learn-from-the-best

██╗     ███████╗ █████╗ ██████╗ ███╗   ██╗
██║     ██╔════╝██╔══██╗██╔══██╗████╗  ██║
██║     █████╗  ███████║██████╔╝██╔██╗ ██║
██║     ██╔══╝  ██╔══██║██╔══██╗██║╚██╗██║
███████╗███████╗██║  ██║██║  ██║██║ ╚████║
╚══════╝╚══════╝╚═╝  ╚═╝╚═╝  ╚═╝╚═╝  ╚═══╝

███████╗██████╗  ██████╗ ███╗   ███╗
██╔════╝██╔══██╗██╔═══██╗████╗ ████║
█████╗  ██████╔╝██║   ██║██╔████╔██║
██╔══╝  ██╔══██╗██║   ██║██║╚██╔╝██║
██║     ██║  ██║╚██████╔╝██║ ╚═╝ ██║
╚═╝     ╚═╝  ╚═╝ ╚═════╝ ╚═╝     ╚═╝

████████╗██╗  ██╗███████╗    ██████╗ ███████╗███████╗████████╗
╚══██╔══╝██║  ██║██╔════╝    ██╔══██╗██╔════╝██╔════╝╚══██╔══╝
   ██║   ███████║█████╗      ██████╔╝█████╗  ███████╗   ██║
   ██║   ██╔══██║██╔══╝      ██╔══██╗██╔══╝  ╚════██║   ██║
   ██║   ██║  ██║███████╗    ██████╔╝███████╗███████║   ██║
   ╚═╝   ╚═╝  ╚═╝╚══════╝    ╚═════╝ ╚══════╝╚══════╝   ╚═╝

A structured framework for finding the best sources, people, and learning paths before you start anything new.
Based on the Mondon Method: "In 99% of cases, someone has already done what you're trying to do. The information exists — you just need to know where to look."

The Problem · The 8 Steps · Quick Start · Examples · FAQ

The Problem

You want to learn something new — a skill, a field, a technology. You could spend weeks reading random blog posts, watching YouTube videos, and following advice from self-proclaimed experts. Or you could spend a few hours mapping the landscape first: Who actually knows this? What are the best resources? What mistakes does everyone make?

Most learning frameworks (Ultralearning, DiSSS, First 20 Hours) assume you already know what to learn. None of them systematically help you figure out where to look and who to trust before you start.

That's the gap this framework fills.

Without / With

The 8 Steps

┌──────────────────────────────────────────────────────┐
│                                                      │
│  1. Scope           → What exactly do I want?        │
│  1b. Prerequisites  → What do I need first?          │
│  1c. Inventory      → What do I already have?        │
│  2. Landscape Map   → Who are the best sources?      │
│  3. Mistake Audit   → What goes wrong for most?      │
│  4. 80/20 Focus     → What 20% matters most?         │
│  5. Learning Path   → How do I get there?            │
│  6. Go/No-Go        → Is it worth it?                │
│                                                      │
│  Output: Reconnaissance Briefing                     │
│                                                      │
│  Stop signals: Convergence · Impact · Blocker        │
│                Quality ceiling · Scope boundary      │
│                                                      │
└──────────────────────────────────────────────────────┘

Step 1: Define Scope

What exactly do I want to know or be able to do — and what explicitly NOT?

Not "learn to write a book" but "preserve my domain expertise as a practitioner in a nonfiction book for the next generation of engineers." The sharper the scope, the better the results.

Step 1b: Prerequisites Check

What do I need to know BEFORE I start?

Catch dependencies early. If writing a book requires learning the craft of nonfiction writing first, that's a prerequisite — not a surprise at the end.

Step 1c: Inventory Check

What do I ALREADY have — and is it enough?

Before searching externally, check what you already have. The most common cause of seemingly missing information isn't missing substance — it's missing connections between existing sources. An impact test reveals whether you need new sources or better mapping of what's already there.

Step 2: Draw the Landscape Map

Who or what are the key sources in this field?

The map looks different depending on the topic type:

Find the signal in the noise:

Guru Filter: In fields without peer review (personal branding, coaching), additionally check: Does this person have provable results OUTSIDE of creating content about the topic?

Knowledge Half-Life: Not all knowledge ages the same way.

Step 3: Mistake Audit

What mistakes do beginners typically make — and how do you avoid them?

Search for "common mistakes in X", "what I wish I knew before X", post-mortems, and honest experience reports. Focus on the most expensive mistakes (time, money, motivation).

Step 4: 80/20 Selection

What 20% gives you 80% of the understanding?

Based on Ferriss' DiSSS principle:

1. Deconstruct — What sub-areas make up this field?
2. Select — Which are most important for your specific goal?
3. Sequence — In what order do they make sense?

Step 5: Sketch the Learning Path

Concrete path from "zero" to "capable of action"

The timeframe adapts to the topic:

• Stable, bounded topics: ~4 weeks (writing, cooking)
• Medium complexity: ~8–12 weeks (algo-trading, new framework)
• High complexity / deep theory: ~6+ months (quantum physics, mathematics)

Four phases: Overview → Foundation → First Build → Reflect

For meta-skills (positioning, communication): start doing early, learn from feedback rather than reading everything first.

Step 6: Go/No-Go

Is it worth it — or not?

Evaluate: effort, risk, fit with your goals, timing. The answer can be "Go", "Go with constraints" (dependencies first), or "No-Go" with a clear reason.

When to Stop Researching

Every research step needs a stop signal. The framework has 5 built-in signals — when any one fires, you can move on:

Safety net: If after 3 rounds of searching none of these signals fires, stop and decide consciously rather than researching indefinitely.

Stable topics (physics, writing craft) typically converge in 1–2 rounds. Volatile topics (AI tools, frameworks) need targeted searches in the right places — docs and changelogs, not books.

Output Format

Every run produces a Reconnaissance Briefing:

# Reconnaissance Briefing: [Topic]

## Scope
[2–3 sentences]

## Prerequisites
[What needs to be in place? Status: available / missing / separate project needed]

## Inventory Check
**Already available:** [What exists already?]
**Diagnosis:** [Substance gap (new source needed) OR Bridge gap (mapping needed)]

## Landscape Map
**Knowledge Half-Life:** [Stable / Medium / Volatile]
**Briefing Expiry:** [Valid until approx. YYYY-MM]
**Topic Type:** [Practice field / Standards-norms / Research]
**Top Sources:** [3–5 with quality grade]
**Top Works:** [3–5 with quality grade]
**Communities:** [1–2]
**State of the Art:** [1–2 sentences]

## Mistake Audit
1. [Mistake]: [How to avoid]
2. ...

## 80/20 Focus
[Prioritized sub-areas]

## Learning Path
**Timeframe:** [adapted to topic]
[Concrete plan]

## Go/No-Go
[Recommendation + reasoning, including dependencies]

When to Use This

Best for: Entering a new field, career pivots, ambitious projects, learning something with real stakes.
Overkill for: Quick how-tos, single-tool lookups, topics you already know well.

The sweet spot is when the cost of learning from the wrong sources is high — in time, money, or motivation.

Quick Start

No installation needed. This is a thinking framework, not software.

Option A — Use it manually:

1. Pick a topic you want to explore
2. Copy the briefing template and walk through the 8 steps
3. Fill in each section as you go

Option B — Use it with an AI assistant:

1. Copy the framework prompt into your AI tool
2. Tell it your topic
3. Walk through the steps together — the AI handles research, you make decisions

Option C — Use it as a Claude Code skill:

# Copy the skill into your Claude Code skills directory
cp -r skill/ ~/.claude/skills/learn-from-the-best/

Examples

See the examples/ directory for complete Reconnaissance Briefings:

• Writing a Nonfiction Book — Stable knowledge field, clear Go
• Algorithmic Trading — Medium complexity, Go with dependency on financial market knowledge

Architecture

learn-from-the-best/
├── README.md              ← You are here
├── framework/
│   └── PROMPT.md          ← The full framework as a reusable prompt
├── skill/
│   └── SKILL.md           ← Claude Code skill version
├── templates/
│   └── briefing-template.md ← Blank template to fill in
├── examples/
│   ├── nonfiction-book.md ← Example briefing: writing a book
│   └── algo-trading.md    ← Example briefing: algorithmic trading
└── docs/
    └── index.html         ← Landing page (GitHub Pages)

Usage Modes

Methodological Foundations

This framework doesn't reinvent the wheel. It synthesizes proven methods and adds what they cover only lightly:

The Mondon Method — named after a mentor who believed that in 99% of cases, someone has already solved your problem — ties these together with a systematic reconnaissance phase that comes before any of these frameworks kick in.

Roadmap

• Core framework (8 steps)
• Source quality grading (A–D + Guru Filter)
• Knowledge half-life classification
• Prerequisites check (dependency detection)
• Inventory check with impact test (v3)
• Topic-type-aware landscape mapping (v3)
• Research depth heuristic with 5 stop signals (v3)
• Emerging patterns from real-world usage (v3, grows with contributions)
• Author-tested across 4 real topics + 3 stress tests (nonfiction writing, algo-trading, EV charging data, scientific quality standards — stress-tested on quantum computing, personal branding, AI agents)
• More example briefings (contributions welcome — see Contributing)

FAQ

Why not just ask ChatGPT to "research X for me"?
You can — and this framework works great with AI. The difference: a single prompt gives you a one-shot answer that varies with phrasing. The 8 steps give you a reproducible process with specific tools (guru filter, source grading, knowledge half-life) that catch things a generic prompt misses. The output is a reusable briefing you can reference, update, and share — not a chat message you'll lose.

Is this just another productivity framework?
No. Existing frameworks tell you how to learn. This one tells you where to look and who to trust before you start learning. It's the reconnaissance phase that comes before Ultralearning, DiSSS, or any other method.

Do I need AI to use this?
No. The 8 steps work with pen and paper, a search engine, and your own judgment. AI assistants speed up the research phase but aren't required.

How do I know when to stop researching?
The framework has 5 built-in stop signals: convergence (same names keep appearing), impact threshold (existing resources are sufficient), blocker (prerequisite missing), quality ceiling (no A/B sources left), and scope boundary (your question is answered). If none fires after 3 rounds, it's time to decide consciously.

How long does a full run take?
With an AI assistant: 30–60 minutes. Manually: 2–4 hours depending on the topic's complexity. Either way, it's a fraction of the time you'd waste learning from the wrong sources.

What if my topic is too niche?
The framework scales. For niche topics, the landscape map will be smaller (fewer experts, fewer works) — but the mistake audit and 80/20 selection become even more valuable because there's less margin for error.

Why "Mondon Method"?
Named after Andrew Mondon, a mentor who consistently demonstrated that the fastest path to mastery starts with finding who's already walked it. The principle is simple: don't reinvent — reconnect.

Contributing

Contributions welcome. Priority areas:

• Example briefings for new topics (use the output format)
• Translations (framework prompt + examples)
• Improvements to the Guru Filter criteria

Please open an issue before submitting large changes.

License

MIT