Notes from brainstorming on possible implementation scenarios with @UOzurumba and others:

• Scenario 1: Community has a list of vital articles (e.g., Telugu Wikipedia)
  • If the community already maintains a list of vital articles:
    • The list can be integrated into CX/SX.
    • These articles can then be surfaced directly through CX/SX. Through tailored onboarding, editors work to:
      • Improve articles section by section, or
      • Create missing articles from scratch.
    • If integration isn’t technically possible:
      • Editors can search for the source articles for existing pages manually in CX/SX and translate them into the destination language.
      • Caveat: This only works if the source article exists in English (or another supported source language). This might require a workaround. See T409222

• Scenario 2: Community doesn’t have a list of vital articles (e.g., Moroccan Arabic Wikipedia)
  • Approach 1: Editors work on topics that align with either their community’s needs or their own individual interests (e.g., gender gap, local literature).
    • If a category only includes existing articles:
      • There’s currently no automated way to surface sections needing improvement. This might require a workaround. See T409222.
    • If a category includes new articles:
      • Editors could look for article ideas that exist on larger wikis (e.g., English Wikipedia) but are missing locally, and use those as sources for translation.
      • Caveat: This only works if the source article exists in English (or another supported source language).
  • Approach 2: Editors work on improving articles in “1000 articles” list already available in CX/SX collection.
    • Caveat: These lists may not reflect what’s most vital or relevant for the local community.

Progress update on the hypothesis for Week 1:

Lots of discussions and planning took place last week related to the hypothesis with @SBisson @ToluAyo @UOzurumba @cchen and the CN admins.

These covered providing communication support, exploring implementation scenarios, defining technical requirements (such as setting up banner tracking and generating one-click links to CX/SX), outlining a measurement plan, obtaining approval from CN admins for the experiment, and designing and documenting tailored onboarding guidance. See more details in the subtasks: T405298

A tailored onboarding material is also a work in progress. It’s still in the early stages and needs feedback and review, but it gives an idea of how we plan to introduce weekly onboarding tips, starting with Week 1, focused on understanding the basics, learning about Wikipedia’s pillars, making simple edits using Visual Editor, and participating in a discussion prompt on the talk page related to Wikipedia’s pillars.

In the following weeks, the focus will shift toward guidance on section and content translation tools and engaging editors in using them to improve vital articles.

New lessons from the hypothesis:
Currently, there is no easy way to select articles outside of CX/SX and start translating them in CX/SX with a single click. The current navigation requires multiple steps in the UX, and if an article doesn’t have a source, the editor ends up seeing empty results. For this work, we plan to programmatically generate SX URLs for a list of vital articles so they can be embedded on a wiki page and linked with the onboarding resource. T409222

Progress update on the hypothesis for Week 2:

First version of the complete onboarding material has been shared with the Community Development team for early feedback here (doc version). It provides step-by-step guidance for Weeks 1–4 on improving vital articles using contribution tools (Visual Editor, Content Translation tools).

First iteration of the functional script to generate Section Translation links has been developed (link). Thanks to @ToluAyo. This script takes an array of article titles in a target language and generates a direct URL to the SectionTranslation page of each article to be localized or improved in that language.

An additional dependency emerged last week: gathering feedback on our initial onboarding material from WMF's Community Programs and Community Development teams before engaging communities for translation support - to understand, based on their experience, whether this approach is appropriate and could achieve the intended impact. Once we receive this feedback, we will reevaluate our timeline.

Another consideration is our current plan to target only contributors from small wikis. We are assessing whether this might make the pool too narrow and whether high-traffic wikis should also be considered. However, given the December timeline and overlap with fundraising plans, involving them may not be feasible.

Progress update on the hypothesis for Week 3:

@MGuadalupeWMF and @AEspinoza (WMF) from the Community Development team provided internal feedback on the onboarding material, based on which follow-up changes were made. A gist of the feedback:

From a learning perspective, the curriculum is solid and offers a clear, well-scaffolded pathway. The weekly sequence builds essential skills in a manageable way for learners unfamiliar with content translation tools, the prompts encourage reflection, and the optional resources offer opportunities for deeper engagement. We recommend adding a brief “Start Here” and “Wrap-Up” highlight to each week, as well as a final prompt that encourages learners to showcase the articles they’ve been working on and reflect on accomplishments or next steps.

The team also thinks the curriculum is a strong fit for WikiLearn.

@UOzurumba sent messages to two communities on Friday. Asks shared:

• Event registration setup
• Translating content
• Identifying vital articles
• Emailing participants from previous experiment

Banner design is ready here: T409227

The experiment is planned to run from November 30th to December 14th. If we receive responses and coordination from communities by November 28th, we will proceed as planned. Otherwise, we will revisit and may consider moving community engagement to the next quarter.

We plan to continue this work next quarter under a new hypothesis, focusing on delivery and piloting. We also plan to include one additional wiki and conduct CN outreach through larger wikis, which was not possible in December due to the quiet period, pending requirements, and overlapping fundraising plans.