Update SCHEMA.md - BlueLedger Enhanced v2.0

SCHEMA.md

@@ -0,0 +1,170 @@
+# BlueLedger Enhanced Dataset Schema v2.0
+
+## Overview
+This document describes the data structure for the BlueLedger Enhanced Police Officer Directory dataset.
+
+## Dataset Structure
+
+### Top Level
+```json
+{
+  "dataset_metadata": { ... },
+  "officers": [ ... ]
+}
+```
+
+### dataset_metadata
+- **name**: Dataset name
+- **version**: Semantic version number
+- **generated_at**: ISO 8601 timestamp
+- **total_officers**: Count of officer records
+- **total_offense_records**: Total offense incidents across all officers
+- **compliance**: App store and data standards compliance info
+- **sources**: List of data sources with verification levels
+
+### officer Record
+
+#### officer_id
+Unique identifier: `BL-{STATE}-{NUMBER}`
+Example: `BL-WA-000001`
+
+#### personal_info
+- **full_name**: Complete name as recorded
+- **name_last**: Last name/surname
+- **name_first**: First name/given name
+
+#### employment
+- **department**: Law enforcement agency name
+- **state**: Full state name
+- **rank**: Officer rank/position
+- **status**: Employment status (Active, Inactive, Terminated, Retired)
+- **badge_number**: Badge/shield number (if available)
+- **hire_date**: Date of hire (ISO 8601)
+- **separation_date**: Date of separation (ISO 8601, null if active)
+
+#### certification
+- **certification_status**: POST certification status
+- **certification_number**: State POST certification ID
+- **certification_date**: Date certification issued
+- **expiration_date**: Certification expiration date
+- **state_post_agency**: State jurisdiction
+- **certifying_authority**: Full name of POST agency
+
+#### verification_sources
+- **primary_gov_source**: Official state POST commission
+  - **url**: Official .gov URL (required for app store compliance)
+  - **agency**: Agency name
+  - **verified**: Verification status (boolean)
+  - **last_verified**: Last verification timestamp
+- **profile_links**: Array of secondary sources (CPDP, etc.)
+
+#### offense_records
+- **total_count**: Total number of recorded incidents
+- **categories**: Breakdown by offense type
+  - use_of_force
+  - misconduct
+  - civil_rights_violation
+  - criminal_charges
+  - policy_violation
+  - excessive_force
+  - other
+- **incidents**: Array of incident objects
+
+##### incident Object
+- **incident_id**: Unique incident identifier
+- **date**: Incident date (ISO 8601)
+- **category**: Incident category
+- **description**: Detailed description
+- **disposition**: Investigation outcome (Sustained, Not Sustained, Exonerated, Unfounded, etc.)
+- **discipline**: Disciplinary action taken
+- **verification_source**: Source documentation
+  - **type**: Source type (Internal Affairs, Court Record, etc.)
+  - **url**: Source URL (preferably .gov)
+  - **document_id**: Official document ID
+  - **verified**: Verification status
+- **status**: Case status (Open, Closed, Under Investigation)
+- **legal_outcome**: Legal proceedings info
+  - **civil_lawsuit**: Boolean
+  - **criminal_charges**: Boolean
+  - **settlement_amount**: Dollar amount if applicable
+
+#### data_quality
+- **completeness_score**: 0-100 score indicating data completeness
+- **has_gov_verification**: Boolean indicating .gov link presence
+- **last_updated**: Last update timestamp
+- **data_source**: Original data source
+
+## Data Standards
+
+### App Store Compliance
+- ✅ All officer records MUST have a valid .gov verification URL
+- ✅ Offense records MUST cite verifiable sources
+- ✅ No unverified allegations included
+- ✅ Clear data provenance and update timestamps
+
+### URL Requirements
+- Primary sources MUST be official government (.gov) domains
+- State POST commission URLs are the gold standard
+- Secondary sources (CPDP, etc.) are supplementary
+
+### Date Format
+All dates use ISO 8601: `YYYY-MM-DD` or `YYYY-MM-DDTHH:MM:SSZ`
+
+### Null Values
+- `null` used for unavailable data (not empty strings)
+- Distinguished from "Unknown" which indicates active uncertainty
+
+## Usage Examples
+
+### Python
+```python
+import json
+
+with open('blueledger_enhanced_v2.json', 'r') as f:
+    data = json.load(f)
+
+# Access officers
+for officer in data['officers']:
+    print(f"{officer['personal_info']['full_name']} - {officer['employment']['department']}")
+    print(f"Gov Verification: {officer['verification_sources']['primary_gov_source']['url']}")
+    print(f"Total Offenses: {officer['offense_records']['total_count']}")
+```
+
+### Pandas
+```python
+import pandas as pd
+
+# Load CSV
+df = pd.read_csv('blueledger_enhanced_v2.csv')
+
+# Filter officers with offenses
+officers_with_offenses = df[df['total_offenses'] > 0]
+
+# Group by state
+state_summary = df.groupby('state').agg({
+    'officer_id': 'count',
+    'total_offenses': 'sum'
+})
+```
+
+## Data Integration Notes
+
+### Adding Offense Records
+Real offense data should be sourced from:
+1. State POST disciplinary databases
+2. Court records (PACER, state court systems)
+3. Internal affairs reports (via FOIA requests)
+4. Verified journalist databases (CPDP, etc.)
+
+### Verification Workflow
+1. Obtain source document
+2. Verify authenticity via .gov source
+3. Extract structured data
+4. Add verification metadata
+5. Update last_verified timestamp
+
+## License
+Public Domain - Government Data
+
+## Contact
+For questions or contributions, see README.md