Setting to critical because this is causing random fails.
Waiting for requests to complete in e2e tests is occasionally necessary when there are no reliable implementation details to wait on instead, but should only be used when absolutely necessary.
These uses can't be used as "it cant hurt to err on the side of caution and wait a little more" tools because the tests will fail if the request isn't made.
Even the wait() uses that seem more technically-sound should instead look for DOM changes etc whenever possible as endpoint requests may happen less often as our caching improves and the UI becomes more complex.
Replace andy wait() uses that can be replaced with implementation details such waiting for elements to change, etc. Any uses that can't be replaced should be accompanied by a comment explaining why (we do something similar for when !important is used in our themes and it really helps!)
Active
Page builder
Enhances developer experience.
(Not a blocker anymore thanks to @bnjmnm's analysis on #3455975.)
This reminds me conceptually of 📌 Stabilize FunctionalJavascript testing AJAX: add ::assertExpectedAjaxRequest() Fixed — I wonder if we can do something similar here? Any inappropriate "wait" would then explicitly fail.