Review and update Vietnamese translation for Inno Setup 6.5.0+#613
Review and update Vietnamese translation for Inno Setup 6.5.0+#613Blackspirits wants to merge 1 commit into
Conversation
This PR performs a full review of the Vietnamese translation for Inno Setup 6.5.0+. Changes include: * restoring the missing `[name]` placeholder in `SelectDirLabel3`; * correcting grammar, punctuation and quotation mark issues; * improving fidelity to the English source; * normalizing Vietnamese terminology and capitalization; * removing overly informal or promotional wording; * improving consistency across installation and uninstallation messages; * preserving all keys, placeholders, line breaks and file structure. The translation has been technically validated against `Default.isl`: * 296 matching entries; * no missing, additional or duplicate keys; * no placeholder errors; * native language name preserved as `Tiếng Việt`. Since memecoder is listed as the original translator, I would appreciate a native review of the proposed changes before considering the translation ready for promotion from `Unofficial`.
|
I have also contacted the original translator by email and asked for a native review of the proposed changes. I will update this PR if I receive any corrections or confirmation. |
|
Thanks for letting me know, and apologies for the confusion. I opened the Serbian Latin PR almost at the same time as you posted your message on the Vietnamese PR, before I had the opportunity to read it and reply. I was therefore not intentionally ignoring your request. I am a translator and speak several languages, and my intention was to help improve as many translations as possible by checking them against the current I will not open any further pull requests without coordinating with you first. For the Vietnamese and Serbian Latin PRs that are already open, please let me know whether you would prefer me to leave them open for the maintainers to review later or close them for now. For Portuguese (Portugal), I believe the PR can be merged, since European Portuguese is my native language, I completed the full review, and I was also responsible for the previous review of that translation. Could you also let me know which languages, if any, would currently benefit from a review, and how you would prefer me to submit the results? I can prepare review reports without changing any files. Would you prefer me to open a GitHub issue for each language, or use another method? |
|
No problem. For the unofficial translations, I would love to make these official, in order of importance:
I have contacted the maintainers of Indonesian, Vietnamese, Greek and Serbian almost 3 weeks ago. The Indonesian and Greek maintainers have not yet responded. Next version (7.0.2, released next monday) makes Lithuanian, Simplified Chinese and Traditional Chinese official because the work for those has finished. Do note that making these official did not impose additional requirements on the maintainers. Other unnamed unofficial translations are either not up to date for 6.5.0, or they don't have a localized version of Windows. Latter applies to Farsi, which is why we don't want to include it as an official translation. For the official translations I have unfortunately noticed they are not of the quality I thought they were. Many do not follow our guidelines, or have other issues. This applies to all but Lithuanian, Simplified Chinese, Traditional Chinese, Corsican, Italian, Dutch and Korean because for those fixes were done recently. The Czech maintainer has proposals for fixes and it going to look at it later. I did fix all mechanical issues already for all official languages, of which there were many. Like missing or added periods, whitespace issues, obvious typos, etc. I think I would prefer a single PR targeting the existing Are your reviews done manually or AI assisted as well? If it's the latter I would like you to use same methodology as I used for consistency. It works well (gotten only positive feedback from maintainers so far about the findings) but it has a weakness: it still takes a ton of time, and it doesn't find everything at once. Can share skills for it. If it isn't clear: I do appreciate your help, I spent quite some time already on making reviews and making proposals, but the result isn't that great yet. Not because findings are incorrect or proposed fixes are incorrect (on contrary, gotten positive feedback) but because of it taking so much time, and because of maintainers not replying. Here's a list of findings I made earlier, some from before further improving the guidance. And all this a single batch of findings: after fixing there would likely be just as many new findings is my experience. Note that some maintainers are present on GitHub, but other aren't. Indonesian.isl
Greek.isl
SerbianLatin.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
SerbianCyrillic.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
BrazilianPortuguese.isl
Spanish.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Arabic.isl
French.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
German.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Japanese.isl
Tamil.isl
Turkish.isl
Thai.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Polish.isl
Ukrainian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Hungarian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Swedish.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Hebrew.isl
Danish.isl
Bulgarian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Catalan.isl
Russian.isl
Finnish.isl
Norwegian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Slovenian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Slovak.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
Armenian.isl
Finally, a few minor, optional points (not blocking, listed only for completeness):
|
|
Thanks for the clarification. Most of my translation and review work is done manually. I compare the translation line by line with As both a translator and a programmer, I also use AI during some review passes because it is very useful for spotting omissions, inconsistencies, placeholder issues and alternative wording that can easily be missed in a large file. I do not apply its suggestions automatically; every finding is checked manually before I accept or reject it. At the end, I also run an additional QA validation to confirm that the file still has the correct keys, placeholders, line breaks, accelerators and structure, and to surface any remaining inconsistencies or possible improvements for a final manual review. I am comfortable working with a single pull request targeting the During my repository review, I noticed a clear placeholder issue in I would also be interested in seeing the skills and methodology you used, so I can align future reviews with the same process. |
|
I also found two unrelated, non-translation issues during the repository review:
Would you be comfortable with me submitting these as two small, separate pull requests, or would you prefer another approach? |
|
Well, that simply sounds great, thank you very much for your help. If you can make that centralized PR and close the others, please do that. Missed the Hebrew.isl issue myself, it's not addressed elsewhere, so please include it. For automated reviews the repo already includes isl.md In my experience it's final paragraph about back translations is what makes these reviews find the correct things. Without this, the findings are not faitful to the source because they try to improve the source while translating, but only in the translation. TBH, as it turns out many (all?) of the existing translations from maintainers are doing this, some excessively, and it's a problem (to me). If you contact me at mlaan @ innosetup.nl I will also send you the skills I used myself for reviewing, and for fixing. BTW, if you weren't aware already, you may be interested to know that in 7.1.0 the IDE will be available in German, Dutch, Japanse and Czech. https://github.com/jrsoftware/issrc/blob/vnext/Projects/Src/IDE.LocalizeFunc.Czech.pas |
That would be great, thanks in advance. |
|
Thank you very much. I will prepare the centralized pull request targeting I will also submit the SBOM correction and documentation typo fixes as two separate small pull requests. I will contact you by email regarding the review and fixing skills. The upcoming IDE localizations for 7.1.0 are also very interesting—thanks for letting me know. |
|
Do note isl-update-proposals already has the Vietnamese update proposals I made myself. For Serbian I also suggest to wait for the maintainer, they will be back soon. Those changes are not in the branch, but in emails to the maintainer. |
|
Superseded by #615, which contains the completed Vietnamese review based on the current |
This PR performs a full review of the Vietnamese translation for Inno Setup 6.5.0+.
Changes include:
[name]placeholder inSelectDirLabel3;The translation has been technically validated against
Default.isl:Tiếng Việt.Since memecoder is listed as the original translator, I would appreciate a native review of the proposed changes before considering the translation ready for promotion from
Unofficial.