Topic on Talk:Structured Data Across Wikimedia/Search Improvements

• Article thumbnail: OK, but I do have some NSFW reservations, which might be solvable via T306246.
• Namespace tags: Not sure about this one. First of all. What is the tag for main ? Second.. it just feels cluttered... I see the challenge, but I'm not sure this is the right solution.
• Sister search: Sure
• Collapse Advanced search: Hmmm. perhaps this requires some more refinement. "Advanced search" seems rather archaic. Maybe we should rename? And I like for instance the Google method with their "Tools" button which drops down several tools into view. As a matter of fact, I don't think you can even find their "Advanced search" any longer. Lastly. Advanced Search was made by WMDE and I know they put in quite a bit of user research over the course of a year. That should be carefully reviewed before tinkering with this.
• Quick View: Like it. Though I have a bit of concern wrt to how this would function. Would it only show the first result ? Or the result you selected ? does that make picking a search result a two step process ? (click a result to open quickview, click something in quick view to navigate to actual result) ? That would seem less than ideal to me.
• Missing article: Not against it, but wouldn't that conflict with quick view then ?
• Did you mean: Sure

Hello @TheDJ, thanks a lot for your feedback!

We already know about T306246 and someone is already looking at that. Hopefully there will be good news in the future, though I cannot promise anything.

We are also already tweaking the feature in order to make it easier for users to hide the thumbnails, if needed. Also, we decided to take our time to re-evaluate our proposal on namespace tags, and not to proceed with collapsing advanced search.

Thank you also for your reply on QuickView. To answer your question, you would see a quick preview of the article you select, by clicking on the snippet, but you will be also able to go straight to the article as you would usually do. No two-step process (that would be indeed not ideal).