When you go online, it feels like you're accessing all the world's information. But the reality is that most of the internet remains out of sight, hidden behind a language filter. You form social media relationships based on shared language, search Google with the language you think in, and algorithms built to maximize attention have no reason to recommend content you won't understand. As a result, the internet you see is limited to your linguistic silo, and you're missing out on a vast array of content and cultural perspectives.

The concentration of internet activity on a small number of large platforms can create the illusion that everyone uses them in similar ways. However, research by our team at the University of Massachusetts Amherst's Initiative for Digital Public Infrastructure has uncovered stark differences in how different cultures harness the internet. We've found that language and culture shape online participation in profound ways, leading to parallel internets that are shaped by local needs, expectations, and norms. For instance, the Russian social media platform LiveJournal was known to English-speaking users as a space for young people to share their feelings, but to Russian speakers, it was an important site of public intellectualism and political discourse.

The biggest technology companies are based in the US, which has created a cultural blind spot where we often assume that the English internet is representative of the rest of the world. This bias is particularly evident in research about YouTube, which tends to focus on English-language videos and neglects the diversity of content and usage patterns on the platform. To study the inner workings of YouTube, our team developed a method to randomly guess video URLs, allowing us to gather a large representative sample of videos and paint a more accurate picture of what's happening on the platform. We examined language-specific samples of English, Hindi, Russian, and Spanish YouTube, working with native speakers to validate our language detection tools.

Our research revealed radical differences in cultural norms and usage patterns on YouTube. For example, Hindi YouTube is characterized by extremely short videos, with a median duration of just 29 seconds, compared to two-and-a-half minutes for Spanish videos. This is likely due to the influence of TikTok, which was incredibly popular in India before being banned in 2020. YouTube's Shorts feature, which was introduced to fill the void left by TikTok, has become a major ecosystem in India, with 58% of Hindi YouTube content consisting of Shorts. We also found that Hindi YouTube has a unique category distribution, with Entertainment and Education dominating, and a popularity metric that suggests a more intimate and human form of engagement, with less popular videos being appreciated and acknowledged by a private audience.

The implications of our research are significant, suggesting that we need to rethink our assumptions about the internet and its users. Language doesn't just shape your view of digital life – it can obscure the diverse, culturally specific ways people use these platforms. We're building businesses, journalism, and regulation on an artificially limited view of the internet, one often filtered through English, popularity, and convenience. It's time we looked deeper and recognized the parallel internets that exist beyond our linguistic silos. By doing so, we can gain a more nuanced understanding of the internet and its role in shaping our global culture.

As we continue to explore the complexities of the internet, it's clear that we have a lot of work to do to uncover the diverse ways people use these platforms. Our research is just the beginning, and we hope that it will inspire a new wave of study and exploration into the hidden corners of the internet. By looking beyond the surface level of popular platforms like YouTube, we can gain a deeper understanding of the internet's role in shaping our global culture and the ways in which language and culture intersect with technology. Ultimately, this knowledge can help us build a more inclusive and nuanced understanding of the internet, one that recognizes the diversity of human experience and the many different ways that people engage with digital technology.