Welcome to the July 2025 edition of the Language and internationalization newsletter by the Wikimedia Foundation Wikimedia Language and Product Localization team! This newsletter provides you with quarterly updates on new feature developments, improvements in various language-related technical projects and support work, community meetings, and ideas to get involved in contributing to the projects.

Subscribe to the newsletter

Key highlights

Language Support for New and Existing Projects and Languages

translatewiki.net is a platform for gathering translations of interface messages for Wikimedia and other open source projects. Recently, three projects were added: CampWiz NXT ^[1], QuickStatements 3.0 ^[2], and Instant Diffs ^[3] while support for two projects was removed: Wikipedia KaiOS app ^[4] and  MismatchFinder ^[5]. 8 languages were also added, including Rajasthani (India) ^[6], Dolgan (Russia) ^[7], Pazeh (Taiwan) ^[8], and Puno Quechua (Peru) ^[9]. These additions enable contributors to begin translating the interface and are key steps in the language onboarding process.

Adding languages to MediaWiki is the next step after Translatewiki.net. With this support, speakers of these languages can use Wikimedia websites like Wikipedia or Wiktionary in their own language and create or contribute content directly. 4 Languages were added to Mediawiki, including Hunde ^[10] and Wali ^[11]. The Wali Wikimedians community in collaboration with Dagaare Wikimedians also recently organized an Art + Feminism Edit-a-thon in Ghana to empower local editors and address critical content gaps on Wikipedia. Learn more in this blog post.

Last quarter, volunteers made contributions that included fixing redirects for Azerbaijani languages ^[12] and restoring Javanese keyboard options ^[13]. Learn more here.

Language Onboarding and Development Initiative Launched

Since July 2023, the Language and Product Localization team has worked closely with language communities around smaller and newly formed wikis, where timely interventions and engagements to resolve technical issues are highly relevant to keep them moving to achieve their goals. In the fiscal year July 2023 – June 2024, as part of a language inclusion initiative aimed at connecting the dots across disparate language diversity and focused work streams inside the Foundation, the team identified recommendations for improving the social and technical infrastructure to support existing and new languages. More specifically, recommendations were gathered on the language onboarding experiences for new language communities.

These insights informed work in the fiscal year July 2024 – June 2025, which included implementation and testing of access to modern features, such as Content Translation, currently unavailable in the Incubator. In collaboration with the Language Committee, five languages were identified to receive access to these features while monitoring and comparing their editing activity in the Incubator. Learnings from the published report indicated that technical improvements alone are not sufficient, growing the editor base for smaller and new language communities is key, and enhancing the onboarding experience for these communities is paramount.

In fiscal year July 2025 – June 26, efforts will build on the learnings gathered in 2024 – 2025, with experiments focused on engaging native speakers to improve vital content in smaller language wikipedias. Learnings gained over the last two years have guided the initiation of the Language onboarding and Development initiative. Led by the Language and Product Localization team, the initiative aims to plan and activate helpful interventions for languages with a smaller presence on the Wikimedia platform through tooling and support strategies that enable them to move toward their goals—particularly around content building and contribution activity.

Please see the Get involved section for ways to collaborate and contribute to the initiative. Additional comments or questions can be shared via the talk page.

MinT for Wiki Readers Entry Points Enabled on Four Wikis

MinT for Wiki Readers is an initiative that uses Wikimedia’s Machine-in-Translation (MinT) service to allow readers to access content in other languages when it is not available in their language, using automatic machine translation. This feature is exposed only on a few wikis. An analysis was done on the usage of MinT for Readers in Lombard, Venetian, Tswana, and Asturian Wikipedias where this feature is exposed ^[14]. The analysis shows that the Asturian Wikipedia showed the most interaction with the feature, both in terms of overall events logged and views of machine-translated and human-created content. Learn more about this and view detailed analysis here: Analysis of MinT for Wiki Readers on pre-pilot Wikis. This analysis is in preparation for an A/B pilot test in this fiscal year.

Two Analysis Reports on Use of Content Translation Published

The first analysis explores the structure of Wikipedia articles created using the Content Translation (CX) tool and not deleted, focusing on how well these articles align with six common wiki formatting and organization criteria: minimum length (8kB), number of categories, number of sections, presence of images, number of references, and intra-wiki links. As of March 2025, about 20% of all translated articles using CX were well-structured according to these criteria. Of those, 13% were well-structured at the time of creation, while another 7% were improved afterward. About 78% did not meet all the formatting criteria at the time of analysis. Interestingly, around 1% of articles that initially met the criteria later fell below them, possibly due to content removal. Read the full report here. The second analysis explores factors affecting the deletion rate of translated articles.

Overall, user experience in creating quality content appears to be more important in predicting the deletion outcome compared to the common formatting and organization criteria. The number of articles a user has created has the most significant effect on the likelihood of deletion – the higher the number of articles created, the lower the probability of deletion. A notable observation is that experience in article creation is more predictive of deletion outcomes than a user’s total edit count or whether they used the Content Translation (CX) tool. Among the formatting related criteria, page length is the most important variable, followed by the number of wikilinks and references. The number of categories and media files present are the least important in predicting deletion. Read the full report here.

A New Developer’s Journey in Language Technical Contributions

User ToluAyod’s interest in computers began at a very early age mostly exploring hardware, dissembling computers and building them. Later on, pursuing computer science at an undergraduate level and getting involved in software related projects. Late February 2025, user ToluAyod began actively contributing to language technical contributions. Getting started entailed setting up the needed tools including Gerrit, Phabricator and Github. Key contributions entailed supporting translation needs and new languages support. user ToluAyod recalls this experience to have enhanced his knowledge, skills and experiences in programming languages and frameworks including Javascript, Vue and PHP. Overall, an enablement in gaining foundational knowledge in internationalization and localization, hands on experience in open source software engineering concepts and acquiring skills such as problem solving. Listen more here (27:00 to 37:40).  Are you a new developer looking forward to getting started in language technical contributions? Here are some first good tasks to try out.

Malayalam Wikipedia in the Press

A recent Indian Express article highlights Malayalam Wikipedia, which saw an increase in viewership after the election of the new Pope. This motivated Malayalam Wikipedia contributors to expand its knowledge base by adding an article on Pope Leo XIV, where India’s largest Catholic population is. For the creation of this article, Wikimedia’s Content Translation tool was used. At the time of the interview, 12,000 articles were reported as having been created using the tool, out of the 86,000 articles that exist on the Malayalam wiki. Indian Express talked to Wikimedia Foundation staff to learn about the tool, trending topics on the wiki, the challenges faced by smaller versions of Wikipedia like Malayalam, and what the organization is doing to address these challenges. Read the interview with Indian Express here.

New Blog Post Published on Content Translation’s Decade-Long Journey and 2 Million Article Milestone

The post by User:UOzurumba (WMF) highlights the achievements of the Content Translation tool, which since its launch in 2015 has been used to create over 2 million articles to date and has helped bridge the gap for underrepresented languages. Several improvements have been made to the tool in recent years, including the integration of Wikimedia’s own machine translation service, MinT, which combines external machine learning models to extend coverage for missing or low-resourced languages. The number of articles created with the help of this tool across some Indic languages over the past decade has also been quite significant and inspiring, close to 60,000 articles. Read more in the blog post.

Community meetings and events

• Sign up here to attend the upcoming Language Community Meeting in August 2025
• In case you missed the language community meeting in May, you can catch up by watching the video recording and reading the notes. This meeting was co-organized by the Language and Product Localization team and the Language Diversity Hub and had over 20 attendees, covering topics such as:
  • Outcomes from the Language Onboarding Experiment
  • Contributions from the Obolo community in Nigeria
  • A New Developer’s Journey in Language Technical Contributions
• Central Asian WikiCon took place in Tashkent, Uzbekistan, on 19–20 April, featuring sessions on the development of small and medium communities and improving software localization, led by User:Amir Aharoni.

Get involved

• If you are looking for technical tasks, take a look at the easy tasks that haven’t been assigned yet in various language project repositories on Wikimedia Phabricator.
• Interested in contributing to language support related technical tasks for new and existing languages See open tasks on LPL Onboarding and Development workboard.
• If you are looking for tools to edit and translate articles and interface messages, you can use Content translation and Special:Translate on Translatewiki.net. These tools make it easier to work with content in different languages.
• Report feedback on talk pages of language tools.

Stay tuned for the next release! You can subscribe to this newsletter.

References

1.  https://phabricator.wikimedia.org/T393850
2.  https://phabricator.wikimedia.org/T393293
3.  https://phabricator.wikimedia.org/T397574
4.  https://phabricator.wikimedia.org/T395932
5.  https://phabricator.wikimedia.org/T395258
6.  https://phabricator.wikimedia.org/T392255
7.  https://phabricator.wikimedia.org/T395396
8.  https://phabricator.wikimedia.org/T392236
9.  https://phabricator.wikimedia.org/T394068
10.  https://phabricator.wikimedia.org/T379137
11.  https://phabricator.wikimedia.org/T384234
12.  https://phabricator.wikimedia.org/T394873
13.  https://phabricator.wikimedia.org/T393745
14.  https://phabricator.wikimedia.org/T390023

Can you help us translate this article?

In order for this article to reach as many people as possible we would like your help. Can you translate this article to get the message out?

Start translation