News for hackers. Stuff that matters.
Re:'Equals' Has More Than One Meaning in Math by interroboink on Tuesday June 18, 2024 @01:58AM (#40714473)
Could you provide an example? I'm not seeing it.

Is there some integer that is not also a rational number?

Re:'Equals' Has More Than One Meaning in Math by Cu3PO42 on Tuesday June 18, 2024 @02:11AM (#40714547)
Conceptually, yes, every integer is also a rational number. But they are represented as different objects.

It's a bit like saying every int32 is also a double. Yes, every value of int32 fits into a double, but the bit pattern is different.

The canonical construction for rational numbers is pairs (a, b) which we interpret as a/b. An integer k "is the same as" (k, 1).

So it might be more correct to say every integer has the same value as some rational number. Of course this distinction is pointless most of the time, so we don't worry about it.

This is not unique to integers and rationals. It also applies to naturals and integers, rationals and reals, etc.

Re:'Equals' Has More Than One Meaning in Math by pyrolistical on Tuesday June 18, 2024 @02:21AM (#40714605)
But an int32 1 isn’t the same as a mathematic object 1. So of course your analogy doesn’t work
Re:'Equals' Has More Than One Meaning in Math by Cu3PO42 on Tuesday June 18, 2024 @02:33AM (#40714685)
And a double 1.0 isn't the same as a rational 1/1. But I also didn't claim either of these things.
Re:'Equals' Has More Than One Meaning in Math by klodolph on Tuesday June 18, 2024 @02:28AM (#40714646)
By definition, analogies are between things that aren’t the “same thing”.
Re:'Equals' Has More Than One Meaning in Math by _nalply on Tuesday June 18, 2024 @06:01AM (#40715847)
I understand that this might be a problem in a programming language, even if you decide to ignore that mathematical sets have infinitively many elements. When I programmed an RPN calculator supporting quotients and complex numbers, I had to grapple with the fact that 1 (one) is a quotient (with denominator 1) and a complex number (with imaginary part 0). I solved it by defining the functions and operators taking a certain type and coercing the arguments to that type. One example, the logarithm can take complex numbers. If I had a quotient argument, I coerce it to a complex number before passing it to the logarithm.

However I assumed that this is only a problem in programming languages. I am a bit surprised that mathematics also seems to be affected. I am going to study it a bit.

Re:'Equals' Has More Than One Meaning in Math by ndriscoll on Tuesday June 18, 2024 @11:54PM (#40724630)
Practical mathematics mostly isn't affected. That's what the earlier commenter was referring to with Tao's stages of maturity.

In the pre-rigorous stage, you don't know it's an issue. In the rigorous stage, you know it's an issue. In the post-rigorous stage, you know it's not an issue (i.e. you know you know how to write down all the details if you needed to, and you know it will all work how you might hope it would, and so you don't need to).

Re:'Equals' Has More Than One Meaning in Math by _nalply on Wednesday June 19, 2024 @04:36AM (#40726158)
I think the magic word is: isomorphism!

An isomorphism is a way to relate two different sets such that each element is paired with exactly one in the other set such that it does not matter if you do operations before or after.

In my calculator the set of reals is isomorph to the set of complex numbers with imaginary part zero:

    ℝ ⬄ { c | c ∈ ℂ and im(c) = 0 }
So I can coerce to complex numbers without impunity because of that isomporphism! And I know it is an isomorphism because if adding two reals then coercing to complex is the same as coercing first then adding.

TIL: In a way mathematics has types like programming languages.

I hope I am not too far off here. I didn't look up anything, this all went into my head this morning.

Re:'Equals' Has More Than One Meaning in Math by ndriscoll on Wednesday June 19, 2024 @09:35AM (#40728161)
This is sort of true, but there are isomorphisms which aren't "good enough" in some sense. E.g. if you're familiar with linear algebra, there's a thing called a dual space, which (for finite dimensions) is isomorphic to your space, but when you change coordinates, dual spaces change their coordinates backwards. Or for complex numbers, when you extend R by adding `i` (sometimes written C=R[i]), if you look carefully, you'll see that you could have instead added an element α = -i, and you'll notice that R[α] is isomorphic to R[i] in a way where the isomorphism maps R to itself; in a sense you can't tell whether you added i or -i, and this degree of freedom is important when studying field extensions.

The closest thing I'm aware of to "pretty much equal" is "unique up to unique isomorphism", so they may not be equal, but they're isomorphic, and there's no flexibility to make any choices of which isomorphism to use. But "isomorphism" also implies you have a particular context/structure in mind that you want to preserve, e.g. a set isomorphism (a bijection) may not be a linear isomorphism (~an invertible matrix). In practice, you may be working in multiple contexts at once, so you invent Functors which map one type of morphism to another, and now you care about Categories.

Re:'Equals' Has More Than One Meaning in Math by SAI_Peregrinus on Tuesday June 18, 2024 @09:26AM (#40717599)
To expand: they're "equivalence classes". The digit 1 represents a value, in the same equivalence class as 1/1, 2/2, 1.0, 1+0i, {0|}, 1.0+0i+0j+0k, etc. ad infinitum.
Re:'Equals' Has More Than One Meaning in Math by klodolph on Tuesday June 18, 2024 @02:04AM (#40714510)
How about this example: the natural number 0 is not a member of the integers, from a set theoretic perspective.

In the natural numbers,

  0 = {}
However, {} is not an element of the integers.

This is not something I expect to be easy to understand. This is the standard set theoretic definition of integers that mainstream mathematicians use. This is not some esoteric, fringe theory.

Re:'Equals' Has More Than One Meaning in Math by Affric on Tuesday June 18, 2024 @02:18AM (#40714586)
Isn’t the empty set a subset of all sets?

A quick look at Wikipedia indicates that there exist other constructions.

Re:'Equals' Has More Than One Meaning in Math by klodolph on Tuesday June 18, 2024 @02:27AM (#40714642)
Yes, the empty set is a subset of all sets. I think some wires got crossed somewhere because whether zero is a subset of some other set doesn’t fit into the questions we are trying to answer.

You could pick a construction where the natural numbers are a subset of the integers. This is trivial, but this is a poor strategy overall, because you can always find a bigger set of numbers to work with. You can’t take the “biggest” set of numbers and then define all other sets of numbers as subsets of that. It would be kind of like trying to count down from infinity.

Re:'Equals' Has More Than One Meaning in Math by Affric on Tuesday June 18, 2024 @03:49AM (#40715151)
I see where I have gone wrong. Element vs subset. Classic error when thinking fast. Thank you for the thought provoking comments.
Re:'Equals' Has More Than One Meaning in Math by SAI_Peregrinus on Tuesday June 18, 2024 @09:22AM (#40717554)
The Surreal Numbers manage to be a sort of "biggest" ordered field, in NBG, though they're a proper class not a set. All other ordered fields are subfields of the Surreals. Of course "ordered" is doing a lot, since it excludes the complex numbers, vectors, bivectors, etc. Whether elements of some object that isn't an ordered field should be considered "nubmers" is related but different question.
Re:'Equals' Has More Than One Meaning in Math by enugu on Wednesday June 19, 2024 @02:36AM (#40725482)
Yes, {} is a subset(⊂) of all sets, but it is not a member(∊) of all sets. For instance, {} is not a member of {{{}}}. In the Von-Neumann definition, 1:={0}={{}}. So, 0∊1, but 0 is not a member of {1} which is the above set. https://en.wikipedia.org/wiki/Set-theoretic_definition_of_na...
Re:'Equals' Has More Than One Meaning in Math by interroboink on Tuesday June 18, 2024 @02:30AM (#40714665)
Ah, thank you. It does actually make some sense, having looked into it some more now (:

It looks like the approach of defining ℤ in terms of ℕ is much more tedious to deal with overall, so can see the advantages.