> In science nothing can be proven.
Who are you referring to? Nobody has mentioned anything about proven. Your entire comment is focused on arguing against science being proofs, but nobody has said science is proofs.
OP said that scientific theories should "explain and predict" which they should. Those that don't should be discarded eventually (which they should). Why are your discussing mathematical proofs of science when nobody brought it up nor proposed that?
- In science nothing can be proven.
- In math things can be proven.
- computer science is not a science because you can prove things
- computer science is a kind of math because you can prove things
You guys don’t understand what math is and you don’t understand what science is and thus you guys are classifying incorrectly that computer science isn’t a math and they are making a completely invalid comparison with physics which is actually a science.
If you start with the premise that computer science is math, of course you'll reach the conclusion that computer science is math. And that's why you're so surprised that anyone else can reach any other conclusion. But the issue is in your premise - that's what you need to re-examine.
The math portions of computer science are math - and the non math portions are not. Your argument is equivalent to saying that Europe is defined by the counties in that continent using GMT - and concluding "Of course Italy isn't in Europe!".
Parts of Computer Science (the theoretical parts) sit in formal science just like logic and math do. But parts of CS like applied CS don't. All of the hardware design and software falls into applied CS and eventually slides into pure engineering.
But CS also is more amorphous than many other disciplines. Parts of it almost touch up against behavioral science with its subfield of linguistics, and syntax and semantics of language. And a failure of CS over the past 50 years has been the utter lack of progress in empirical data around human modeling of computation. We still don't have any empirical framework for evaluating what languages or modeling approaches are more efficient than others, or even what makes a more efficient programing language. As a result that is stuck in alchemy and cargo culting.
Models of computation and information theory begin to intersect with the natural sciences n the realm of physics. Physicists begin to push up against what is computable by nature in a given region of space. And Quantum computation begins to redefine what the physical bounds are on computation.
Don't begin by looking at what math is - if you do then the only thing possible to conclude is that it's math . An informal definition you could begin with might be "The theory and application of tools (ie formal cs) and models (ie applied cs) that predict and explain behavior related to Computation"
How do you model the real world via computation? What are the limits of computation? How do you implement computation?
Computer Science is the study of all aspects of computation.
The term is "theoretical computer science." Not just "computer science"
https://www.wikiwand.com/en/articles/Computer_science
https://www.wikiwand.com/en/articles/Theoretical_computer_sc...
The OP is making a claim on "theoretical computer science" NOT "computer science". Hence the title "What is theoretical computer science".
If you look at the wiki entry you will see that "Theoretical computer science" is highly mathematical and different from "computer science".
Honestly even I mis-used the term and this leads to tons of confusion. The naming of everything really changes the way people think and mixes everyone up. It's entirely a linguistic problem.
Some things can, some cannot (be either proven or shown to be false).