Table of Contents

• [2018-11-24] my thoughts on that
• [[diffforms.pdf]] [[study]]
  • [[a 1-diff form is an **expression** F(x, y) dx + G(x, y) dy]]
  • [[total differential is an example of diff. form:]]
  • [[a diff form is similar to vector field]]
  • [[a form is exact if it’s a total differential of a scalar function]]
  • [[closed – if pF/py = pG/px]]
  • [[exact means that integral is path independent!]]
    • [[this is sort of similar to complex analysis?]]
  • [[wedge space: given vector space V:]]
  • [[2-form: an expression, built using wedges on pairs of 1-forms]]
    • [hmm. are all formal experessions looking like F dx^dy + G dx^dz + H dy^dz looking like that??](#hmmrllfrmlxprssnslknglkfdxdygdxdzhdydzlknglktht TIDDLYLINK)
  • [[rules for derivative:]]
  • [[orientation: for a curve, direction; for a surface – normal direction]]
    • [[so basically, by definition: integral of 2-form would be]]
    • [[what would it mean to eval surface integral of dx ^ dy + dx ^ dz + dy ^ dz over a sphere?]]
    • [ah! so it’s actually a flux: int int_S F \dot dS = ∫ int_S F₁ dy^dz + F₂ dz ^ dx + F₃ dx ^ dy](#hstsctllyflxntntsfdtdsntntsfdydzfdzdxfdxdy TIDDLYLINK)
    • [[NOTE arclength is NOT a differential form (always > 0)]]
    • [[the manifold has to be parameterised for defining diff forms (or at least, for defining integrals?)]]
  • [[page 50: Maxwell’s equations. EM field is a 2-form]]
• [https://en.wikipedia.org/wiki/Differential_form#Intrinsic_definitions](#snwkpdrgwkdffrntlfrmntrnscdfntns pull TIDDLYLINK)
  • [[By the universal property of exterior powers, this is equivalently an alternating multilinear map]]
• [2018-11-06] differential forms ^ : wedge/exterior product
  • [[do in agda?…]] [[agda]]
• [2018-11-25] ok, so diff forms are special types of tensors. Not all tensors are diff forms [[tensor]]
• [2018-11-14] calculus - What does (dx) mean in differential form? - Mathematics Stack Exchange
  • [[right, here it’s a 1-form! so no concerns about antisymmetry etc..]]
  • [[ok, but what if it was curvilinear? What would dx mean then??]]
• [[what are the other examples of 1-forms?]]
  • [[gradient – ok, definitely pretty intuitive]]
    • [[hmm. gradient is a vector and total derivative is covector?? https://math.stackexchange.com/questions/47618/definition-of-the-gradient-for-non-cartesian-coordinates]]
    • [[The 1-form ̃dfsuch that for any infinites-imally small vectordxfromTp ̃df(dx) =df(3.5)is called the gradient at point P . ok – that makes way more sense!]]
    • [[dual 1-form. dual covector field for vector field; is that a thing???]]
    • [[ugh. I guess alla of them are basically derivatives since well.. there isn’t much else apart of derivatives there and they form complete basis]]
• [[some intuitive stuff from NotesOnVectors.pdf]]
• [2018-11-10] Дифференциал (дифференциальная геометрия) — Википедия https://ru.wikipedia.org/wiki/%D0%94%D0%B8%D1%84%D1%84%D0%B5%D1%80%D0%B5%D0%BD%D1%86%D0%B8%D0%B0%D0%BB_(%D0%B4%D0%B8%D1%84%D1%84%D0%B5%D1%80%D0%B5%D0%BD%D1%86%D0%B8%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D1%8F_%D0%B3%D0%B5%D0%BE%D0%BC%D0%B5%D1%82%D1%80%D0%B8%D1%8F) pull
• [2018-11-10] differentiation in nLab https://ncatlab.org/nlab/show/differentiation
• [2018-11-18] differential geometry - What is a covector and what is it used for? - Mathematics Stack Exchange
• [2018-11-24] Closed and exact differential forms - Wikipedia
• [2019-01-24] A Visual Introduction to Differential Forms and Calculus on Manifolds | Jon Pierre Fortney | Springer [[viz]]
  • [2019-06-12] ok, so illustrations are ok and there are many of them. explanations seemed ok as well I guess

[2018-11-24] my thoughts on that

so basically, treat dx, dy as formal symbols? you can multiply them by scalar, add them up
NOTE in agda it would be just n-form with ordered requirement? or just a subset?

diffforms.pdf [[study]]

a 1-diff form is an expression F(x, y) dx + G(x, y) dy

total differential is an example of diff. form:

df = pf/px dx + pf/py dy

a diff form is similar to vector field

a form is exact if it’s a total differential of a scalar function

closed – if pF/py = pG/px

exact impllies closed, but not in reverse

exact means that integral is path independent!

this is sort of similar to complex analysis?

wedge space: given vector space V:

wedge rules: v^u = -u^v; u ^ u = 0; and linearity
build the space: \Wedge² V = {Sum_i u_i ^ v_i | u_i ^ v_i | u_i, v_i in V} are imposed

2-form: an expression, built using wedges on pairs of 1-forms

hmm. are all formal experessions looking like F dx^dy + G dx^dz + H dy^dz looking like that??

rules for derivative:

linearity
d(f alpha) = df ^ alpha + f d alpha
d (dx) = d (dy) = d(dz) = 0 # TODO hmm what does that one mean??

orientation: for a curve, direction; for a surface – normal direction

basically, continuous normal field. If it exists, there are two of them: n and minus n
parameterise the surface by two coordinates u, v so that du ^ dv is the direction of normal

so basically, by definition: integral of 2-form would be

int F dx^dy + G dy^dz + H dz ^dx = int int [ F p(x, y)/p(u, v) + G p(y, z)/p(u, v) + H p(z, x)/p(u, v)] du dv

what would it mean to eval surface integral of dx ^ dy + dx ^ dz + dy ^ dz over a sphere?

ah! so it’s actually a flux: int int_S F \dot dS = ∫ int_S F₁ dy^dz + F₂ dz ^ dx + F₃ dx ^ dy

NOTE arclength is NOT a differential form (always > 0)

but what we can do is to define a metric tensor

the manifold has to be parameterised for defining diff forms (or at least, for defining integrals?)

page 50: Maxwell’s equations. EM field is a 2-form

https://en.wikipedia.org/wiki/Differential_form#Intrinsic_definitions [[agda]]

[2018-11-25] ok, so diff forms are special types of tensors. Not all tensors are diff forms [[tensor]]

https://en.wikipedia.org/wiki/Antisymmetric_tensor https://en.wikipedia.org/wiki/Volume_form#Riemannian_volume_form https://math.stackexchange.com/questions/664648/what-does-dx-mean-in-differential-form/664674#664674
dx₁ is a differential 1-form, linear map acting on tangent space. Action is: (1, 0, …. 0)

dx_i is a covector field. {dx_i}_i span all covector fields. For f ∈ Rⁿ -> R, you can write df = [f₁(x) f₂(x) .. f_n(x)] as f₁(x) dx₁ + … + f_n(x) dx_n
in that sense, dx₁ is the derivative of coordinate function f(x_1…xn) = x₁

right, here it’s a 1-form! so no concerns about antisymmetry etc..

so, to compute the 1-form along the path, we:
split the path gamma(t); t from 0 to 1 into k parts
each part can be considered a tangent vector at gamma_i, and plugged into the form. then sum up the numbers

ok, but what if it was curvilinear? What would dx mean then??

what are the other examples of 1-forms?

gradient – ok, definitely pretty intuitive

hmm. gradient is a vector and total derivative is covector?? https://math.stackexchange.com/questions/47618/definition-of-the-gradient-for-non-cartesian-coordinates

hence gradient depends on metric
ok, so I guess definition might involve coordinates; but otherwise they are coordinate free once we prove they transform properly

The 1-form ̃dfsuch that for any infinites-imally small vectordxfromTp ̃df(dx) =df(3.5)is called the gradient at point P . ok – that makes way more sense!

dual 1-form. dual covector field for vector field; is that a thing???

ugh. I guess alla of them are basically derivatives since well.. there isn’t much else apart of derivatives there and they form complete basis

some intuitive stuff from NotesOnVectors.pdf

Examples of how you can picture contravariant and covariant vectors. A contravari-ant vector is a “stick” with a direction to it.  Its “worth” (or “magnitude”) is proportional tothe length of the stick.  A covariant vector is like “lasagna.”  Its worth is proportional to thedensity of noodles; that is, the closer together are the sheets, the larger is the magnitude ofthe covector.  These and other pictorial examples of visualizing contravariant and covariantvectors are discussed in Am.J.Phys.65(1997)1037.Figure  3:   Pictorial  representation  of  the  innerproduct between a contravariant vector and a co-variant  vector.   The  “stick”  is  imbedded  in  the“lasagna”  and  the  inner  product  is  equal  to  thenumber of noodles pierced by the stick.  The in-ner product shown has the value 5.

[2018-11-10] Дифференциал (дифференциальная геометрия) — Википедия https://ru.wikipedia.org/wiki/%D0%94%D0%B8%D1%84%D1%84%D0%B5%D1%80%D0%B5%D0%BD%D1%86%D0%B8%D0%B0%D0%BB_(%D0%B4%D0%B8%D1%84%D1%84%D0%B5%D1%80%D0%B5%D0%BD%D1%86%D0%B8%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D1%8F_%D0%B3%D0%B5%D0%BE%D0%BC%D0%B5%D1%82%D1%80%D0%B8%D1%8F)

https://ncatlab.org/nlab/show/differentiation

1. Idea
Differentiation is the process that assigns to a function f:X→Y f : X \to Y its derivative, sometimes denoted df d f. The derivative is a function that, roughly speaking, assigns to each point x∈X x \in X the linear transformation dfx d f_x that maps infinitesimal differences y−x y - x (for points y y infinitesimally close to x x) to infinitesimal differences f(y)−f(x) f(y) - f(x).

[2018-11-18] differential geometry - What is a covector and what is it used for? - Mathematics Stack Exchange

https://math.stackexchange.com/questions/240491/what-is-a-covector-and-what-is-it-used-for

[2018-11-24] Closed and exact differential forms - Wikipedia

https://en.wikipedia.org/wiki/Closed_and_exact_differential_forms

[[viz]]

• State "START" from "TODO" [2019-02-27]

https://www.springer.com/us/book/9783319969916

[2019-06-12] ok, so illustrations are ok and there are many of them. explanations seemed ok as well I guess