"Write in English, Nobody Understands Your Language Here": A Study of Non-English Trends in Open-Source Repositories
This study analyzes billions of GitHub interactions to reveal that while non-English participation in open-source software is steadily increasing across code and documentation, this growing linguistic diversity simultaneously fosters inclusivity while creating visibility barriers and collaborative tensions rooted in the community's historical dominance of English.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine the world of open-source software (like the code that runs your phone, your browser, or your favorite apps) as a giant, global potluck dinner.
For decades, this dinner had one strict rule: "Everyone must speak English at the table."
If you wanted to bring a dish (code), write a recipe (documentation), or chat with the other guests (discussions), you had to do it in English. This made sense because the first people to start the dinner were mostly from the US and UK, and the early tools were built in English. It was the "lingua franca" (common language) of the kitchen.
But recently, the dinner has gotten crowded. People from Korea, China, Russia, Brazil, and India are joining in huge numbers. And just like in real life, they want to talk to each other in their own languages.
This paper, titled "Write in English, Nobody Understands Your Language Here," is a massive investigation into what happens when the potluck becomes truly multilingual. The researchers looked at 9.14 billion messages and 62,500 code projects over ten years to see how the language of the kitchen is changing.
Here is what they found, explained simply:
1. The Table is Getting Louder (and More Colorful)
The Trend: The "English-only" rule is breaking down.
The Analogy: Imagine the dinner table used to be 96% English speakers. Now, about 11% of the chatter is in other languages, and that number is growing fast.
The Details:
- Korean is the fastest-growing language at the table (up 1,700%!).
- Chinese, Russian, Japanese, and Vietnamese are also filling up the room.
- This isn't just in the "chat" (issues and comments); people are even writing their recipes (code comments) and labels on the jars (string literals) in their native tongues.
2. The "Secret Sauce" vs. The "Main Dish"
The Trend: People are mixing languages, but they are keeping the core structure in English.
The Analogy: Think of a computer program like a house.
- The Foundation and Beams (Identifiers/Variables): These are still almost 100% English. You won't see a beam labeled "Muro" (Wall) in a Spanish house; it's still called "Wall" in the blueprint. This is because computer tools (compilers) are picky, and it's hard to type non-English characters without switching keyboards.
- The Decor and Notes (Comments & Documentation): This is where the magic happens. Developers are writing notes to each other in Korean, Chinese, or Russian. They are saying, "This function does X," but in their own language. It's like writing a sticky note on the fridge in your native tongue.
3. The "Language Police" and the Friction
The Trend: Even though everyone is welcome, there is still tension.
The Analogy: Imagine a guest arrives and starts speaking Portuguese. Another guest might say, "Hey, please speak English so we can all understand!"
- The researchers found thousands of these moments. Sometimes it's a polite reminder; sometimes it's a harsh rejection.
- Interestingly, it's not just English speakers doing this. In a Chinese project, someone might say, "Please write in Chinese!" to a Russian speaker.
- The Result: This creates a "gatekeeping" effect. If you don't speak the dominant language of that specific project, you might feel unwelcome or ignored.
4. The Popularity Contest (The "Star" System)
The Trend: Speaking English gets you more fame.
The Analogy: Think of GitHub "Stars" (likes) as the applause at the dinner.
- English Projects: They get the most applause, the most new guests (contributors), and the most attention. They are the "famous chefs."
- Non-English Projects: They often get fewer stars and fewer new helpers, even if their code is excellent. It's like a delicious dish served in a language only a few people understand; fewer people know it exists, so fewer people try it.
- The Catch: However, there is a twist! Non-English projects actually fix problems faster. Because the teams are smaller and more focused, they reply to complaints quicker. English projects are so huge and chaotic that they take longer to respond.
5. The Future: AI as the Universal Translator
The Trend: Artificial Intelligence might be the hero that saves the day.
The Analogy: Imagine if every guest at the dinner had a magical earpiece that instantly translated what anyone said into their own language.
- Tools like GitHub Copilot and ChatGPT are already helping developers write code and comments in their native languages.
- The hope is that in the future, a developer in Brazil can write a bug report in Portuguese, and a maintainer in Japan can read it in Japanese, with the computer doing the translation seamlessly.
The Big Takeaway
The open-source world is becoming more inclusive (more languages, more voices), which is a beautiful thing. But it's also becoming more complicated.
- The Good: More people can participate without needing to be fluent in English.
- The Bad: Projects that don't use English get less visibility and fewer helpers.
- The Challenge: We need better tools (like automatic translation) and better community rules to make sure that speaking a different language doesn't mean you get left out of the conversation.
In short: The potluck is no longer just an English-speaking affair. It's a global festival. But to make it work, we need to figure out how to keep the table connected while everyone speaks their own beautiful language.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.