Technology

  1. Bashkir becomes the first language collated inside MediaWiki

    Are the names of articles in the wiki in your language shown in the correct alphabetical order? If not, perhaps you can now fix this similarly to how it was done for the Bashkir language.... Read more

  2. Admittedly loopy but not entirely absurd—Understanding our Search Relevance Survey

    We’re experimenting with machine learning as part of our work to improve search on Wikipedia and the Wikimedia projects. This can sometimes lead to amusing or even strange queries for users—but they’re not an error.... Read more

  3. Introducing the Cloud Services Team: What we do, and how we can help you

    24% of Wikipedia edits over a three month period in 2016 were completed by software hosted in Cloud Services projects. In the same time period, 3.8 billion Action API requests were made from Cloud Services. We are the newly formed Cloud Services team at the Foundation, which maintains a stable and efficient public cloud hosting platform for technical projects relevant to the Wikimedia movement.... Read more

  4. What ten students made for Wikimedia while participating in Google Summer of Code and Outreachy round 14

    Ten students from India, Israel and United States contributed thousands of lines of code to Wikimedia projects this past summer under the mentorship and guidance of thirteen mentors.... Read more

  5. The great Wikipedia bot-pocalypse: Challenging an established narrative

    Headlines assumed that when Wikipedia bots reverted edits made by other bots, the two were in conflict or ‘fighting’. This assumption sounds reasonable, but it is not the case for an overwhelming majority of bot-bot reverts.... Read more

  6. Wikipedia, search, and the “Цкщтп” keyboard

    People sometimes search for things on Wikipedia that look like gibberish. Unlocking that mystery was up to Trey Jones, a senior software engineer who spends a lot of time uncovering the secrets of search on Wikipedia. ... Read more

  7. Pulling ‘Puppet’ strings on Discovery’s Dashboard framework

    The Wikimedia Foundation’s Discovery Analysis team recently switched to a Puppet-based configuration for their metrics dashboards. In this post, we share resources and tips for learning Puppet for non-Technical Operations (Ops) people, and—as an educational exercise for newcomers—explain how the new configuration works.... Read more

  8. Honoring our friend Bassel: Announcing the Bassel Khartabil Free Culture Fellowship

    To honor Bassel Khartabil’s legacy and his lasting impact on the open web, a slate of nonprofits are launching a new fellowship in his name.... Read more

  9. Investing in our shared future, supported by AI: Announcing the Scoring Platform team

    The Wikimedia Foundation’s new Scoring Platform team, led by Aaron Halfaker, will be working on democratizing access to AI, developing new types of AI predictions, and pushing the state of the art with regards to ethical practice of AI development.... Read more

  10. Investigating a mysterious performance improvement

    Late last month, large wiki articles were appearing to load faster in our testing environment. We wanted to find out why.... Read more