Erdős Problem #657
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: geometry, distances.
Read this status correctly
The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.
The question in plain language
Can a set of points avoid every isosceles triangle without producing an enormously large variety of distances? If \(n\) distinct points lie in the plane and every three points have three different pairwise distances, Erdős Problem #657 asks whether the complete set must determine more than a constant times \(n\) different distances, by a factor that increases without bound as \(n\) grows.
Write
The conjecture is that \(F(n)/n\to\infty\). Equivalently, for every fixed \(M>0\), every sufficiently large admissible planar configuration must have at least \(Mn\) distinct distances. Constructing just one family with many distances would not answer that universal question.
The maintained catalogue still lists the two-dimensional question as OPEN.
A small example
Consider the four corners of a nonsquare rectangle,
Any triangle formed by three of these corners has side lengths \(1\), \(2\) and \(\sqrt5\). It is therefore scalene, as required. The entire set has only these three distinct distances, even though it contains six unordered pairs of points:
Why is that possible? Every repeated distance belongs to two disjoint edges: the two vertical sides, the two horizontal sides or the two diagonals. Reusing a length at two edges meeting at a vertex would instead create an isosceles triangle.
A similar three-point fixture is any scalene triangle, which has three distances. Such examples illustrate what the geometry permits at small \(n\), but they do not produce arbitrarily large planar configurations with \(D(X)=O(n)\).
What is known
A useful exact linear bound. Regard the \(n\) points as vertices of the complete graph \(K_n\), and label each edge by its geometric length. The absence of isosceles triangles means that edges with the same label cannot meet at a vertex. Thus every collection of equal-length edges forms a matching: a collection of disjoint pairs.
A matching in \(K_n\) has at most \(\lfloor n/2\rfloor\) edges. Since \(K_n\) has \(\binom n2\) edges and each of the \(D(X)\) distances accounts for at most one matching,
Consequently,
The rectangle realizes the even \(n\) equality at \(n=4\). This elementary graph argument also appears in the maintained #657 public discussion. It is not a new GCL theorem. Crucially, an \(n-1\) lower bound is still only linear; it supplies no factor tending to infinity.
Why a one-dimensional analogy matters. For points on the real line, take any ordered triple \(a<b<c\). It makes an isosceles triangle in the distance sense exactly when
Thus line configurations without equal incident distances are the familiar three-term arithmetic-progression-free sets. This links the one-dimensional problem to additive combinatorics.
Adrian Dumitrescu studied related midpoint-free and \(\lambda\) free point sets in On distinct distances and \(\lambda\) free point sets, Discrete Mathematics 308 (2008), 6533–6538. The paper's publisher abstract explicitly addresses one-dimensional estimates and related geometry. The maintained Erdős entry summarizes subsequent progress on that one-dimensional question, including results using Bloom–Sisask and Kelley–Meka bounds for three-term progressions. None of those cited one-dimensional inequalities is claimed here as a theorem for arbitrary planar sets.
Erdős originally discussed the problem with Davies (in [Er73]) and later mentioned investigations with Füredi, Ruzsa and Pach (in [Er97e]), as reported by the maintained catalogue. Those original historical passages were not independently inspected in this audit.
Dimension matters. There are constructions in Euclidean spaces of dimension increasing with \(n\) that avoid isosceles triples with relatively few distances. Their existence does not disprove an assertion about the fixed two-dimensional plane. A proof or counterexample must keep the ambient dimension exactly two.
GCL's current state
The original authenticated RESULT/1 source return has historical disposition PARTIAL. This status and the original research exceptions are preserved, not silently converted into a theorem verdict. The maintained Erdős source and source-specific mathematical definitions were reviewed, and this article is editorially reviewed as a source-exception-annotated explanation under EC-BULK-005-R10, effective only after the exact-hash protected admission merges.
Source interpretation and unresolved mathematical qualification: Equal-distance edges form a matching, proving only the linear graph bound; the fixed-plane superlinear problem remains open.
The explanation is accepted for publication of established background, examples, and explicitly attributed current claims. It does not establish or formally certify the disputed original theorem, external model-assisted proof, historical interpretation, or any unreplayed Lean kernel file. Original source-fidelity and proof-review research holds remain.
What remains
To improve the known elementary argument, one must show that disjoint pairs of edges of the same length cannot be arranged as freely in the Euclidean plane as the matching bound allows in an abstract complete graph. In other words, the missing ingredient concerns the geometric compatibility between different matching color classes, not the elementary fact that each class is a matching.
A bounded research experiment could enumerate small configurations using exact squared distances, counting how often disjoint edge pairs may reuse a length without any vertex seeing equal lengths. It would need an explicit stopping rule and a rigorously stated geometric lemma before promoting numerical observations into mathematics. In particular, checking all configurations up to a fixed size is not a proof of superlinear growth.
The historical source-fidelity work remains to locate and read the exact [Er73], [Er75f, p.101], [ErPa90] and [Er97e] passages. Editorially, this draft still requires source-critical checking, consolidated protected merge, real public MathJax rendering verification and a separate exact-hash V2 admission.
Sources and claim boundaries
- T. F. Bloom, Erdős Problem #657: maintained statement, OPEN label and historical bibliography.
- Erdős #657 public discussion: recorded matching bound and comments distinguishing planar from one-dimensional results.
- A. Dumitrescu, On distinct distances and \(\lambda\) free point sets, Discrete Mathematics 308 (2008), 6533–6538. Publisher abstract inspected; full proof not independently reviewed.
- T. F. Bloom and O. Sisask, An improvement to the Kelley–Meka bounds on three-term arithmetic progressions, 2023: related one-dimensional additive-combinatorial interface, not a planar proof.
- GCL authenticated #657 return: pinned source identity and elementary verified fixture.
- Original Erdős [Er73], [Er75f], [ErPa90], [Er97e] historical references: bibliographic attribution only, exact relevant pages not inspected.
Authority boundary: established elementary graph counting and a finite explicit rectangle; no new theorem, resolution of superlinear planar growth, full original-source lock, Lean certification, protected production merge, deployed browser MathJax check or editorial admission.
Qualified editorial critical review: R10 source-exception dossier. The source-semantic exception is visible and its original research hold remains. Publication classification: REVIEWED EXPLANATORY ARTICLE WITH UNRESOLVED SOURCE/PROOF QUALIFICATIONS. Independent theorem proof or Lean certification: NOT CLAIMED.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1928 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: STATEMENT_CAPTURE_REQUIRED.
- Formal-statement coverage: No statement was confirmed in GCL's pinned formal-provider snapshot. Presence does not mean the statement or proof has been checked.
- External mathematical status: consult current problem page and its dated bibliography.