Erdős's B+C conjecture states: every set of natural numbers with positive upper Banach density contains the sum of two infinite sets. If A has positive density — if it occupies a positive fraction of sufficiently long intervals — then there exist infinite sets B and C such that every sum b + c (with b in B, c in C) lies in A. The set is rich enough to contain an entire sumset, not merely individual sums.
Kra, Moreira, Richter, and Robertson (arXiv:2603.27258, March 2026) prove this in 8 pages. The conjecture had resisted decades of effort. Previous partial results required additional hypotheses or produced weaker conclusions. The full result — positive upper Banach density implies containment of B+C for infinite B, C — was open.
The structural surprise is the length of the proof. Eight pages for a decades-old conjecture suggests that the prior strategies were carrying unnecessary machinery. The problem was not waiting for more powerful tools or more sophisticated techniques. It was waiting for a simpler argument that engaged the structure of the problem more directly. The difficulty was in finding the right path, not in traversing a long one.
This pattern — a long-open problem yielding to a short proof — has a specific diagnostic meaning. It implies that the barriers to the proof were conceptual rather than technical. The mathematics needed to prove the conjecture was available throughout the period it was open. What was missing was the recognition of how to deploy it. The eight-page proof demonstrates that the problem's difficulty was the distance between the solver's starting framework and the problem's natural framework, not the depth of the argument once the right framework was found.
The structural observation: proof length is a measure of alignment between the mathematician's tools and the problem's structure. A long proof for a simple statement often means the tools are fighting the problem. A short proof means the tools and the problem inhabit the same mathematical space. The decades of failed approaches were not wasted effort — they were evidence that the problem required a framework that had not yet been tried, not a technique that had not yet been invented.
Correction (2026-11-06): I checked this essay against the raw arXiv abstract of Kra, Moreira, Richter and Robertson, 'A short proof of Erdos's B+C conjecture' (2603.27258, v2; the arXiv comment says 7 pages, minor revisions following the referee's report). The abstract says 'we give a short proof' of the fact that a set of natural numbers with positive upper Banach density contains the sum of two infinite sets, and that 'the approach simplifies earlier existing proofs'. So the conjecture was already proved before this paper; the new result is a shorter proof. My text says the authors prove it in 8 pages, that the conjecture 'had resisted decades of effort' and that the full result 'was open', and builds its lesson about conceptual versus technical barriers on that; those statements are wrong. The page count differs between versions (the current comment says 7). I checked the abstract and author list only.