Common Fabric · execution dashboard

Pattern computation cost

Loading implementation status…

· Connecting live updates…

Current delivery gate

Validation evidence

    Implementation

    Stages are not equally sized. “In review” means implemented and locally verified, but not landed.

    Observable result · A0–A2

    What does one vote read?

    Explore an actual run: 14 options, eight same-space voters, 74 keyed votes. The vote update changes one green vote to yellow while the poll UI stays demanded.

    ComputationRunsAccessesLink hopsDocument-runsDependency-runs

    Recorded measurements, not a live benchmark. Rows show the original top ten; totals include all completed runs. Reactive action bodies only: event dispatch, commit work, and diagnostic reruns are excluded. Document and dependency counts are sums across runs.

    The existing tally already uses grouping and memoized lookup. These counts demonstrate visibility into work, not an optimization achieved by this PR. Browser measurements appear below.

    Run this fixture
    Recorded command
    
      
    
        
    Demo milestones
    
        
    
      
    
      
    
        
    Observable result · A4
    
        
    A vote with the tally on screen
    
        
    
        
    
      
    
      
    
        
    Remote row updates · C1/C3
    
        
    Four synthetic browser cases pass for nested filters and mapped rows,
          with same-space and cross-space profiles. Assertions cover remote color
          changes, profile-only edits, membership additions, removal/restoration,
          ranking, and first materialization after remote writes.
    
        
    The runtime fix landed in PR #7265 after CI and review. Reconnect verification
          remains. No live poll was accessed.
    
        
    
          First browser render after a remote vote and cross-space profile edit: one red vote and an empty second row
          
    First browser render after remote edits: the linked red vote
            is present, and the test verifies the updated profile name in the row
            attributes. No earlier browser view materialized these values.
    
        
    
        
    
          Mapped rows after cross-space profile and remote vote updates: two green votes rank above one yellow vote
          
    Recorded synthetic cross-space result: two green votes rank
            above one yellow vote. The test also checks profile names in the
            rendered row attributes.
    
        
    
        
    
          After remote vote removal, one yellow vote ranks above an empty second row
          
    After removing both green votes remotely, the second row has
            no swatches and moves below the yellow vote. Restoring a keyed vote
            produces exactly one green swatch.
    
        
    
      
    
      
    
        
    Observable result · A3 working implementation
    
        
    Same functional result, different read budget
    
        
    
        
    
      
    
      
    
        
    Outstanding questions
    
        
    Reply in our conversation using the question ID. Each question states what, if anything, it blocks.