Luke's message (below) continues to mischaracterize various things and misstate my positions and generally obscure things.
So I'll use my legendary patience to attempt yet again to clarify, and will try to refrain from sarcasm, which Luke tries to use as a substitute for argument. (I was going to say, as a simulation of an argument, but that would be sarcasm.)
>If he wants me to agree that simulated hydrogen
>atoms are not real, he needs to nail down what
>he means by "real" in the first place -- in a
>manner that doesn't simply assume the conclusion.
I did not say that simulated hydrogen atoms are not real. I said that a description in writing of a hydrogen atom is not a hydrogen atom. And I spelled out reasons--e.g. that the pencil marks on paper do not and cannot do and be everything that an atom can do and be. E.g., 2 atoms can form a molecule, whereas 2 pieces of paper cannot. Whar could be more obvious?
Luke's remarks about simulations being real, i.e. that the parts and events in a computer are material, is irrelevant.
Again I appeal to other readers to chime in and reveal whhch arguments seem sound to them and which not, and why.
Response to To be or not to be
Robert has presented us with a brilliant defense of assuming the conclusion. Simulated space? Doesn't count. Simulated hydrogen atoms? Not real enough. By definition. How can we argue with that?
The fact is he is importing a bias that has no place in this conversation. If he wants me to agree that simulated hydrogen atoms are not real, he needs to nail down what he means by "real" in the first place -- in a manner that doesn't simply assume the conclusion.
Objects in the physical universe do have limitations that digital objects don't. For example, it is more trivial to duplicate a digital object, given enough storage space, and thus save a history or create an alternate timeline for the object (separate object, to the degree that they diverge). You can also clock it's speed up and down at a push of a button.
But these added properties are not sufficient to render them non-real. Non-real objects would be things like beauty and truth -- concepts of an abstract nature. Programs and files are not like that. They are encoded numerically, but that does not make them numbers in the abstract sense like "the number three" is a number.
They are even less abstract than a mathematical function, because they are specific material phenomena. Even though their exact physical location, shape, etc. are rather transient and largely irrelevant, it bears remembering that they remain physical machines with genuine physical functions and mechanisms all the way down, every bit as real as an automobile or sewing machine -- if yet ever so much more precise.
As to the hydrogen atom represented on a non-computed piece of paper, it is an example that generates more confusion than it dispels in my opinion. There are things (like suspending the passing of time) that are easy to do for digital objects but not physical ones. A piece of paper is like a simulation that has been frozen in place, and thus (according to my argument) like a physical object in an instant of time. The difference is that the paper is implied to stay that way for a long while (or forever), whereas an object or simulation would only remain in that state briefly. But as I've said, the fact that you can do this for digital objects and not physical ones doesn't make digital objects less real -- it merely means you have more control over them.