Content Quality: Well-structured News piece (Overview, What the Ballot Required, Contracts at the Center of the Dispute, What Happens Next, Effect on Compilers, What We Don't Know), about 1,000 words (1,002 by the pipeline counter; News range 400-1200). Neutral tone, no AI self-reference, no relative dates, no vote tally asserted. Title is 136 characters (under 150). Hedging is accurate: the cause (contracts) is attributed only to Fertig and Pusz, and Pusz's caveat that the national-body comments must be published to be sure is reproduced. No other published Herald article covers the ballot failure (only the March C++26 approval piece and the April GCC 16 piece exist), so this is a new story.
Source Verification: All five snapshots (source-0..4.html.gz, all HTTP 200, suspicious_patterns null for every source, so no scanner false positives to override) were decompressed and read from raw text (HTML stripped; PDFs via pdftotext). INDEPENDENCE RECOUNT: 0 of 5 sources are independent of the dispute in the strict sense. heise (source-0) is an allowlisted editorial outlet but the piece is a guest column by Andreas Fertig, a WG21 member who is part of the German national-body vote and who writes a personal-opinion section calling the result a 'riesengroßes Desaster' (not reproduced by the article); Pusz (source-1) is a WG21 voting member who states he loves contracts and prefers to keep them; the three papers are WG21 participants' own positions. This is a News item, not an Analysis, so the 2-independent minimum does not apply, but the central fact (ballot disapproved) is reported by two separate participant accounts (Fertig; Pusz quoting convenor Seacord's LinkedIn post) that agree, and every contested claim is attributed by name. No wire or journalist report was cited. (0) heise, Fertig, published 02.11.2026: 'Am 23. Oktober ist das Undenkbare geschehen', 'Nach zwölfwöchiger Abstimmungsphase liegt das Resultat vor: Disapproved!'; 'Soweit mir bekannt, ist es das erste Mal in der Geschichte von C++, dass eine DIS-Abstimmung ... mit Nein endete' (article: 'as far as he knows, first time', attributed to Fertig only, correct); he says he is a committee member and forms the German vote with colleagues (confirmed); 'Aus den von den NBs abgegebenen Kommentaren lässt sich der Grund für das Nein recht gut ablesen: Contracts' with 'eine Anzahl von NBs ... dieses Feature bisher nicht in einem Release-fähigen Zustand' (confirmed); two-thirds majority not reached (confirmed); four options, all mean delay, naming question open (confirmed); 'Es beginnt am 16. November im brasilianischen Búzios' (confirmed). Fertig does not state a vote tally. Note: Fertig, as a committee member, reads the comments, while Pusz says they are not yet published; the article attributes each correctly. (1) train-it.eu, Pusz, 25 Oct 2026: 'C++26 failed its DIS ballot' (exact); Seacord post 24 Oct 2026 'Balloting has completed for C++26. Unfortunately it was Disapproved' (exact, line-wrapped); 'The ballot closed on 23 October 2026'; two-thirds / one-quarter criteria quote exact; ISO policy bars sharing how any NB voted ('So I report only that the ballot failed'); full report within 12 weeks, 'mid-January 2027 at the latest'; 'I expect the No votes to be about contracts, and we will know for sure only when the comments are published' (exact); four courses and SC22 chair chooses one of the first three 'in cooperation with the SC22 secretariat and the WG21 convenor' (confirmed; ISO central-office consultation omitted, immaterial); Búzios 16 to 21 November 2026; Garcia and Vandevoorde quotes exact (the article's closing period inside the Vandevoorde quote is not in the source, immaterial); Vandevoorde as WG21 Direction Group member, 'C++27 would make sense', ISO/IEC 14882:2024, GCC 16 -std=c++26 / Clang -std=c++2c, Clang contracts not merged: all confirmed. (2) P4433R0, 'Do not remove Contracts from C++26', Date 2026-10-23, 16 authors listed (counted: Doumler, Berne, Ažman, Bindels, Dimov, Fiselier, Fracassi, Halpern, Honermann, Lakos, Liber, Lippincott, Rivera Morell, Rosten, Schultke, Xie): 'On 22 October 2026, the DIS ballot for C++26 concluded' (so the 22 vs 23 October conflict is real; the article states it and does not pick one); 'a small number of individuals have actively lobbied NBs against approving the C++26 DIS' exact; 100/14 and 114/12 WG21 plenary votes exact; 'we should seek another ballot without altering contract assertions or removing them from C++26' exact; poll request for Búzios confirmed; 'This is not realistic' follows the P4238R1 delay estimate and the C++29 return suggestion. (3) P4238R1, 'Returning C++26 for the Evaluation It Skipped', 2026-09-01, authors Falco, Voutilainen, García Sánchez, Spicer (confirmed): 'the National Bodies should vote No, with technical comments, on the C++26 DIS ballot' exact; 'one ballot cycle: a revised draft and a second vote' and 'a delay of months' exact. The paper states it is the C++ Alliance's position (Falco is its president). (4) P4334R0, 'P2900 Contracts' fundamental flaws', 2026-08-09, authors Stroustrup, J-Daniel Garcia, Falco, Spicer, Voutilainen (confirmed): 'We are not fundamentally against idea of contracts' and 'What we are against is the bloated and incomplete P2900 contracts' exact. HOST VERIFICATION: open-std.org is the official ISO/IEC JTC1/SC22/WG21 site (its home page is titled 'ISO/IEC JTC1/SC22/WG21 - The C++ Standards Committee', its 2026 papers index lists P4238R1, P4334R0 and P4433R0, and isocpp.org's Submit a Proposal page links to open-std.org/jtc1/sc22/wg21/docs/papers/); train-it.eu is Mateusz Pusz's company site (github.com/mpusz links https://train-it.eu, names Train IT as his company and says 'Voting member of the ISO C++ Committee (WG21) since 2017, active in Library Evolution (LEWG)'; P4433R0 also cites [Pusz26]). Both hosts added to config/source_allowlist.txt with justification comments. Internal link: the article's link to the earlier Herald GCC 16 piece resolves to src/content/articles/2026-04/09-gcc-16-nears-release-with-c26-reflection-and-contracts-algol-68-frontend-and-c20-as-the-new-default.md, is described exactly ('GCC 16 nearing release with C++26 reflection and contracts'), uses the correct singular /article/ form, and that article does cover the GCC 16 contracts implementation and the 114-to-12 plenary vote.
Factual Accuracy: No fabricated specific found. The date conflict is reported faithfully: Pusz says the ballot closed on 23 October and Fertig dates the unthinkable event and result to 23 October, while P4433R0 says the ballot concluded on 22 October. The Overview says only 'in October'. 'First DIS ballot failure' is attributed to Fertig only (Pusz says 'as far as I know, it is the first C++ draft to fail one', also hedged; Fertig additionally notes the 2006 C++/CLI DIS 26926 did not pass, which concerns a different project and is not claimed by the article). The cause (contracts) is attributed to Fertig and Pusz only, with Pusz's caveat and the mid-January 2027 timing. Every 'all/every' statement is attributed or sourced: 'all four options mean a delay' (Fertig), 'every' does not appear as an unattributed universal. The two P4238R1/P4334R0 descriptions are exact in authors, dates and quotes. P4433R0's 'These claims are not true' refers to the specific claims the paper lists (no genuine consensus, hasty design, objections not heard, alternatives not considered); the article's wording 'claims in the petition and papers that accompanied that effort' is slightly broader than that list but is attributed to the paper and is not a misquote. The phrase 'two positions that preceded the ballot' covers two papers from an overlapping author group on the same side (Falco, Spicer, Voutilainen, Garcia appear on both); the opposing view is then given via P4433R0, which post-dates the ballot's close but was written before the result was known, as stated.
Overall Assessment: Accurate, well-attributed and appropriately hedged coverage of a significant development in C++ standardization; all claims verified against raw source text and no correction is warranted. Featured candidacy: 2 - first-ever failed ISO DIS ballot for C++ and a possible delay to the standard, significant for the broad C++ developer community, but still unfolding and sourced entirely to participants