Preferences stay in this browser. Browser zoom (Ctrl/⌘ +) and pinch zoom remain available.
04 / POINTS OF DISAGREEMENT
Several questions inside one dispute.
Arguments are compared without assigning unstated positions or equating one member’s comment with project-wide consensus.
01Localization
Is Traditional Chinese just character conversion?
The contributor and ddman emphasize Taiwanese terminology and cultural context. arp242 later explicitly says he does not oppose adding Traditional Chinese. Demand for the locale and the method of this contribution should be assessed separately.
The contributor says tools do not determine quality and he takes responsibility. The maintainer expects carefully vetted human translations. A claim of reading through the work, an AI review and a named human language review are different kinds of evidence.
The contributor asks for a written AI prohibition. brlin-tw also emphasizes observing established conventions. These are different expectations, not proof that the whole team adopted a complete AI policy.
The PR changes shared templates as well as prose, creating technical review work. The contributor offers a smaller scope. brlin-tw criticizes excessive explanation while also asking maintainers for actionable feedback.
The maintainer alleges agitation; brlin-tw criticizes the publicity strategy; pcchou sees possible benefits. The original post does not explicitly invite an influx. In the separate publication post, Brune Kz suggests that several Taiwanese readers comment to attest to readability. That proposal does not prove author coordination or actual collective action.
Participants challenge the prompt characterization, mechanical-translation argument and disparaging wording. Wording and status changes can be checked. Allegations about motives, discrimination and honesty must remain distinct from established facts.
Russian review, tools and locale labels: what does the wording support?
Two GitHub comments were rechecked using public links supplied by a reader, and the BBC Chinese article was read in the browser. This is sourced editorial analysis, not new participant comments. It does not add to the comment count or change existing capture cutoffs. Checked: 27 Oct 2026, 14:44:12 (Taiwan, UTC+8); UTC 2026-10-27T06:44:12Z.
01Russian comparison
Does saying he cannot check Russian imply he speaks Chinese?
Verified wording: The 2025 comment in #86 identifies two limitations: arp242 does not speak Russian and cannot even build the website. It therefore says he cannot check the work. Extracting only the language limitation omits the build context.
Limits of inference: The comment does not say that speaking a language is a prerequisite for judging a contribution method. Assessing the AI workflow, submission description or trust in #99 does not require Chinese ability. These two comments do not prove that he speaks Chinese or guarantee that opening the next issue in Chinese would communicate effectively.
Unconfirmed: These sources supply no evidence of arp242’s Chinese ability or a line-by-line language review of this translation. Differences between PR responses merit comparison but do not, by themselves, establish a motive for differential treatment.
Is rejecting an AI translation while suggesting readers use machine translation contradictory?
Verified wording: #99 distinguishes readers obtaining mechanical translations for themselves from toml.io formally publishing carefully vetted human translations. Personal reading and accepting responsibility for published content and maintenance are different use cases. The comment does not require all Traditional Chinese readers to use machine translation.
Remaining disagreement: That distinction explains the maintainer’s position but does not settle how to assess AI drafts reviewed line by line by a human. The contributor reports reviewing his work; whether that review is sufficient and how to make it verifiable still require inspection of the result. A categorical tool-based exclusion and an outcome-based assessment remain materially different approaches.
What kind of generalization extends one dispute to future zh-TW translations?
Verified wording: arp242 first says he does not object to Traditional Chinese or other languages. Later he extends suspicion to any future zh-TW translation, using any zh-TW and AI translated slop, and ends sarcastically. Keeping only his stated openness to a locale would omit this expansion.
Editorial analysis: Extending one submission dispute to unseen future work in the same locale broadens suspicion on a language-locale basis. Readers may feel disparaged by association; criticism of this wording and its fairness has a concrete textual basis. This analysis of possible effects does not speak for every reader’s feelings.
Limits of inference: In context the statement expresses suspicion about zh-TW translations. It does not literally declare every Traditional Chinese user or every term they use to be an AI product. Suspicion is also not a verified finding. The passage alone cannot establish an internal motive of hostility toward an entire community or one cause for everyone’s anger.
What does the BBC’s AI-translation disclosure demonstrate?
Verified article: The BBC Chinese article of October 14, 2026 discusses efficiency, human review and contextual errors in AI translation, using professional examples to show that fluent text can still contain significant mistakes. Its closing disclosure says that this Chinese article itself was translated with AI assistance and reviewed by BBC journalists before publication.
Meaning of the comparison: This is a publicly disclosed publishing workflow combining AI assistance with human review. It helps distinguish tool provenance, review responsibility and result quality. It is not reporting on the TOML dispute or reviewing this translation, and cannot establish that the submission passes, must be accepted or necessarily contains errors.