An apple pie from scratch $\succ$ Outline $\succ$ Drawing arrows and coding functions |
Here is an outline structure (work in progress):
Eventually, make a list of mutually properly distinct physical frameworks - i.e. mathematical theories used in physics, usually coming with some state space and dynamics, see Perspective. In the latter sense, they must be exhaustively specified, of course, together with a concise list of their respective axioms. Some frameworks contain black boxes that are further given insight into in more detailed theories. Here, when it comes to physics, I tend to like to have the rougher theories first.
'that list' |
Now one can reference real life systems and associate mathematical models in this and that framework with them. Develop them like this:
mechanics |
todo: examples of experiments
… phenomenological thermodynamics |
todo: examples of experiments
(Phenomenological thermodynamics leads to a lot of basic chemistry)
electronics |
todo: examples of experiments
electromagnetic field theory |
Preface |
---|
An apple pie from scratch |
Guideline |
Outline (this entry) |
Logical prerequisites
Meta |
---|
On syntax |
Symbols |
Types, terms and programming |
up to dependent type theory (?MLTT, etc.) |
type of natural numbers |
type of categories |
informal Curry-Howard correspondence |
Foundational temp1 |
Formal logic |
Intuitionistic logic (independent of type theory) |
First order logic (i.e. adding LEM to Intuitionistic logic) |
Axioms for set theory (e.g. GT-set-theory, and then U= type of sets) |
Mathematical core
todo: the following line of “apple pie from scratch” entries have some gaps
I have neither been able to find a good n-category basis instead of the 1-category approach I take here
nor have I formalized “up to iso” anywhere (this point is relate to the one above)
Formal part |
---|
On category theory basics |
On universal morphisms |
Foundational temp3 |
Foundational temp4 |
… |
… mathematics of statistical physics |
On universal morphisms |