Open stop-cran opened 3 years ago
A nice way to define spheres, would be as SubspaceTopology (inverse_image (euclidean_norm_on_R^n) (Singleton 1))
. So some theory about finite products and finite dimensional euclidean spaces would be useful in this setting.
Using the Heine-Borel property of R^n we can then show that spheres are compact.
I made a sketch on a separate branch.
While playing around, I noticed that working with SubspaceTopology
is very clunky and needs a lot of boilerplate, translating between X
and { x : X | In S x }
. Doing this to functions & ensembles makes everything a lot more difficult.
How about introducing a notion of "open_in_subspace" or somesuch, as
Definition open_in_subspace
{X : TopologicalSpace}
(S : Ensemble (point_set X))
(U : Ensemble (point_set X)) :=
Included U S /\ exists U', open U' /\ U = Intersection S U'.
And similar notion for compactness? Or at least a lemma that maps from U : Ensemble (point_set (SubspaceTopology S)), open U
to open_in_subspace
.
The definition open_in_subspace
is unnecessary, now that subspace_open_char
is an equivalence. But more a definition of compact_subspace
would still be nice.
Some more example spaces & constructions:
Some possible constructions and isomorphisms:
Maybe it'd be useful to give some simple and practical examples to the defined topological constructions and proven theorems. E. g. a homeomorphism between a unit interval with identified ends and a unit circle.