{"id":"88dec02e-47d9-478e-aca4-703a476cff84","arxiv_id":"2607.28444","paper_version":1,"verdict":"UNVERDICTED","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"A documented lattice QCD authorship dispute shows structural gaps in how long-term research software and conceptual contributions are credited and adjudicated.","lead":"A lattice QCD researcher documents a credit dispute over software and ideas, arguing that authorship rules and German integrity procedures fail long-term research software work. The piece is a first-person case study meant to push community standards for software recognition.","discovery_kind":"review","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified beyond the Reader's already-flagged one-sidedness; the meta-claim does not require adjudicating the confidential dispute.","rationale":"The Reader correctly classifies the work as a first-person research-practice case study, sets UNVERDICTED, and isolates the real soft spot: interested-party chronology plus inaccessible ombuds evidence. That soft spot is real but not decisive for the claim the paper actually advances. The argument does not need the OWID outcome to have been substantively wrong; it only needs to show that existing mechanisms and authorship norms leave a recognition gap for long-lived research software—an observation already supported by the open software record, the parallel papers, DFG/OWID texts, and external RSE studies the paper cites. Treating the piece as requiring a machine verdict on named misconduct would over-reach; leaving it UNVERDICTED is appropriate. No stronger technical or logical fracture (hidden equation-level assumption, non-reproducible central derivation, etc.) exists to move the needle. Agreement with the Reader is therefore full on both the weakest link and the verdict.","tokens_in":21928,"tokens_out":558,"duration_ms":12253,"concrete_test":"Compare the public v1/v2 texts of Ref. 14 (acknowledgements, software citations, absence of citation to Ref. 13) against the GitStats summary in Fig. 1 and the DOIs for BaHaMAS, CL2QCD, Monte Carlo C++ analysis tools, and Script utilities; if those public records already document substantial long-term software reuse without authorship, the structural claim stands without needing the sealed OWID file.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's strongest claim is structural and normative (traditional authorship + German integrity/collaboration machinery poorly capture long-term research-software and related conceptual credit in lattice QCD; community standards may help), not a finding of misconduct. Publicly checkable pieces already carry most of that load: software DOIs and GitStats (Fig. 1) showing sustained infrastructure work after Ref. 15; parallel arXiv records (Refs. 13–14); DFG guideline 14 on conceptual design; OWID procedural limits the author quotes (no reciprocal evidence access, no appeal, confidentiality bars); and the cited RSE literature on under-crediting. The confidential OWID file and contested priority narrative are illustrative colour, not a necessary premise. Once the piece is read as a practice reflection rather than an empirical misconduct adjudication, the Reader's weakest_assumption does not undercut the central claim enough to demand a different verdict. No further internal inconsistency or hidden technical assumption is load-bearing.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript presents a first-person case study from lattice QCD on authorship and credit for long-term research software and related conceptual work. Using a documented sequence involving parallel arXiv submissions (Refs. 13–14), reuse of analysis and infrastructure codes, an OWID ombuds procedure, and subsequent collaboration decisions, it argues that traditional authorship norms and existing German research-integrity and collaboration-governance mechanisms do not adequately capture substantial software and conceptual contributions. Public elements include software DOIs, GitStats contribution summaries (Fig. 1), DFG guideline 14 on conceptual design, and quoted OWID procedural limits. The piece is framed as a factual practice reflection aimed at stimulating community discussion (e.g. via LDIC/ReSA) rather than as a physics result or a formal misconduct adjudication.","tokens_in":21972,"tokens_out":1341,"duration_ms":34591,"significance":"If accepted as a practice reflection, the paper is timely: lattice QCD depends on multi-year software stacks often built by early-career researchers, while career metrics remain publication-centric. Strengths include transparent Competing Interests disclosure; checkable public artifacts (arXiv records, software DOIs, Fig. 1 GitStats on named repos); explicit engagement with DFG guidelines, FAIR4RS, and the RSE/credit literature (e.g. under-crediting of code contributors); and a clear normative ask for community standards rather than a claim of proven misconduct. The structural point—that open reuse, legal ownership, and integrity procedures can leave software-heavy contributions under-recognised—does not require settling the confidential priority dispute and is of genuine interest beyond the single case.","major_comments":[{"comment":"§2 and §3: The central structural claim is supportable from public materials, but the manuscript still leans on contested, one-sided chronology (conception priority of the Ref. 15 follow-up; what OWID did or did not engage; relative developer status) as if those points were established facts. Please separate more sharply (i) independently checkable items (parallel uploads, software DOIs/GitStats, quoted DFG/OWID rules, published acknowledgements) from (ii) the author’s interpretation of confidential procedure and priority. The latter should be explicitly labelled as the author’s account of a disputed record, especially given the Competing Interests statement and the OWID finding of no wrongdoing as reported in the text.","section":"§2 Context; §3"},{"comment":"§3 (DFG guideline 14 / conceptual design) and the comparison of Refs. 13 and 14: The argument that conceptual design confers authorship rights is load-bearing for the case illustration. As written, readers cannot verify the claimed match between the programme suggested in Ref. 13 and the work in Ref. 14 without a concise, neutral side-by-side (scope, observables, simulation strategy) that does not presuppose the author’s priority narrative. Add a short factual comparison table or paragraph limited to published content so the guideline-14 discussion rests on citable overlap rather than assertion.","section":"§3; Refs. 13–14"},{"comment":"Fig. 1 and the software-credit claim: Commit/line counts are used to support “substantive technical contributions” and to criticise diluted developer attribution in Ref. 14. GitStats totals do not by themselves establish scientific indispensability or exclusivity of contribution to the specific analysis in Ref. 14, and the caption notes that main/master branches barely evolved after Ref. 15. Qualify what Fig. 1 does and does not show (sustained stewardship vs. novel code for the new study; multi-author history of CL2QCD etc.), and tie credit claims to the reused packages actually cited for that work rather than cumulative career commits alone.","section":"Figure 1; §2"}],"minor_comments":[{"comment":"Figure 1 is extremely dense (multiple stacked GitStats panels, tiny type). Consider a simplified summary table of commits/lines/active periods per codebase for the principal authors, with full GitStats moved to supplementary material or a repository.","section":"Figure 1"},{"comment":"§1–§4 occasionally shift from descriptive practice analysis into rhetorical questions that read as advocacy. Tonally aligning those passages with the Abstract’s “factual account and reflection” framing would improve suitability for a formal journal.","section":"§3; §4"},{"comment":"Footnote 1 and footnote 3 add important citation-propagation detail but interrupt the main line; consider folding the essential facts into the body and shortening the footnotes.","section":"Footnotes 1 and 3"},{"comment":"Minor language/typos: e.g. “wheresuchinfrastructuredoesnotyetexist”, “This situationraisesnontrivial”, “in medio stat virtusand”; normalise spacing and punctuation throughout.","section":"§1; §4"},{"comment":"Keywords line is unpunctuated and hard to parse; use semicolon- or comma-separated terms consistently.","section":"Front matter"},{"comment":"Ensure all software citations (BaHaMAS, CL2QCD, Monte Carlo C++ analysis tools, Script utilities) consistently list version, DOI/URL, and access date where relevant, matching Data Accessibility.","section":"References [19–23]; Data Accessibility"}],"recommendation":"minor_revision","confidential_remarks":"Scope fit is the main editorial question: this is a research-practice/ethics case study, not a lattice result. It is appropriate only if the journal (or a designated section) publishes community/practice pieces. Substantively I do not treat the piece as a misconduct finding—the OWID outcome and one-sided evidence are disclosed—and the structural RSE-credit argument can stand after the framing revisions above. If the journal cannot host contested first-person institutional narrative even when labelled as such, transfer or reject-on-scope is cleaner than forcing a physics-style rewrite. I see no need to adjudicate the underlying personal dispute to accept a revised practice reflection."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The useful part of this paper is not a new physics claim and not a finding of misconduct. It is a dated, named case chronology from lattice QCD that ties long-term software infrastructure (CL2QCD, BaHaMAS, analysis tools, Script utilities) and a conceptual follow-up (arXiv 13 vs 14) to how authorship and German integrity/collaboration machinery actually behaved. That is more specific than the usual RSE/FAIR/ReSA citations, and it is what makes the piece worth reading.\n\nWhat it does well: it stays mostly on structure. Figure 1 and the software DOIs give checkable volume and longevity of code work after the 2021 JHEP paper. Parallel arXiv records and the PoS withdrawal episode are public. DFG guideline 14 on conceptual design and the quoted OWID procedural limits (no reciprocal evidence, no appeal, confidentiality bars) are cited in a form you can verify. The institutional “short-circuit” point—integrity bodies punt governance issues back to the CRC, while the CRC waited on integrity—is the sharpest general observation. The call for community norms (LDIC, ReSA-style) is modest and already in the literature; the case just makes the gap feel real.\n\nSoft spots, in proportion: the narrator is a direct party with a professional stake, Competing Interests says so, and the confidential OWID file is not available. So you cannot treat the priority narrative or “they diluted my software credit” as settled fact. OWID found no wrongdoing. That weakens any misconduct-adjacent reading, but it does not erase the structural claim once you treat the dispute as illustration rather than proof. Self-citation of the author’s software and idea papers is inevitable in a first-person case study; the public artifacts still carry most of the load. No math to break; citation pattern to prior RSE work is adequate.\n\nWho it is for: people who care about research-software credit, early-career incentives, or lattice/HEP collaboration practice. Not for someone hunting a QCD result. I would send it to peer review as a methods/practice piece—editors should expect pushback on one-sidedness and may want tighter separation of public record from contested narrative, but it is serious enough for referee time. I would cite the structural discussion and the GitStats framing if I write on software credit; I would not cite it as establishing who conceived the follow-up. Engage if that is your beat; skip if you only want physics.","headline":"A named lattice-QCD case study that makes the familiar software-credit problem concrete with public repos, GitStats, and German integrity procedure limits; useful as practice reflection, not as adjudication of the dispute.","tokens_in":22757,"tokens_out":601,"would_cite":true,"duration_ms":12387,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A lattice QCD case study argues that long-term research software and conceptual work still fall outside fair authorship and integrity rules.","keywords":["research software","authorship","credit attribution","software sustainability","lattice QCD","research integrity","career paths","rewards and incentives"],"falsifier":"Independent comparison of the conceptual proposal paper with the later results paper, plus public commit histories of the cited codebases, showing either that the disputed ideas and software were not load-bearing or that equivalent credit was already given under clear prior agreement.","tokens_in":22601,"feed_emoji":"💻","tokens_out":866,"duration_ms":17510,"temperature":0.7,"pith_summary":"This paper uses a documented dispute in lattice QCD to show that traditional authorship and German research-integrity procedures do not reliably credit long-term research software and the conceptual extensions that grow out of it. The author recounts how analysis tools, simulation infrastructure, and a follow-up idea developed while finishing earlier joint work were reused in later papers while his authorship was withdrawn after an ombuds process that found no misconduct. He argues that software has become essential scientific infrastructure, yet career incentives still treat publication counts as the main currency, so people who invest years in maintainable code can lose both papers and positions. The piece is offered as a factual reflection meant to open community discussion, not as a re-litigation of the physics results. A sympathetic reader is meant to see why lattice QCD and similar fields may need explicit standards that treat sustained software contributions as first-class intellectual work.","feed_headline":"Software builders still lose authorship fights in lattice QCD","feed_subtitle":"A first-person case study says integrity rules and paper culture miss long-term research code","key_machinery":"The documented case chronology itself: parallel arXiv uploads, reuse of shared analysis and management codebases (with contribution statistics), an initial authorship agreement later withdrawn, an OWID ombuds assessment of no wrongdoing, and subsequent institutional hand-offs that left broader governance questions unaddressed.","core_discovery":"In software-intensive lattice QCD, the combination of long-term software infrastructure, numerical strategy, and conceptual follow-on ideas can create authorship and priority disputes that neither conventional publication norms nor existing German ombuds and university misconduct channels resolve in a way that captures the full contribution, producing structural tension between open collaboration, software sustainability, and traditional credit.","pith_inferences":["Similar credit gaps are likely in any HPC-heavy field where PhD and postdoc labor builds multi-year simulation stacks that outlive individual appointments.","Making software DOIs citable is necessary but not sufficient; without authorship or equivalent career currency, the incentive problem remains.","Transparent, two-sided access to evidence in integrity procedures would make later public case studies less one-sided and more usable for community learning.","Explicit contribution taxonomies agreed at project start could reduce later fights over whether infrastructure work counts as authorship."],"forward_implications":["If the diagnosis is right, lattice QCD groups that treat major internal software as uncredited tools will systematically under-reward the people who build them.","Career paths tied mainly to paper counts will continue to discourage investment in sustainable, well-documented research software.","Disputes that mix authorship, software, supervision, and collaboration governance can fall between institutional mandates even when each body follows its own rules.","Community bodies could usefully develop shared norms for when long-term software and conceptual infrastructure warrant co-authorship versus citation or acknowledgement.","Open-sourcing code without matching recognition norms may still leave early-career builders professionally exposed inside the same institute or collaboration."],"fun_headline_variants":["Lattice QCD case shows software infrastructure loses authorship fights","Long-term code work sparks unresolved credit disputes in lattice QCD","Authorship norms miss research software builders in lattice QCD study","Ombuds channels fail full credit for lattice QCD numerical infrastructure","Open collaboration clashes with authorship rules on lattice QCD software"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The author’s first-person timeline and contribution weights must be complete and accurate enough to support a general structural critique, even though he is a direct party and the confidential process found no wrongdoing.","fun_headline_variants_meta":{"raw":{"variants":["Lattice QCD case shows software infrastructure loses authorship fights","Long-term code work sparks unresolved credit disputes in lattice QCD","Authorship norms miss research software builders in lattice QCD study","Ombuds channels fail full credit for lattice QCD numerical infrastructure","Open collaboration clashes with authorship rules on lattice QCD software"]},"model":"grok-4.5","effort":"low","cost_usd":0.00329,"raw_usage":{"total_tokens":1082,"prompt_tokens":746,"num_sources_used":0,"completion_tokens":63,"cost_in_usd_ticks":32904000,"prompt_tokens_details":{"text_tokens":746,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":273,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":746,"tokens_out":63,"duration_ms":5812,"temperature":1.0,"reasoning_tokens":273,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T07:11:41.454744+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Independent comparison of the conceptual proposal paper with the later results paper, plus public commit histories of the cited codebases, showing either that the disputed ideas and software were not load-bearing or that equivalent credit was already given under clear prior agreement.","supporting_citations":[],"review_version":1}