Animation Example Descriptions

These examples describe the visual sequence of what happens on screen and intentionally leave out how the DebateCore is updated or how that DebateCore update interacts with the visuals. The planner can still depend on that interaction; this document is only describing the visible sequence.

Current Implementation Scope

• The first production planner implementation currently covers add-confidence claim animation only.
• The first production planner sequence currently stops after the first wave from the command target through its outgoing connectors.
• Add-relevance animation and subsequent propagation-wave sequencing are future scope.
• The settled display state still needs to support debates that already contain relevance structures elsewhere in the graph.
• The planner input boundary is the pre-command DebateCore state plus the command payload.
• Planner-facing math should stay DebateCore-shaped, so the planner should consume DebateCore-first math entrypoints rather than building a separate scoring adapter layer.

Scale Rule

• sourcesScale is the size factor applied uniformly to every structural element in a local area: claim boxes, confidence connectors, and all layout distances.
• Scale is self-similar — zooming into a scaled-down area produces geometry identical to the same area at full size.
• A claim's score controls fluid fill and contributes to its sibling group's shared base scale. It never makes that claim alone structurally smaller than a sibling.

Group Scale Rule

Terms:

• relevanceMultiplier: the combined effect of all relevance connectors on one confidence connection. Equals 1 when no relevance connectors are attached.
• deliveryScore: the authoritative contribution weight delivered to the target.

Rules:

• All source claims connecting to the same target share one sibling base scale: targetSourcesScale / max(1, sum(deliveryScores)).
• Every claim in the group and its confidence connector use this base. Scores indirectly shrink the whole group only when total delivered contribution exceeds capacity.
• Each source's delivery connector scale = sourcesScale × relevanceMultiplier, making it wider or narrower than the confidence connector when relevance is not 1.
• Parent fluid share is deliveryScore / max(1, sum(deliveryScores)). Zero shares retain their ordered point and visible empty shell.
• Examples without relevance: 100/0/0/0 keeps parent-sized siblings; 100/100/0/0 makes every sibling base 50%; 20/30 keeps parent-sized siblings and leaves half the target empty; 60/80 gives base 1/1.4; all-zero groups retain parent-sized empty shells.
• With score 100%, relevance 2, and a zero-score sibling at relevance 1, both sibling bases are 50%; the first delivery shell becomes parent-sized after relevance and owns the full parent fluid interval.

Layout Rule

• claimLaneAxisGap is the edge-to-edge gap between sibling claim boxes along the lane axis, not a center-to-center distance.
• That gap resolves at the same local sourcesScale as the surrounding geometry, so it shrinks and grows with the source-side potential scale.
• In the current orientation, claim boxes are left-justified within the claim-lane band.
• Sibling source claims form local clusters that stay mostly centered on their source claim when surrounding constraints permit it.

Add Confidence Claim to Existing Debate Example

• Voila: The new claim scales in from zero to its calculated size at its settled position and remains in the foreground throughout the animation sequence.
  • Adds in the new claim setting the scale to tween from zero to its planned full pipe scale.
  • That planned claim scale is the shared child sourcesScale solved for that target's sibling group from current scored delivery demand. The outgoing delivery side uses that same base scale and then applies the child's continuous relevance multiplier after the junction.
  • Add in the connectors, junctions and agregators for the new claim.
    • visible is false for the ones that support that.
    • Delivery Connector scale and score is set to zero
• Sprout: All affected claims and their connectors transition together throughout the step. The claims move once into their planned compact positions and scales while each connector remains attached to its source and target, adjusts to its planned width and target stack position, and the new Delivery Connector traces from the new claim toward the target claim in the sourceToTarget direction. The compact claim order uses the same shared ordering rule as the target-side connector stack so the lines do not cross.
  • See Debate Animation Data Model Design for how those target-side attachment positions are determined.
• First Fill: The score fluid progressively fills the new pipe. Its moving frontier uses the path primitive's overflow-preserving curved extremity collapsed toward the bottom edge. The tip enters through the source boundary before the full slant is visible, and the full slant flows beyond the target boundary before the fill becomes flush.
• Wave: After First Fill completes, interpolate logical scores through the settled layout. The separate moving frontier carries old fluid ahead and new fluid behind while traveling right to left. Subsequent propagation toward the Main Claim remains future scope.

Add Relevance Claim To A New Junction

• Voila: The new claim scales in from zero to its calculated size at its settled position.
• Sprout: All affected claims and connectors transition together throughout the step while remaining attached to their connected geometry.
  • The pipe wall and pipe interior progressively trace out the path of the Relevance Connector from the new claim to the top or bottom side of the relevance aggregator attached to the affected junction on the affected confidence connection.
  • The affected confidence connection shows a visible junction and a visible Display Confidence Connector leading into that junction.
  • The Relevance Connector uses the top side if the relevance claim is above the junction and the bottom side if the relevance claim is below it. It reaches that side with the same slope as that side.
  • If multiple relevance connectors land on that same relevance-aggregator edge, they restack according to the shared Debate Animation Data Model Design rules.
  • If there is only one relevance claim, the relevance aggregator may remain hidden or collapsed even though the Relevance Connector is still associated with that relevance aggregator, which can make it look like the connector is landing directly on the junction in this example orientation.
  • The junction will grow from zero to its planned size on the affected confidence connection. That planned size includes how wide the relevance landing area is, how thick the incoming Display Confidence Connector side is, and how thick the outgoing Delivery Connector side is.
  • Relevance can make that outgoing Delivery Connector side either wider or narrower than the incoming Display Confidence Connector side.
  • If the relevance aggregator needs to become visible, it will grow out from the junction edge as a separate item from the junction.
  • The Display Confidence Connector stays attached to the source-facing edge of the junction and the Delivery Connector stays attached to the delivery aggregator on the target claim side. The delivery aggregator remains visible even when it has only one incoming connector.
• First Fill: The score fluid progressively fills the new pipe.
• Wave: Start the update wave at the target Relevance Aggregator Adjust step.

Update Wave

The Update Wave is a process that propagates changes through the graph starting when a claim receives an update to its score.

• Start at a claim that finished adjusting.
• If the outgoing connector is a Relevance Connector:
  • Relevance Connector Adjust:
    • The Relevance Connector progressively adjusts to the new score.
  • Relevance Aggregator Adjust:
    • If visible, the relevance aggregator (progressively?/uniformly?) adjusts to the new score.
• Else if the outgoing connector is a Confidence Connector:
  • Confidence Connector Adjust: If visible, the Display Confidence Connector progressively adjusts to the new score.
• Junction Adjust: If visible, the junction progressively adjusts to the new score.
• Delivery Connector Adjust: The Delivery Connector progressively adjusts to the new score.
• Delivery Aggregator Adjust: The delivery aggregator (progressively?/uniformly?) adjusts to the new score.
  • Do we want the scales or orders to happen here instead of at the end?
• Claim Adjust: The claim adjusts to its new score.
• If it is the Main Claim, then continue to the next step. If not, then go to the start of this list.
• Scale: The whole graph will uniformly adjust scales when the potential pipe-size calculation changes. Current score still only controls the fraction of fluid shown inside those pipe sizes.
• Order: The whole graph will uniformly adjust the order of claims and targets to and inside the aggregators based on all the new scores. The same shared ordering rule must still be used for both claim order and target-edge connector stack order.