Discernatron is a search relevance tool developed by the Discovery department. Its goal is to help improve search relevance - showing articles that are most relevant to search queries - with human assistance.
We are asking for people to use Discernatron to review search suggestions across multiple search tools. Respondents will be presented a set of search results from four tools - CirrusSearch (Wiki search), Bing, Google, and DuckDuckGo. For each query users will be provided the set of titles found asked to rank each title from 1 to 4, or leave the title unranked.
You can access Discernatron at https://discernatron.wmflabs.org/ and authenticate with your unified user account.
When evaluating potential changes to search those changes will be rated by how much closer results can get to putting the most relevant articles, as judged by humans in Discernatron, at the top of the search results page.
The plan is to release a new set of 500 queries into Discernatron each month to be graded, and ideally each query will be graded by 10+ people.