docs: add Chinese language support and improve markdown formatting#326
docs: add Chinese language support and improve markdown formatting#326hamberluo wants to merge 7 commits intoRetroAchievements:mainfrom
Conversation
Made-with: Cursor
Made-with: Cursor
|
Just wanted to add in that this PR was generated via AI, clearly not specified by this user. Would double check whatever "consistency" this inaccurate AI-generated translation contributes to this PR. |
thank you for your check, i will also check and learn. |
Made-with: Cursor
Made-with: Cursor
…umentation - Fix untranslated 'unforeseen' in why-delta.md - Fix English content in VitePress component examples in contributing-to-the-docs.md - Translate all alert box examples to Chinese (info, tip, warning, danger, details) Made-with: Cursor
Made-with: Cursor
Made-with: Cursor
|
I'm a volunteer and can't possibly review a 14,000 line long PR. As a GDE, you should know better. There is insufficient data to support this change. Our Discord does not have a Mandarin-speaking community. Our website does not support zh_CN. Traffic levels from China, Taiwan, and Hong Kong are very low, with the exception of IP addresses known to be associated with data centers. Given these facts, I'm unsure how this translation wouldn't drift over time, which would be a disservice to any Mandarin-speaking users who are interested in RA, as they would be getting the wrong information. My perception is those users would be better-served by a current machine translation rather than an outdated one. |
Yes, I have checked all the articles. My mention of MR itself indicates that this was my voluntary effort. Actually, having AI maintain it is the lowest-cost approach. Just look at the other two languages; They are already outdated. |
|
Translations should be driven by user need, not speculative. 120+ translated files will drift the moment any English source doc changes. To your point, this is already happening with other languages. Any plan to maintain these using AI is unsustainable for us. We're volunteer-run, and my personal belief is outdated translations are worse than none at all because they erode trust. Furthermore, this PR is unreviewable. I have a responsibility to not merge what I can't verify. I have no choice but to close this one out. We don't have the capacity to maintain this at the moment, and if this changes in the future, we should start with a smaller-scoped proposal with some sort of plan to keep things sustainable. What I'd really like to see is our docs somehow hooked into Crowdin, but that's a larger-scale project I haven't explored yet. |
|
It depends on who takes the first step—it's a chicken-and-egg problem. Does having Chinese documentation attract more Chinese contributors? I haven't done large-scale documentation integration with Crowdin before. However, if a maintainer of Chinese documentation is needed, I'm willing to take on the responsibility, provided the organization accepts me. |
|
I understand what you mean about a potential chicken-and-egg problem. I think the answer for us is actually neither side of the chicken-and-egg. Browser-based machine translation isn't too far off from what Cursor has output in this PR, and it has zero maintenance cost. Shipping static translations means those translations start drifting the moment any English source doc changes, and then users get a worse experience than they would from enabling Google Translate on the up-to-date English page. The existing PT and ES translations are a good example of this problem. They're well-intentioned, but already well out of date. Without someone constantly on them, that's how the situation will be ad infinitum. Those languages have significant community presence - they account for roughly 20% of RA's traffic. If anything, I think those translations are something we should reconsider maintaining long-term rather than a precedent we should expand on. For a volunteer-run project our size, I think leaning on machine translation for these docs is the right call. I appreciate the offer to maintain this, but I have to keep this one closed. |
|
I understand your point about document updates. In that case, I actually think it might be better to delete other outdated translations. Perhaps in this scenario, machine translation is more suitable. |
This PR adds comprehensive Chinese (zh) language support to the RetroAchievements documentation and improves markdown formatting consistency across multiple files.
Major Changes:
Chinese Language Support
docs/.vitepress/config/zh.ts)Markdown Formatting Improvements
Files Modified:
docs/.vitepress/config/index.ts,shared.ts, newzh.tsThis contribution significantly expands the accessibility of RetroAchievements documentation for Chinese-speaking users while maintaining consistency with existing documentation standards.