Skip to content

FAQ.md

Latest commit

 

History

History
265 lines (185 loc) · 7.93 KB

FAQ.md

File metadata and controls

265 lines (185 loc) · 7.93 KB

Frequently Asked Questions (FAQ)

License & Legal Questions

Q: Can I use LinguaLint in my commercial product?

A: Yes! You can use LinguaLint commercially under AGPL-3.0, but with conditions:

  • ✅ Internal use: Use LinguaLint within your company for analysis, reports, etc.
  • ✅ Desktop applications: Distribute LinguaLint as part of desktop software
  • ✅ Open source SaaS: Run LinguaLint as a web service if you share your source code
  • ❌ Proprietary SaaS: Cannot run as a closed-source web service without commercial license

Need proprietary licensing? Contact jefferson@richards.plus

Q: What does "network service" mean in AGPL?

A: Web services, APIs, or cloud applications. If users interact with LinguaLint through a network (web browser, API calls, etc.), you must provide your source code under AGPL-3.0.

Examples:

  • ✅ OK: Using LinguaLint to analyze your company's 10-K filings internally
  • ✅ OK: Distributing LinguaLint as part of an open-source desktop app
  • ❌ Requires commercial license: Running LinguaLint as a proprietary web service

Q: Can I modify LinguaLint without sharing my changes?

A: Depends on how you use it:

  • ✅ Internal modifications: Keep private if only used within your organization
  • ❌ Distributed modifications: Must share if you distribute the software
  • ❌ Network service modifications: Must share if you run it as a web service

Q: Is AGPL compatible with my MIT/BSD project?

A: No, not directly. AGPL is a copyleft license that's incompatible with permissive licenses like MIT/BSD. You would need:

  1. Separate the components: Keep LinguaLint as a separate service/process
  2. Commercial license: Get a proprietary license for integration
  3. Dual licensing: We offer commercial licenses for this use case

Q: What's the difference between GPL and AGPL?

A: Network services.

  • GPL: Only requires source sharing when you distribute the software
  • AGPL: Also requires source sharing when you run it as a network service

AGPL closes the "SaaS loophole" where companies could use GPL software in web services without sharing improvements.

Technical Questions

Q: What are the system requirements?

Minimum:

  • Python 3.10+
  • 4GB RAM
  • 2GB disk space
  • Internet connection (for Wikipedia enrichment)

Recommended:

  • Python 3.10+
  • 8GB RAM
  • 5GB disk space
  • SSD storage for better performance

Q: Can I run LinguaLint offline?

A: Mostly yes. Core NLP processing works offline, but:

  • ✅ Text analysis: Works completely offline
  • ✅ Report generation: Works offline
  • ❌ Wikipedia enrichment: Requires internet connection
  • ❌ Package installation: Requires internet for initial setup

Q: What file formats does LinguaLint support?

Input:

  • Plain text (.txt)
  • JSON files
  • Direct text input via web interface or CLI

Output:

  • JSON (structured data)
  • HTML (interactive reports)
  • PNG (visualizations)
  • PDF (comprehensive reports)
  • CSV (project plans)

Q: How accurate is the NLP analysis?

A: Depends on the content type:

  • Financial documents: Optimized for 10-K filings, SEC reports
  • News articles: Good performance on structured news content
  • General text: Reasonable performance, but may need tuning
  • Technical documents: May require domain-specific customization

Accuracy factors:

  • Text quality and structure
  • Domain-specific terminology
  • Language complexity
  • Document length

Q: Can I customize the analysis for my domain?

A: Yes, several ways:

  1. Semantic Primes: Extend Wierzbicka's 65 primes for your domain
  2. Custom Models: Train SpaCy models on your data
  3. Wikipedia Sources: Use domain-specific knowledge bases
  4. Vector Weights: Adjust warm/cold vector calculations
  5. Custom Reports: Modify HTML/PDF templates

Usage Questions

Q: How do I get started quickly?

A: Three options:

  1. Web Interface (easiest):

    node web-server.js
# Open http://localhost:3001

  2. Command Line:

    python3.10 run.py "Your text here"

  3. MCP Server (for AI assistants):

    python3.10 server.py

Q: What's the difference between the interfaces?

  • Web Interface: User-friendly, good for occasional use
  • Command Line: Scriptable, good for batch processing
  • MCP Server: Integration with AI assistants like Claude
  • Python API: Direct integration into other Python applications

Q: How do I process multiple documents?

A: Batch processing:

# Process multiple files
for file in *.txt; do
    python3.10 run.py --file "$file"
done

# Or use Python API
from src.nlp_processor import ModernNLPProcessor
processor = ModernNLPProcessor()

for filename in file_list:
    with open(filename, 'r') as f:
        result = processor.process_text(f.read())

Q: Can I integrate LinguaLint with my existing system?

A: Yes, several integration options:

  1. REST API: Run the web server and make HTTP requests
  2. Python Library: Import and use directly in Python code
  3. Command Line: Call via subprocess from any language
  4. MCP Protocol: Integrate with AI assistants and tools
  5. File Processing: Process files and read JSON outputs

Performance Questions

Q: How fast is LinguaLint?

A: Typical performance:

  • Short text (< 1KB): < 1 second
  • Medium document (10KB): 2-5 seconds
  • Large document (100KB): 10-30 seconds
  • Very large (1MB+): 1-5 minutes

Performance factors:

  • Text length and complexity
  • Wikipedia enrichment (adds network latency)
  • Hardware specifications
  • Python/SpaCy model size

Q: How can I improve performance?

A: Several optimizations:

  1. Disable Wikipedia enrichment for faster processing
  2. Use SSD storage for better I/O performance
  3. Increase RAM for larger documents
  4. Batch processing for multiple documents
  5. Custom SpaCy models (smaller = faster)

Q: Can I run LinguaLint in production?

A: Yes, with considerations:

  • Scaling: Single-threaded by default, use multiple processes
  • Memory: Monitor memory usage for large documents
  • Caching: Cache Wikipedia results for repeated concepts
  • Error handling: Implement proper error handling and logging
  • Monitoring: Monitor performance and resource usage

Support Questions

Q: How do I report bugs or request features?

A: GitHub Issues:

  1. Search existing issues first
  2. Use issue templates for bugs and features
  3. Provide detailed information (version, OS, steps to reproduce)
  4. Include sample data if possible (anonymized)

Q: How do I get help with implementation?

A: Multiple channels:

  • GitHub Discussions: Community Q&A
  • Documentation: Check /docs/ directory
  • Examples: See /examples/ directory
  • Code: Read the source code (it's open!)
  • Email: jefferson@richards.plus for general questions

Q: Do you offer commercial support?

A: Yes! We offer:

  • Commercial licenses for proprietary use
  • Professional support contracts
  • Custom development services
  • Training and consulting

Contact jefferson@richards.plus for enterprise options.

Q: How do I contribute to LinguaLint?

A: We welcome contributions!

  1. Read CONTRIBUTING.md
  2. Follow CODE_OF_CONDUCT.md
  3. Fork the repository
  4. Create a feature branch
  5. Submit a pull request

Types of contributions:

  • Bug fixes and improvements
  • New features and enhancements
  • Documentation and examples
  • Tests and quality assurance
  • Translations and localization

Still have questions?