Wikipedia Transformed into a Social Media Feed

A New Experience in Accessing Information

In the tech world, a project that could reshape information consumption habits is drawing attention. Developer Lyra Rebane has created a web application called 'Xikipedia' that presents articles from Wikipedia's Simple English version in a social media feed-like format.

Locally Running Algorithm and Privacy

According to the explanation on Xikipedia's homepage, the application uses a basic algorithm that suggests similar articles based on the content users interact with. The most notable feature of the system is that it does not use machine learning and does not collect any user data. The algorithm runs entirely locally, and all data is deleted when the user closes the tab or refreshes the page.

How Does It Work?

The application contains 'posts', each being a summary of an article from Simple Wikipedia. Users can like posts or choose to see content from specific categories. Liking a post increases the likelihood of content from the same category, parent categories, or linked articles appearing more frequently in the feed. Users can click on a post they are interested in to access the full article via Wikipedia.

The project offers a transparent alternative that increases users' awareness of the impact of AI algorithms on content. However, the application requires a short wait time to load 40 MB of data, and since it pulls text and images from random articles, there is a possibility of encountering content not suitable for adults when scrolling for a long time.

Evaluated as a Discovery Tool

The presence of over 278,000 articles in Simple English Wikipedia provides a vast scrollable content pool. However, it is observed that this version is not updated as frequently as the main Wikipedia. Therefore, Xikipedia is evaluated as a starting point that encourages users to discover new topics.