Show Notes
Parker argues that vibe coding will burn out if you skip a core skill: effective planning. This daily dives into the Abe Lincoln method for planning, how to apply it with AI tools, and why you should lock in a repeatable workflow before chasing new tech toys.
Abe Lincoln method: sharpen before you swing
- The core idea: spend most of your time planning (planning > coding).
- Apply it to AI coding: map out questions, constraints, and steps before touching code.
- For noobs: ask for a step-by-step plan first, then execute.
- For experienced coders: use planning to raise your game and avoid scope creep.
How to apply in practice
- Start with a clear prompt that defines the plan (e.g., “Generate a plan to move blog content to the homepage with no SEO tagging for now.”).
- Use planning tools or modes (Plan mode / Architect mode) to lay out the steps before writing any files.
- Expect to spend roughly two-thirds of your time planning; execution follows more smoothly.
The three cohorts of users and the market reality
- Noobs: enthusiastic but often overwhelm themselves with tech ladders (Next.js, TypeScript, UI kits) without fundamentals.
- Mid-level coders: solid skills but need better systems and processes; may cycle through tools unless there’s a repeatable method.
- Daily coders: strong hands, need branding and a repeatable method to scale, often overpromising prompts without structure.
- Opportunity: codify a universal planning method, brand it, and apply it across versions (zero to one, one to 1.1, etc.).
actionable takeaway
- Create a simple repeatable plan framework you can reuse for any new project, then tailor the details per project.
Tooling hype vs. real-world use
- Don’t overreact to VC hype (Windsurf, Cursor, etc.). One bad day with a tool doesn’t prove it’s useless; all major players converge on similar capabilities.
- Tool-hopping is costly. Pick one that fits your flow and stick with it; reassess after a few weeks if you need to.
- Figma vs Canva: design-to-code bridges are accelerating, but each has strengths. Figma’s AI features are evolving; Canva is chasing AI for marketing collateral, not UI design.
- Stitchware/bridges exist (Cursor, Windsurf, Builder, Webflow). Expect them to iterate; the best approach is to understand the core problem (handoff, design-to-code) and choose a tool that tackles that well.
- Klein and Ader show how “architect mode” and “map commands” can give you structured control over context and workflow.
actionable takeaway
- Don’t chase every new tool. Pick one that aligns with your planning workflow, learn it deeply, then expand as needed.
A practical workflow you can use tomorrow
- Plan mode (or equivalent) to architect the task:
- Define the target state (e.g., move /blog content to /home, drop SEO tagging for now).
- List nested pages and metadata implications you’ll ignore in the first pass.
- Outline success criteria and potential blockers.
- Act mode (execution):
- Let the tool generate a concrete plan.
- Review, refine, then let it execute step-by-step.
- Iterate quickly on a few cycles to validate outcomes.
- Keep the planning: execution ratio around 2:1 in favor of planning when starting something new.
What to watch for in the tooling world
- Expect design-to-code ecosystems to blur the lines between design and implementation.
- Look for tools that offer robust planning modes and stable context handling (e.g., MCP-based workflows).
- Be mindful of overclaiming features (e.g., “no indexing” or “instant perfect code”)—test with real projects.
Closing thought
If you want to stay ahead, build a portable planning method you can drop into any project. The Abe Lincoln approach isn’t anti-AI; it’s a guardrail that keeps you from boiling the ocean every time you start something new.