font, is really an entirely different letter.
Here is where it starts to get interesting. Suppose you make an observation (think of trying to read a letter written by someone with "bad handwriting") of something that appears to be familiar, but appears noisy or distorted (think about why your new HDTV image is so much cleaner than your old TV image). Is the difference between the familiar, and the distorted copy significant? Maybe, maybe not. It all depends on what the entities know about each other. An entity that knows that the sender is always using a very limited alphabet might be able to assume that an observed "badly written" character is in fact just a noisy, distorted version of a member of the known alphabet. He may then replace it with a "clean" copy. When your new digital TV started doing things like this, then like magic, the image on your new HDTV screen suddenly became noise-free.
You stated that: nature decides what that selection will "look like"
But in the HDTV case just described, it is no longer just nature doing the "deciding"
Nature decided to add noise and distortion (uncertainty) to the received TV symbols.
Then the TV receiver decided that the received symbols were noisy and distorted.
Then the TV receiver decided to replace the noisy and distorted symbols with clean ones.
Then, like magic, you see a noise-free image. Then uncertainty has been eliminated.
Where does the uncertainty lie? In the decision as to which "clean symbol" should be used to replace the "noisy and distorted symbol".
And then, in the next steps in the process, things start to get really interesting. But that is another story.
I hope you can see how these types of behaviors can have a major impact on "observing" "attributes of objects" and "attributes of the relationship between an object and an observer".
If you do, then you are seeing something that the physicists have heretofore missed. And it is something not to be missed.