Add K8s SIG Docs L10n project idea

[ianychoi]

Adds Kubernetes SIG Docs - Localization Subproject idea for LFX mentorship 2026 term 1.

[ianychoi]

Please note this proposal was discussed during today's SIG Docs meeting. Sharing Meeting minutes from https://docs.google.com/document/d/1emuO4nmaQq3K8JZ9-MQeIygtrCPO9kWv7U7RzTaW4F8/edit?tab=t.0 for visibility.

Recording: https://youtu.be/s0JQjml6Eqk?t=2614 (43:34~)

[Natali/Ian] https://github.com/kubernetes/website/issues/54075 LFX mentorship option
Mentioned in the localization channel, there is a proposal on AI and localization. The Korean localization team have stepped up to participate with Seokho as a mentor
[Ian] Participated in Korean localization mentoring program last year. Good timing to revisit how to enhance workflow with AI assistance

[nate-double-u]

Thanks for this @ianychoi

CC @seokho-son since they were mentioned in your comment.

I'm a little concerned that this looks like a tool exploration project. The delay in tool adoption is not understanding the tools, it's usually the CNCF/LF review of the tools (legal review, PII, partnership setup, etc.). So I'm not sure that this is a good candidate for a mentorship.

@natalisucks, I had to leave the meeting a bit early, so missed this part of the conversation. What are your thoughts on this mentorship proposal?

[ianychoi]

Thank you for the feedback @nate-double-u .

I'd like to clarify that this proposal is not just about tool exploration, but rather enhancing existing SIG Docs localization workflows using infrastructure we already have like Prow/CI.

To give context, we're building directly on established work including the outdated L10n diff tooling [1], Japan team's i18n tracker discussions [2], and Korean kubelingoassist plugin [3]. For example, the Korean team deliberately chose not to follow the main/master branch strategy [4] because it didn't fit our workflow - this project addresses these actual SIG Docs localization pain points.

The AI component specifically focuses on encouraging better localization contributions through workflow improvements like upstream change detection, review prioritization, and consistency hints via Prow comments/labels - not on translation quality itself.

@seokho-son (SIG Docs Localization Subproject co-chair) could you share your thoughts on how this aligns with our team's priorities? Would greatly appreciate your perspective!

[1] kubernetes/website#45844
[2] https://kubernetes.slack.com/archives/C0191RDKHU1/p1750781081578559
[3] https://kubernetes.slack.com/archives/CA1MMR86S/p1765301066771019
[4] https://kubernetes.io/docs/contribute/localization/#branch-strategy

[seokho-son]

Hi @ianychoi @wonyongg @Eundms
Thank you for putting together a mentorship proposal focused on the Kubernetes SIG Docs Localization Subproject.
From the perspective of someone who has been involved in SIG Docs and localization work at a broader level, it is clear that this proposal is grounded in real operational experience and a sincere desire to improve how localization works in practice. That alone makes it a proposal worth supporting. :)

My opinion and suggestion as one of the Kubernetes SIG Docs Localization Subproject Leads on this proposal as follows.

From the point of view of the LFX Mentorship program, the proposal could become stronger with a clearer focus that matches the current strengths of the mentor group. The mentors have solid experience with SIG Docs localization, especially with the Korean localization (@ianychoi is approver/reviewer of Korean Localization, @wonyongg and @Eundms are active contributor of Korean Localization). I guess this makes localization itself a very natural and strong main focus for the mentorship. Framing the project around direct contributions to Kubernetes SIG Docs (localization).

The AI assisted localization automation ideas described in the proposal are interesting and meaningful, especially from a longer term SIG Docs perspective.
At the same time, this area is quite broad and exploratory, and it may be challenging to treat it as a central mentorship goal. Narrowing this part of the project to a clearly defined advanced scope, for example limiting it to one or two specific automation tools or improvements that can be realistically implemented and proposed to the repository, could help make the project more focused and easier to execute. Positioning AI related work as an advanced or optional track rather than the core of the mentorship may also better reflect the current experience profile of the mentor group. I think this might reduce the concerns from @nate-double-u. (is it @nate-double-u ?)

---

Many of the improvement ideas outlined in the proposal are valuable for the SIG Docs Localization Subproject as a whole, regardless of the mentorship context. Some of them may naturally extend beyond what the LFX Mentorship program typically expects in terms of scope and concrete deliverables, and in that sense, it feels appropriate to separate what can be achieved within a mentorship timeframe from what should continue as a broader SIG Docs discussion and effort.

Overall, this proposal shows clear care for the Localization Subproject and raises important challenges that are worth addressing. With a slightly more focused framing that emphasizes localization contributions as the core outcome and treats automation and AI as a constrained advanced area, the proposal has strong potential to become a very solid and successful mentorship project.

Thank you again for the work that went into preparing this proposal, and I look forward to seeing how it continues to evolve.

[seokho-son]

@nate-double-u I have a couple of questions.
I understand that the official deadline for mentorship applications has passed. Would it still be possible to refine the current proposal over the next few days before the final announcement, or would it be better to target the next term instead?

As mentioned earlier, the proposed mentors have strong experience mainly in Korean localization. Given that, would it be acceptable to limit the primary mentee applicants to Korean speakers?

[seokho-son]

Hi @natalisucks,
I have added my thoughts on this proposal. Please feel free to leave your comments as well if you would like.

[nate-double-u]

Thanks for clarifying @ianychoi & @seokho-son!

I understand that the official deadline for mentorship applications has passed. Would it still be possible to refine the current proposal over the next few days before the final announcement, or would it be better to target the next term instead?

The proposal came in on time, so we can target this term. Discussion and clarification can go past the deadline. @ianychoi, if you'd like to make updates to the description based on this conversation, please go ahead.

---

@wonyongg, @Eundms, could you each 👍 or approve this PR to confirm your participation?

Also, since this is a proposal for a Kubernetes SIG project, we'll need a sig-contribex lead's approval as well. Usually we'd require a project maintainer's approval, but things are a touch different with K8s projects, sub-projects, and SIGs.

@kaslin, @mfahlandt, or @palnabarun, if one of you could take a look and give a 👍 (or approve this PR) I'd appreciate it.

[natalisucks]

As we chatted about in the SIG Docs meeting, I see this proposal focusing more on improvements to localization workflows, as opposed to translation on its own, hence the value. I agree with you @seokho-son that choosing a specific tool advancement (e.g. CI, upstream change detection and review priority) could strengthen the proposal, but the framing of this mentorship plan is indeed very positive and I really like the focus on keeping the human element in localization.

For approvers, please note this concept has a +1 from SIG Docs, and we appreciate @ianychoi's work, @seokho-son's feedback, and @nate-double-u's offer to tailor the proposal based on some of the feedback received 🚀

[nate-double-u]

As mentioned earlier, the proposed mentors have strong experience mainly in Korean localization. Given that, would it be acceptable to limit the primary mentee applicants to Korean speakers?

It's a global call for participation, so I ask that the mentors consider all candidates. If this is a Korean-language only project then it isn't a good fit as the primary working language of the program is English.

[wonyongg]

Thank you for the thoughtful feedback and discussion.
I really appreciate the opportunity to be considered and the guidance shared here💙

[wonyongg]

@nate-double-u

It's a global call for participation, so I ask that the mentors consider all candidates. If this is a Korean-language only project then it isn't a good fit as the primary working language of the program is English.

I agree. Since this is a global call for participation, English should remain the primary working language.

[Eundms]

@nate-double-u
I agree as well. This is not a Korean-only project, and English is the primary working language.

[ianychoi]

Thanks everyone for the thoughtful discussion and feedback throughout this process! 🙏

As mentor with @Eundms and @wonyongg, I want to confirm - English will be the primary working language with global participation from any language welcome. The focus remains on sustainable contributor growth through practical workflow improvements that make localization contributions easier and more effective across all languages.

Really appreciate the strong support from SIG Docs (@natalisucks and @seokho-son). This validates that localization sustainability is a shared priority worth tackling together!

[kaslin]

+1 to this project. I know a lot of folks are eager for opportunities to try out real-world AI use cases. Seems like the mentee could learn a lot. lgtm.

[nate-double-u]

Thanks everyone, that's all the approvals we need.

Before I merge this in, @ianychoi, I'd like to give you a chance to make any clarifying updates you feel you need to the description. I'll wait till (my) Monday to merge to give you some time. If you're OK with it as is, let me know and I can merge sooner.

[ianychoi]

Thank you @nate-double-u ! @wonyongg, @Eundms and I worked together to clarify the updates. Thank you!!

Thanks everyone, that's all the approvals we need.

Before I merge this in, @ianychoi, I'd like to give you a chance to make any clarifying updates you feel you need to the description. I'll wait till (my) Monday to merge to give you some time. If you're OK with it as is, let me know and I can merge sooner.

[seokho-son]

Hi @ianychoi, @wonyongg, and @Eundms (@nate-double-u),
This is a bit last minute, but I am happy to step in and support this mentorship as a mentor. As a Localization Subproject co-lead, I feel that I should contribute in any way I can. ;)

[nate-double-u]

I'll merge this and get it loaded into the LFX platform today.

[ianychoi]

That’s wonderful – thank you so much for stepping in and supporting this mentorship as a mentor! 🙏
I really appreciate your willingness to help, and I’m looking forward to collaborating closely together on this mentorship.

Hi @ianychoi, @wonyongg, and @Eundms (@nate-double-u),
This is a bit last minute, but I am happy to step in and support this mentorship as a mentor. As a Localization Subproject co-lead, I feel that I should contribute in any way I can. ;)