Stop Guessing. Start Building.#

The Great Trigonometric Survey of India (1802-1880) mapped an entire subcontinent—including Mt. Everest—using triangulation from known points. We’re doing the same for software: systematically surveying the library landscape so you can assemble your own mountains.

The Origin Story#

Back in September 2025, at PuPPy Talk Night, I presented research on comparative AI coding methodologies. I took the Claude Code task tool and said “build this thing” with four different approaches working in parallel:

• Immediate implementation
• Specs-driven development
• Test-driven development (TDD)
• Adaptive TDD with mutation testing

The result: a model showing which methodologies fit which programming problems. Adaptive TDD excels at input validation. Specs-driven is ideal for LLM wrappers. Great! Now I knew how to tell AI to write code for different purposes.

But then came the real question: WHY WRITE CODE AT ALL?

We have thousands of excellent Python libraries. All I needed was a way to figure out which library was right for what I’m building, then write glue code to make it all work together.

The Great Trigonometric Survey (But for Software)#

Think about the Great Trigonometric Survey of India, started in 1802 to map the geography of the Indian subcontinent. If you want to know the exact height of Mt. Everest, you don’t climb it with a measuring tape. You start from two points X and Y at sea level, take sightings of point Z on land, and use triangulation to calculate Z’s height. Keep adding triangles and soon you’ve mapped not just the peak, but the entire landscape.

Now think about the most complex software system you can imagine—call it “Mt. Everest.” What is it really? Just an accumulation of smaller pieces: libraries, frameworks, modules, APIs. And now that we can generate software modules quickly with AI, what’s stopping us from assembling our own code mountains?

Answer: Nothing! We’re limited only by our imagination. (Oh, and token budget.)

The Dewey Decimal System for Algorithms#

I created a classification system for software libraries, organized like the Dewey Decimal System:

• 1.001-1.009 - Sorting & Searching Libraries
• 1.010-1.019 - Graph & Network Analysis
• 1.020-1.029 - Math & Statistics
• 1.030-1.059 - String & Text Processing
• 1.060-1.069 - Cryptographic & Hashing
• 1.070-1.099 - Data Structures & Algorithms
• 1.100-1.149 - Machine Learning & AI
• 1.150-1.159 - Geometric & Spatial
• 1.160-1.169 - Compression & Encoding
• 1.170-1.209 - LLMs & Natural Language

The system starts from 1.001 Sorting Libraries and continues expanding across algorithms, data structures, ML frameworks, and infrastructure components.

The 4PS Methodology: Four-Pass Survey#

I conduct deep research in each category following what I call 4PS - The Four-Pass Survey:

S1: Rapid Discovery#

Find the popular libraries. Quick comparison table. When you need an answer now.

S2: Comprehensive Analysis#

Understand the entire solution space. Feature matrices. Deep comparisons. When you need to know everything.

S3: Need-Driven Discovery#

Start with requirements, find exact-fit solutions. Scenario-based selection. When you know exactly what you’re building.

S4: Strategic Selection#

Think long-term. Consider maintenance, team expertise, ecosystem fit, vendor stability. When you’re making architectural decisions.

Plus: An explainer pass for technical generalists who need context without overwhelming detail.

The Research#

Each survey piece represents deep investigation using the 4PS methodology:

• Systematic coverage across algorithms, data processing, ML, and infrastructure
• 40,000+ tokens of analysis per category
• Publicly accessible at research.modelcitizendeveloper.com

Having this repository makes it possible to do extremely rapid development with greater confidence because I don’t have to:

• Guess which library is right
• Let Claude pick one randomly
• Roll the dice on unmaintained packages
• Stop what I’m doing to do deep research

The map already exists. I just consult it and build.

What Makes This Different#

NOT: Another Tutorial Site#

• Tutorials teach “how to use X”
• We teach: “Why X over Y, with measured evidence”

NOT: Another Awesome List#

• Awesome lists are curated links
• We provide: Deep analysis with methodology and confidence levels

NOT: Another Framework#

• Frameworks lock you in
• We give you: Components you understand and own

YES: A Surveyed Landscape#

• “Here’s what we measured”
• “Here’s how we measured it”
• “Here’s the working implementation”
• “Replicate it yourself”

Build Your Own Mountain#

If you build something using this survey, I want to hear about it!

I write case studies and would love to feature your project. Whether it’s a production system, a research project, or a weekend hack—if you assembled your code mountain using this map, let’s share the story.

The vision: Make it possible for developers and organizations to DIY solutions they would have paid vendors for before. The map is open. The methodology is transparent. The code is yours.

Quality Bar & Limitations#

Research-Grade Confidence#

Overall confidence: 70-80% (directionally strong guidance) Not: Consumer Reports 95%+ certainty

This research provides strategic direction, not absolute guarantees:

• S1 Rapid: 70-80% confidence (speed-optimized)
• S2 Comprehensive: 80-90% confidence (depth-optimized)
• S3 Need-Driven: 75-85% confidence (context-specific)
• S4 Strategic: 60-70% confidence (forward-looking analysis)

Information Decay#

Library ecosystems evolve constantly:

• Pricing changes
• Feature updates
• New entrants
• Vendor pivots
• Acquisitions

Estimated accuracy:

• At publication: 70-80%
• 12 months: 50-70%
• 36 months: <30%

We aim for quarterly updates to major categories and treat research as living documents that guide investigation, not gospel truth.

The Road Ahead: Systematic Expansion#

The Great Trigonometric Survey took 70 years to map India. We’re mapping a different kind of landscape, but the principle holds: systematic measurement, one triangulation point at a time.

Each new survey makes the next easier. Patterns emerge across categories. Reusable methodologies compound.

Ready to start building? Browse the Research Library →