spec: add linked-views-selection specification (#3348)

## New Specification: `linked-views-selection`

Related to #3344

---

### specification.md

# linked-views-selection: Multiple Linked Views with Selection Sync

## Description

Multiple coordinated plots where selecting data in one view
automatically highlights or filters corresponding data points in all
other views. This brushing-and-linking interaction pattern is
fundamental for exploratory data analysis, allowing users to discover
relationships across different visual representations of the same
dataset. By selecting a cluster in a scatter plot, for example, users
can instantly see how those points distribute in histograms, how they
appear in parallel coordinates, or which categories they belong to.

## Applications

- Exploratory data analysis dashboards where analysts investigate
multivariate datasets by brushing across scatter plots, histograms, and
categorical views
- Customer segmentation tools where selecting a group in one dimension
reveals their characteristics across demographics, behavior, and
transactions
- Scientific data exploration comparing different measurements or
experimental conditions with synchronized highlighting
- Financial portfolio analysis linking asset allocation views with
performance metrics and risk indicators

## Data

- `x` (numeric) - Values for scatter plot horizontal axis
- `y` (numeric) - Values for scatter plot vertical axis
- `category` (categorical) - Group labels for bar chart or color
encoding
- `value` (numeric) - Additional numeric dimension for histogram or
secondary visualizations
- Size: 100-1000 points (interactivity works best with moderate dataset
sizes)
- Example: Iris dataset or simulated multivariate data with numeric and
categorical columns

## Notes

- Implement at least 2-3 coordinated views (e.g., scatter plot +
histogram + bar chart)
- Selection in any view should highlight corresponding points in all
other views
- Use consistent color encoding across all views for the same data
points
- Unselected points should be visually de-emphasized (reduced opacity or
gray)
- Include a clear selection mechanism (brush, click, or lasso)
- Provide a reset/clear selection button to restore full view
- Consider showing selection count or summary statistics
- Libraries with native linked selection support (Altair, Bokeh, Plotly)
should leverage those features

---

**Next:** Add `approved` label to the issue to merge this PR.

---
:robot: *[spec-create
workflow](https://github.com/MarkusNeusinger/pyplots/actions/runs/20832699525)*

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

Author: github-actions[bot]

plots/linked-views-selection/specification.md

@@ -0,0 +1,32 @@
+# linked-views-selection: Multiple Linked Views with Selection Sync
+
+## Description
+
+Multiple coordinated plots where selecting data in one view automatically highlights or filters corresponding data points in all other views. This brushing-and-linking interaction pattern is fundamental for exploratory data analysis, allowing users to discover relationships across different visual representations of the same dataset. By selecting a cluster in a scatter plot, for example, users can instantly see how those points distribute in histograms, how they appear in parallel coordinates, or which categories they belong to.
+
+## Applications
+
+- Exploratory data analysis dashboards where analysts investigate multivariate datasets by brushing across scatter plots, histograms, and categorical views
+- Customer segmentation tools where selecting a group in one dimension reveals their characteristics across demographics, behavior, and transactions
+- Scientific data exploration comparing different measurements or experimental conditions with synchronized highlighting
+- Financial portfolio analysis linking asset allocation views with performance metrics and risk indicators
+
+## Data
+
+- `x` (numeric) - Values for scatter plot horizontal axis
+- `y` (numeric) - Values for scatter plot vertical axis
+- `category` (categorical) - Group labels for bar chart or color encoding
+- `value` (numeric) - Additional numeric dimension for histogram or secondary visualizations
+- Size: 100-1000 points (interactivity works best with moderate dataset sizes)
+- Example: Iris dataset or simulated multivariate data with numeric and categorical columns
+
+## Notes
+
+- Implement at least 2-3 coordinated views (e.g., scatter plot + histogram + bar chart)
+- Selection in any view should highlight corresponding points in all other views
+- Use consistent color encoding across all views for the same data points
+- Unselected points should be visually de-emphasized (reduced opacity or gray)
+- Include a clear selection mechanism (brush, click, or lasso)
+- Provide a reset/clear selection button to restore full view
+- Consider showing selection count or summary statistics
+- Libraries with native linked selection support (Altair, Bokeh, Plotly) should leverage those features

plots/linked-views-selection/specification.yaml

@@ -0,0 +1,32 @@
+# Specification-level metadata for linked-views-selection
+# Auto-synced to PostgreSQL on push to main
+
+spec_id: linked-views-selection
+title: Multiple Linked Views with Selection Sync
+
+# Specification tracking
+created: 2026-01-08T21:41:58Z
+updated: null
+issue: 3344
+suggested: MarkusNeusinger
+
+# Classification tags (applies to all library implementations)
+# See docs/reference/tagging-system.md for detailed guidelines
+tags:
+  plot_type:
+    - scatter
+    - histogram
+    - bar
+    - multi
+  data_type:
+    - numeric
+    - categorical
+  domain:
+    - general
+    - statistics
+    - business
+  features:
+    - interactive
+    - multi
+    - coordinated
+    - brushing