Imagine standing in a dark room with a single candle. When you touch the flame of one candle to another unlit candle, the darkness recedes further, while your original flame continues to burn brightly. This act symbolizes an ancient Italian proverb: “One candle loses nothing by lighting another.” This metaphor illustrates one of Wikipedia’s transformative innovations—the Content Translation tool. Just as candles spread light, this tool has helped “ignite” more than 2.4 million translated articles across various languages within Wikipedia.
Before the launch of the Content Translation tool, a significant knowledge gap existed in underrepresented language Wikipedias. While editions in English and other major languages thrived with millions of well-sourced articles, less represented language Wikipedias faced a daunting reality with a significantly low number of well-sourced content. Also insights from Scott Hales’s research on the influences and editing patterns of multilingual editors on Wikipedia revealed that multilingual contributors who are native speakers of underrepresented languages often edit similar types of content on both Wikipedias. For example, languages in West Africa, such as Hausa and Igbo, had little content in their Wikipedias, and the multilingual contributors worked to expand knowledge in those languages by translating content from English and other bigger language Wikipedias.
As a result, being monolingual in these native languages severely restricted access to humanity’s collective wisdom. With easy and better access to translation tools, more multilingual editors can contribute towards improving both the knowledge base as well as the contributor base in their native language.

Bilingual contributors faced challenges in translating Wikipedia articles prior to the content translation tool. They had to copy content to a separate document, translate it while considering context, and then format and cite it using Wikitext—all while following editing policies. This cumbersome process hindered volunteer participation, especially in underrepresented language Wikipedias where digital literacy is often limited. Writing articles from scratch in those languages was also time-consuming.
The Introduction of the Content Translation tool
The Content Translation tool, launched by the Wikimedia Foundation in 2015, automated the tedious translation process, making it easier for Wikipedia’s wealth of knowledge to flow across languages with a human-review process, ensuring context preservation which defines Wikipedia as highlighted by Catalan speakers in a roundtable discussion. Over the years, as the Wikimedia Language and Product Localization team (formerly the Language team) has iteratively refined and enhanced the tool, we have witnessed remarkable growth in content creation across underrepresented languages, contributing to knowledge equity. Importantly, the knowledge gap in Wikipedia isn’t just about the number of articles but also the diversity of topics covered. Initially, Wikipedia editions in Southern Asia and West Africa were characterized by a focus on geographical entries and biographies, leaving vital subjects in STEM, culture, and history underrepresented. The tool is gradually helping to bridge these topic gaps in the sub-regions.
The spread of knowledge across continents
Through collaborative initiatives with Wikipedia volunteer communities and campaigns championing knowledge equity, contributors have made significant strides using this tool to add content and increase the number and coverage of articles in various Wikipedias. Examples would be:
- With the distinctive contributions made in Igbo, and Hausa language Wikipedias, the tool has created 80 percent of all content, illuminating key topics around STEM, culture, history and society, that were previously unavailable to these languages. Volunteer editor Aliyu from Nigeria shared in an article that before the tool, he couldn’t relate to what it would be like to translate health articles manually from English to Hausa Wikipedia.


- The Indic language Wikipedias (such as Telugu, Tamil, Nepali, Bengali, Kannada, Marathi, Urdu. Odia, Assamese and Kashmiri language Wikipedias) have seen consistent yearly growth in translated articles. These language Wikipedias, which are widely spoken in Southern Asia, have expanded from translating fewer than 11,000 STEM topic articles to over 59,641 in just three years.

- During the Covid-19 pandemic, the tool helped the translation taskforce to make COVID-19 articles in the WikiProject COVID-19 initiative available in more than 130 languages in a short time. This helped reduce the risk of misinformation about COVID-19 and provide reliable information to the world during and after the lockdown.
Notably, articles generated through the translation tool have been measured to be less likely to be deleted than those created from scratch. Indicating that when carefully edited and reviewed by humans, these machine-assisted outputs yield high-quality content in many language Wikipedias.
A decade of consistent improvements
Like a gardener nurturing seedlings, the Wikimedia Language and Product Localization team has nurtured the Content Translation tool over the past decade. As the Chinese proverb goes, “A journey of a thousand miles begins with a single step.” The launch of the Minimum Viable Product marked that first step—planting seeds in Wikipedia’s vast garden. Each enhancement has evolved the tool, leading to notable improvements:
- Google Machine Translation Integration (2016) that sped up the translation process across more language Wikipedias by incorporating automatic first drafts from Google Translate. You can read more about the impact from a researcher’s findings.
- Improved Template and Citation Handling (2018) that easily converted complex templates in scientific articles during translations, preserving references and content integrity.
- Collaborative translation mobile experience (2021) which allowed multiple editors to work on different sections of the same article simultaneously, improving collaboration.
By 2021, the tool had evolved to tackle both sociological and technological challenges, becoming a robust tool for desktop and mobile users to collaboratively translate articles section by section while detecting misuse of machine translation.
Further technical advancements continued to expand the tool’s reach:
- MinT Machine Translation: Expanding translation support to over 40,000 language pairs. It increased the productivity level of translators in 55 Wikipedia language editions, which got MinT translation support for the first time. More details are available in this article.
- Personalized Translation Suggestion feature: This is a recent improvement from the team. This work has improved the Content translation tool’s article suggestion list support, enabling users to customize and receive article recommendations based on 41 topic-based filters. This means they can discover and translate articles in their area of interest and expertise, as well as discover community curated lists called Collections that are inspired by content growth campaign initiatives. Currently, this feature has about 50 collections and counting, with further improvement in the pipeline.
- A unified dashboard made it easier to access the same capabilities on both desktop and mobile devices. With this, desktop users were given the ability to translate articles section by section, and personalised filters to get translation suggestions that align with their area of interest.

Your role in knowledge equity
This tool’s evolution reflects a commitment to continuous improvement, expanding access to high-quality, human-reviewed information across various fields, including STEM, Culture, and History. Each translated article creates new connections and encourages more people to participate in this global effort. Here’s how you can get involved:
For Multilingual Editors: Use the Content Translation tool to translate an article into languages you know. Even one article can help someone find information in their native language.
For content growth campaign organisers: Host translation events focused on topics important for underrepresented languages and use the page collection tag to attract more translators to translate your article list.
The Content Translation tool shows Wikipedias’ commitment to making knowledge available to everyone. Together, we can help by sharing knowledge and lighting pathways for others to broaden the curiosity and understanding for all.

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