Erdős Problem #67
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: proved (not automatically the live result).
Subject areas: discrepancy.
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
Imagine an infinite row of plus and minus signs. Choose a step size \(d\), and read the signs in positions \(d,2d,3d,\ldots\). Must some initial portion of some such progression show an arbitrarily large imbalance between pluses and minuses? Erdős conjectured yes; Terence Tao proved it.
The exact question asks whether every \(f:\mathbb N\to\{-1,+1\}\) satisfies
A small example
For a row of \(N\) signs, define the finite discrepancy
A complete enumeration of the \(4096\) possible twelve-sign rows shows that the smallest achievable \(D_{12}\) is \(2\), attained by \(370\) rows. The calculation is finite evidence, not a proof of the infinite theorem. If one appends another sign, \(D_N\) cannot decrease, because all the old progressions remain available.
What is known
Tao proved in 2015–2016 that the discrepancy is unbounded for every infinite sign sequence. His argument uses a Fourier-analytic reduction developed in the Polymath5 project, results on correlations of multiplicative functions, and the structure of functions resembling Dirichlet characters. The theorem also holds for sequences of unit vectors in real or complex Hilbert space. See Tao, Discrete Analysis 2016:1 and arXiv:1509.05363v6.
There is a distinct, stronger quantitative question: must the maximum discrepancy among progressions with \(md\le x\) grow at least as fast as a constant times \(\log x\)? The maintained catalogue records partial quantitative progress, but Tao's yes/no theorem alone does not establish this rate.
GCL's current state
The pinned catalogue and the maintained Erdős Problems page classify #67 as PROVED. The GCL work here is documentary: original source issue #1336. We have checked the finite \(N=12\) example, not independently certified Tao's deep proof. This article remains an unaccepted local draft pending an authenticated source return.
The original issue contains an authenticated RESULT/1 source report. This is unadjudicated source evidence only, not independent mathematical certification or editorial admission.
What remains
A bounded follow-up is to formalize the finite functional \(D_N\), its monotonicity, and the twelve-sign enumeration. Compactness supplies a useful logical bridge: the infinite theorem is equivalent to saying that for each integer \(C\) some finite length \(N(C)\) forces discrepancy greater than \(C\) for every sign vector of that length. The existence of these witnesses for every \(C\) is the substantive result, not something established by checking \(N=12\).
Sources and claim boundaries
- Tao, The Erdős discrepancy problem, Discrete Analysis 2016:1, arXiv v6.
- Maintained Erdős problem #67, captured 2026-10-12; pinned catalogue revision 536408716724a2402bdfa7c1d764bf83d64a8c89.
- GCL original source issue #1336.
No new solution, quantitative theorem, independent proof certification or editorial acceptance is claimed.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1336 — 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.