Summary
A reflective essay by creative coder Ahmad Moussa arguing that code sketches are not throwaway experiments but seeds of long-term creative development. Through the metaphor of “planting seeds,” the author shows how small exploratory sketches compound into mature works over time — an irregular grid algorithm becoming a dungeon generator, a simple shape study unlocking an entirely new output space. Sketching is framed as a dialogue with the machine: imperfect, exploratory, and fundamentally generative.
創意程式設計師 Ahmad Moussa 的反思文章，主張程式碼素描不是可拋棄的實驗，而是長期創意發展的種子。透過「播種」的隱喻，作者展示小型探索性素描如何隨時間累積成熟——一個不規則網格演算法演變成地牢生成器，一個簡單的形狀研究解鎖了全新的輸出空間。素描被定義為與機器的對話：不完美、探索性，且從根本上是生成性的。
Key Points
- Sketching is “solving problems you don’t have yet” — small experiments build a repertoire of reusable techniques
- Ideas develop across the span of your entire body of work, not just within a single project (Zach Lieberman’s “sense of time”)
- Style emerges through addition and subtraction: incorporating what you enjoy, excluding what you don’t
- A sketch is a dialogue: you input thoughts as code, the machine returns interpretations — including flawed ones that push exploration forward
- In long-form generative art, the system and its output space become the artwork, not individual outputs
- Save your random seeds — once a beautiful output is gone, it may be gone forever
- Creative coding is an effective attention-engagement practice; it reliably induces flow state
Insights
- The “dialogue with the machine” framing maps directly onto modern AI-assisted coding — the flaws in the machine’s interpretation are not bugs but fuel for refinement; this is structurally the same as iterating with Claude
- “Sketching is solving problems you don’t have yet” is a strong framing for why exploratory side projects compound: the techniques you discover become available when a real problem matches them
- Vera Molnar’s practice (systems-first, even before computers) is a reminder that generative thinking predates generative tools — the mindset is the skill, not the software
- The essay’s implicit argument is that quantity of sketches matters: you can’t know which seed will bear fruit, so planting many is the right strategy
- This connects to the broader theme of deliberate practice and taste development — the author building a style through “hundreds of hours studying” parallels the taste-building discussed in other clippings
Connections
- Generative Art
- Creative Coding
- Practice
- Everyone using AI has about 12 months to develop these 3 moats
- p5.js
- Long-form Generative Art
Raw Excerpt
Sketching is a practice where you’re solving problems that you don’t have yet.