The current load indicator (the small spinner in the upper left of reports) isn't really sufficient for users. On our larger reports, which can take a while to load, the user sees a large empty area ...
fbcideas_migusr
6 years agoNew Member
This is a great idea. Even the spinner wasn't enough for some end-users to know when the filter was applied and I would have to train them on that. Greying out the visual until the data is ready is so intuitive that it would require no training for the end users to understand.
Recent ideas
Select Entire Hierarchies in Power BI Personalize Visuals
Feature Request: Select Entire Hierarchies in Power BI Personalize Visuals Hi Power BI Team, I'd like to suggest an enhancement to Personalize Visuals that allows users to select an entire hierarch...davejames121 hour agoNew MemberNew3Views1like0CommentsFile level commit - Need the ability to edit the commit description
I tried the recently announce file level commit for Fabric items. One notable issue I have with the feature (testing this with a Fabric warehouse, I'm assuming this would hold true for other items): ...DataBard1 hour agoMost Valuable ProfessionalNew30Views0likes2CommentsRename notebook cell name “SparkR (R)” to “R” and “PySpark (Python)” to “Python”
It's a simple idea. In Fabric notebooks, the cell language for R is labeled “SparkR (R)”. Since Spark 4.0, the SparkR package is deprecated, and the Fabric Runtime 2.0 documentation itself notes th...lorin_s3 hours agoNew MemberNew11Views0likes0CommentsAuto Display Units Should Dynamically Change Granularity
When auto is selected for display units, the granularity should automatically change to best suite the size of the data. For example, we often have charts that have numerical ranges from the milli...acason3 hours agoNew MemberNeeds Votes351Views26likes5Comments