Skip to content

Erdős Problem #1082

Editorial depth: Authored explanatory DRAFT — NOT editorially accepted.
Snapshot classification: falsifiable (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.

UNACCEPTED DRAFT — mathematical editorial review pending

This material is a provisional, source-qualified manuscript published for line-by-line scrutiny and real HTML/LaTeX testing. It has NOT received separate GCL editorial acceptance or independent mathematical certification. Its presence on this website is not evidence of a solution, reviewed coverage, or final publication.

Review the underlying draft on GitHub.

The question in plain language

Take \(n\) distinct points in the plane, with no three collinear. How many different distances must occur?

Erdős Problem #1082 actually contains two questions. The first asks whether the entire point set determines at least \(\lfloor n/2\rfloor\) distinct distances. The stronger version asks whether one particular point must have that many different distances to the other points.

The distinction is decisive. The stronger version is false, while the general, global version remains open.

A small example

Build a square of side length \(1\), and erect an outward equilateral triangle on each side. Retain the square's four corners and the four outer triangle tips.

Write \(r=\sqrt3\). Exact coordinates for the eight points are

\[ (0,0),\ (1,0),\ (1,1),\ (0,1), \]

and

\[ (1/2,-r/2),\ (1+r/2,1/2),\ (1/2,1+r/2),\ (-r/2,1/2). \]

No three points are collinear. Among all \(28\) unordered point pairs, the complete set of squared distances is

\[ \{1,\ 2,\ 2+\sqrt3,\ 4+2\sqrt3\}. \]

Thus the configuration has four distinct global distances, meeting \(\lfloor8/2\rfloor=4\).

But each square corner sees only the three squared distances \(\{1,2,2+\sqrt3\}\). Each triangle tip sees only \(\{1,2+\sqrt3,4+2\sqrt3\}\). No point sees four different distances. The stronger, single-point conjecture therefore fails.

These identities, and the noncollinearity of all \(\binom83=56\) triples, were independently checked with exact symbolic algebra. No floating-point approximation is required.

What is known

The maintained catalogue credits Szemerédi with a weaker bound involving roughly \(n/3\) distances and associates the eight-point counterexample with Heiko Harborth, as described by Erdős and Fishburn (1997) and Fishburn (2002).

The historical result already disproved the stronger local conjecture before the recent formal work. The 2026 discussion reports a Lean-verifiable eight-point construction found by a DeepMind prover, while explicitly acknowledging the older construction.

The original global lower-bound conjecture, however, has not been disproved by this example. It holds with equality for these eight points.

GCL's current state

Authenticated GCL source audit #2353 separately records the two mathematical claims and gives the complete algebraic coordinates. It checks all \(28\) pairwise squared distances and the orientation determinants of all \(56\) triples.

The frozen formal-conjectures statement likewise separates the open global theorem from the disproved single-point theorem and links an external Lean proof. GCL has not independently kernel-replayed that development, and the original journal sources remain to be page-audited.

What remains

For the single-point claim, the next obligation is provenance and formal-verification review, not rediscovery: check the 1997 and 2002 constructions, then independently replay the external Lean proof with its exact axioms.

For the global claim, the substantial problem remains whether every noncollinear \(n\)‑point set determines at least \(\lfloor n/2\rfloor\) different distances. A finite eight-point configuration with exactly four distances cannot settle the general question.

Sources and claim boundaries

The finite geometric counterexample is independently checked. This article is still unaccepted, and no external theorem or formal proof has been certified by GCL.

Catalogue and evidence provenance

  • Pinned catalogue source: teorth/erdosproblems at 536408716724.
  • GCL issue: #2353 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
  • GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
  • Formal-statement coverage: A provider statement was included in the pinned snapshot. Presence does not mean the statement or proof has been checked.
  • External mathematical status: consult current problem page and its dated bibliography.

Back to catalogue · Compendium introduction