告诉HN：氛围编码分类法
我现在有一个分类体系。前三级属于程序员领域，第四级则对程序员和非程序员都开放。
1. 传统编码（不使用人工智能。你只是手动完成）
2. 强力编码（“添加一个函数 fooBlarg，并将其连接到这些数据结构”）
3. 旁观编码（“你能重构一下这段代码，让它更优雅吗？”）
4. 气氛编码（完全的瑞克·鲁宾模式……“什么是变量？”）
我认为区分这些是有帮助的，因为我经常听到人们将第二和第三级称为“气氛编码”，而原本的意思是“根本不看代码，仅凭感觉。”
查看原文
So I have this taxonomy right now. The first 3 levels live in Programmer Land. The 4th one is available to both programmers and non-programmers.<p>1. Tradcoding (no AI. You just do it manually)<p>2. Power Coding ("Add a function fooBlarg and hook it up to these data structures")<p>3. Backseat Coding ("Can you refactor this code it so it's more elegant")<p>4. Vibe Coding (Full Rick Rubin Mode... "What's a Variable?")<p>I think it's helpful to distinguish between these, because I often hear people call levels 2 and 3 "vibe coding" as well, whereas the original meaning was "not even looking at the code, just going on vibes."