GOLD STANDARD · TOPOLOGY · HOW X WORKS
Find the decision hidden inside the model
A crumpled cube is bounded by a sphere yet need not be a ball. The decisive evidence lies in how the boundary sits inside the chosen side: collars and controlled contractions of small complementary loops distinguish a genuine cell from a wild one.
This guide connects the Gold Standard of Topology, Gold Standard of Education, and How X Works hub.
Chapter contents and reading routes
Choose the route closest to your present task. Every chapter remains expanded, so the article also reads naturally from start to finish.
Orient crumpled cubes
Work the local test
Separate sides and sewings
Transfer and diagnose
Use evidence and next routes
CHAPTER 1 OF 12 · ORIENT THE CUBE
1. A sphere boundary does not guarantee a ball
Embed an (n−1)-sphere in the n-sphere and take the sphere together with the closure of one complementary domain. The resulting space is a crumpled n-cube. Its designated boundary is the embedded sphere; its interior is the chosen complementary domain.
If the embedding is wild, the closure can fail to be an n-cell even though the boundary is abstractly a standard sphere. The word cube records the intended role, not a promise of tame local geometry.
CHAPTER 2 OF 12 · ORIENT THE CUBE
2. The two sides can have different topology
A wild sphere has two complementary domains, and their closures need not behave symmetrically. For the familiar Alexander horned sphere, one chosen side may be a 3-cell while the other has complicated complementary loops.
Always name the crumpled cube as a side plus its boundary. Statements about the embedded sphere alone can hide which complement supplies the obstruction.
CHAPTER 3 OF 12 · WORK THE TEST
3. A collar is a product neighbourhood from one side
A boundary collar in C is an embedding of Bd C×[0,1] into C that fixes Bd C at the zero level. It supplies parallel copies of the boundary and a controlled inward direction.
For crumpled cubes, being a cell is equivalent to having the appropriate boundary collar. This is a one-sided assertion inside C; an ambient bicollar is stronger and corresponds to tameness from both sides.
CHAPTER 4 OF 12 · WORK THE TEST
4. Work the epsilon–delta 1-LCC test
Fix x on Bd C and a neighbourhood scale ε. The 1-LCC test asks for a smaller δ such that every loop in the δ-neighbourhood of x, missing Bd C and lying on the interior side, bounds a disk in the ε-neighbourhood that also misses Bd C.
A loop that survives at arbitrarily small scales signals local wildness. One loop contracting at one scale is not enough: the quantifiers require every point, every ε and all sufficiently small loops.
| Test element | Required statement | Common shortcut to reject |
|---|---|---|
| Point | Every x in Bd C | Only regular-looking points |
| Outer scale | Every ε>0 | One convenient neighbourhood |
| Inner scale | Some δ for that ε | Same δ used globally without proof |
| Loops | Every small loop bounds a disk | Selected meridians only |
CHAPTER 5 OF 12 · WORK THE TEST
5. Why 1-LCC detects a missing collar
A collar lets small loops be moved onto controlled product levels, where standard sphere-side topology provides contractions. Conversely, the deep theorem uses local loop control to tame the boundary sufficiently to recover cell structure, with the proof depending on dimension.
The intuition is useful, but the theorem is not merely an exercise in fundamental groups. Approximation and manifold-recognition machinery enter in difficult dimensions.
CHAPTER 6 OF 12 · SEPARATE THE SIDES
6. One-sided collarability is not ambient local flatness
A boundary can be well behaved from the chosen crumpled-cube side and wild from the other. A collar inside C proves that C is a cell; it does not automatically give a collar in the opposite closure or a bicollar in the ambient sphere.
To conclude the embedded sphere is tame, check both sides or invoke the exact local-flatness theorem. Preserve which complement every 1-LCC statement refers to.
CHAPTER 7 OF 12 · SEPARATE THE SIDES
7. Sewing tests expose boundary compatibility
Sew two crumpled cubes by a homeomorphism of their boundaries. Even when each boundary is an abstract sphere, the quotient can depend on how wild sets and complementary obstructions are matched.
Mismatch conditions and disjoint-disks properties provide sufficient or necessary controls in higher-dimensional sewing theorems. The identity map is not automatically harmless when the boundary carries wild embedding data.
| Question | Evidence needed | Possible conclusion |
|---|---|---|
| Is C a cell? | One-sided collar or 1-LCC complement in range | C≈Bⁿ |
| Is Bd C tame in Sⁿ? | Control from both sides or local flatness | Ambient bicollar |
| Does C∪hD give Sⁿ? | Sewing theorem hypotheses | Sphere quotient |
| Does a fibre shrink? | Whole decomposition control | Near-homeomorphism |
CHAPTER 8 OF 12 · TRANSFER AND DIAGNOSE
8. Changed-context transfer: one wild sphere to a quotient boundary
A crumpled cube starts with an embedded sphere. A cell-like quotient may instead produce a singular set whose link or boundary resembles a wild sphere. The same loop test can guide diagnosis, but the target may be a generalised manifold rather than an ambient sphere.
Transfer succeeds when dimension, local homology and the exact pair are rebuilt. Copying the word collar from a manifold boundary into a singular quotient can assume the result.
CHAPTER 9 OF 12 · TRANSFER AND DIAGNOSE
9. Dimension changes the proof route
Bing established key three-dimensional results, while dimensions four and five require special low-dimensional methods and higher dimensions use other general-position machinery. The visual epsilon–delta statement can look uniform even when its proof is not.
Cite the dimension-specific theorem being used. A result about a crumpled 3-cube cannot be promoted to every n solely because the definitions share symbols.
CHAPTER 10 OF 12 · TRANSFER AND DIAGNOSE
10. Limits and common misconceptions
Contractibility does not make a crumpled cube a cell, and an abstractly spherical boundary does not make it locally flat. Global simple connectivity of the interior is weaker than controlled 1-LCC behaviour at every boundary point.
A collar on one side does not certify the other side. Likewise, good individual pieces do not ensure an arbitrary sewing or decomposition shrinks; compatibility and simultaneous control remain separate questions.
CHAPTER 11 OF 12 · CHECK AND CONTINUE
11. Questions readers often ask
Is every crumpled 3-cube homeomorphic to a 3-ball?
No. The boundary is a 2-sphere, but wild embedding can make the chosen closure non-cellular. A collar or the relevant 1-LCC criterion is needed to identify a genuine 3-cell.
Why test loops instead of just points and neighbourhoods?
Wildness can be invisible in the intrinsic shape of the boundary yet recorded by linking in its complement. Small loops reveal whether that linking persists at arbitrarily fine scales.
CHAPTER 12 OF 12 · CHECK AND CONTINUE
12. Evidence, connected routes and the next check
The research on sewings of crumpled cubes develops boundary-mismatch and disjoint-disks conditions for obtaining a sphere. Bestvina, Daverman and Venema’s 1-LCC shrinking theorem shows how controlled complements support approximation in dimension four.
Continue through cellularity and 1-ULC complement tests, DDP and low-dimensional shrinking, the Alexander horned sphere, and Bing shrinking. The next check is to write the chosen side beside every collar and 1-LCC statement.