[soulofmischief]

Actually, all numbers are functions in Peano arithmetic. :)

For example, S(0) is 1, S(S(0)) is 2, S(S(S(0))) is 3, and so on.

There is no end of a number line. There are lines, and line segments. Only line segments are finite.

> There is no infinity'th prime number. There is a process by which you can show that a bigger primer number must always exist.

You misunderstand the concept of infinity. Cantor's diagonal argument proves that such a bigger number must always exist. "Infinity'th" is not a place in a number line; Infinity is a set that may be countable or uncountable, depending on what kind of infinity you're working with.

There are infinities with higher cardinality than others. Infinity relates to set theory, and if you try to simply imagine it as a "position" in a line of real numbers, you'll understandably have an inconsistent mental model.

I highly recommend checking out Cantor's diagonal argument. Mathematicians didn't invent infinity as a curiosity; it solves real problems and implies real constraints. https://en.wikipedia.org/wiki/Cantor's_diagonal_argument

[psychoslave]

>You misunderstand the concept of infinity. Cantor's diagonal argument proves that such a bigger number must always exist. "Infinity'th" is not a place in a number line; Infinity is a set that may be countable or uncountable, depending on what kind of infinity you're working with.

Diagonal argument doesn’t work in a constructive ground. It’s not a matter of whether the conclusion is valid, but if we have blind faith in the premises and are fine about speaking of something we can’t build.

They are things that humans will never be able to construct, no matter how far their control over the universe surrounding them might go. To start with, humans can create the universe, — whether it’s infinite or not.

[drdeca]

We can construct a function G, which, given a function f : N -> N -> {0,1} returns a function G(f) = h : N -> {0,1} defined by h(i):=not(f(i)(i))

This h=G(f) has the property that, for all i, there exists a j such that f(i)(j)≠h(j) . In particular, j=i will work for this.

It seems to me that this is all constructive.

The only out that I see is to not consider the class of “functions from N to {0,1}” to be something that exists (as a set, or type, or whatever).

Like, you can fairly reasonably hold the position that there is no powerset of the natural numbers, but you can’t reasonably hold the position that it exists and that there is a surjection from the natural numbers to it. (Likewise with any other set N. This isn’t specific to the natural numbers.)

We have a constructive refutation of that claim, in the sense that we have a construction of a function which, given such a surjection (as in, a function along with a promise that the provided function is such a surjection), produces a contradiction.

[drdeca]

> For example, S(0) is 1, S(S(0)) is 2, S(S(S(0))) is 3, and so on.

S is a function symbol. S(0) (in PA) is not a function. It is an expression involving one.

[soulofmischief]

You're right, I was way too imprecise in my language. Thanks for the correction. I'd rather say that all numbers can be interpreted as being constructed via a process, just as the parent comment described their interpretation of Infinity.

[alexnewman]

Let’s keep it simple. What physics or engineering is easier? Let us ignore mathematics for its own sake . If you can’t show use there … id argue it’s our math aesthetics that are wrong

[drdeca]

Why have you picked that comment of mine to reply with that reply? I was just correcting an error. Your reply seems irrelevant to my comment.

I was just saying that “one more than 0” isn’t a function just because in Peano arithmetic, the successor function, along with the constant 0, is used to denote natural numbers.