Structured Data Across Wikimedia/Search Improvements

Languages:

The Search Improvements project will use structured content to give users a more inviting and more efficient way to search and find content on the Wikipedias. By improving Special:Search, we want to enable users to find the information they are looking for, or that they may not have noticed, or previously come across through existing search.

We aim to identify and define incremental “special search” improvements that use structured content, to assist users in finding the content they are looking for, especially in those language wikis that have fewer articles.

Background[edit]

Existing search functionalities mainly focus on article matches. This is a problem for Wikipedias in emerging languages, because there is a higher risk for readers to end up on the Special:Search page without finding relevant articles or content. Linguistic factors and inexperience in using Special:Search are also among the main obstacles in finding content.

Moreover, it is likely that the information that the user is seeking might be “hidden” somewhere on the projects – i.e., under sections of existing articles, or inside articles with a different name, or in sister projects and other sources in our ecosystem.

The project[edit]

What we want to do[edit]

Work in this area will include things like:

Stylistic UI improvements to the Special:Search page
Creating a quick view panel for each result that:
- shows key information about the result
- displays interwiki links related to the result

Search Preview panel[edit]

Tracked in Phabricator
Task T306341

One of the proposed changes that we want to introduce is a Search Preview panel for each result, to discover relevant content and go directly to a desired section of the article. The following mockups illustrate how this panel would work.

The changes will be initially rolled in early 2023 on three partner Wikipedias - Portuguese, Russian and Indonesian Wikipedia - for a first evaluation period.

Desktop

Click on the snippet to open a search preview on the desktop. Users can still click on the article name to go straight to the article.
Preview opened for the first search result. The preview is opened by clicking on the snippet and stays open until the user taps on the close button or clicks on the snippet again.
All the content that could appear in quick view: 1) Article image 2) Expanded snippet 3) Top few sections of the article 4) “Related images” - media in this widget would come from MediaSearch on Commons. 5) Sister projects associated with the article.

Mobile

Tap on the arrow to open a search preview on mobile.
The preview is opened by tapping on the arrow and stays open until the user taps on the arrow again to close it.
Users can swipe left...
...and right to go through the contents of the preview.

Improving search experience through a tool/gadget[edit]

Additionally, we are working with interested contractors to ideate and develop a tool or gadget that experiments with using structured data/content to improve the search experience on one or more Wikipedias.

The first tool that has been ideated is “View it!”, and is developed by Dominic Byrd-McDevitt (User:Dominic), Kevin Payravi (User:SuperHamster), and Jamie Flood (User:JamieF). The tool will enrich Wikipedia content by offering an illustration of a given subject. This tool will increase the discovery of Wikimedia Commons uploads and encourage contributors to utilize Commons and structured data. While the number of images displayed in a Wikipedia article is finite and highly curated by editors, this tool will allow readers to access the full catalog of images available on Wikimedia Commons, and help editors easily add relevant items to a given article.

Feedback[edit]

Project feedback is and will always be welcome. We are especially interested in your ideas, and we are looking forward to hearing from you on the talk page about what you think about the project, and especially about the following open questions:

What do you think about the approaches outlined above?
Do any of these changes affect the contribution flow in your opinion? If yes, how?
Are the images, sections, and other content in search preview helpful to you? If so, how would you use it?
Is there anything else you would like to see added or removed from the search preview or desktop or mobile?
What do you think about the interaction of viewing the search preview on mobile?

Previous work[edit]

Round 1 mockups[edit]

The following are the mockups of our first round of suggested improvements to Special:Search, which have been discussed in a multi-project request for comment with the Wikimedia community from June 27 to July 10. After the consultation, some of the improvements have been approved and are being currently implemented, while others have been put on hold, pending further investigation.

1. Add article thumbnail to search results to allow for better scanning of the content, using the same thumbnails that appear in the Go bar (implementing, T306883)
2. Add namespace tags in front of the article to be able to better distinguish between various page types (on hold, T307605)
3. Rearrange the sister search section and add a header for better context and awareness of these projects (implementing, T307468)
4a. Collapse advance search so that there is more focus on the results itself
4b. The advanced search can then be expanded if the user needs to use it (on hold, T307041)
5. Introduce a quick-view panel for each result to discover relevant content and go directly to a desired section of the article (implementing, T306341)
6a. Re-style metadata to put more focus on the content itself (implementing, T307463)
7. New location for “create missing article” message to allow more focus on the search results and at the same time provide a clearer and consistent call to action to add an article (on hold, T307467)
8. Change the color of the “did you know” suggestion from red to blue as red is more appropriate for missing articles (Message A: “did you mean” message)
8b. (Message B: “Show results for” message) (implementing, T307464)