When code is abundant quality becomes the differentiator, but what is good code? New blog post from Earendil engineer Sebastian B on why quantifying the ‘sloppiness’ of code is difficult, which eval methods don’t work well, and what metrics matter. Link to the post below

Read “If code is solved, what now?: Measuring the sloppiness of code”

@SebastianBaye We need to talk more about this. More is not better. Simple is not easy.

@SebastianBaye 现在代码不缺了，拼的就是质量，怎么量化代码质量这个问题确实挺头疼，这篇文章得去看看。

@SebastianBaye One indicator I use is CRAP , and also AST to see the depth of the code being written. I feel crap is a good way to capture the cc with unit test . It’s a good idea on the unnecessary complexity that ai provides

@SebastianBaye That post doesn't seem to get the attention it deserves. Do people not care? They should.

@SebastianBaye Love the design of the blog. And it's content.

@SebastianBaye 现在终止模型输出的快捷键是Esc，但有时你需要在多个窗口中切换焦点，而esc通常是其他一般软件的退出。当你本意是在另一个窗口按下esc来退出或返回，而焦点却没能正确的切换出去，这会导致错误的中断模型输出。 因此我建议使用两次esc作为pi中模型终止输出的快捷键，而不是一下。

@SebastianBaye once codegen is cheap, measuring sloppiness matters more than counting LOC — failed eval methods are half the story

@SebastianBaye agent 一天塞几十个 PR 时，「好代码」很难打分。改动面、失败复现、审不过的次数，比抽象草率分更管用。

@SebastianBaye Will there be a desktop version?

@SebastianBaye What's more cringe is the amount of AI generated comments on this thread. Your em dashes and non-ASCII characters are a dead giveaway. You are perpetuating slop, unrelated to code.

@SebastianBaye For verbosity, could @CFDevelop deslop help?

@SebastianBaye Good read

@SebastianBaye Cool

@SebastianBaye The hard part seems to be measuring sloppiness without rewarding cosmetic style. I wonder if the useful signal is semantically correct code with brittle interfaces, rather than a cleaner looking diff.

@SebastianBaye What metric actually held up something you can automate in CI, or does it still require a reviewer in the loop?

@SebastianBaye Sloppiness metrics still lie on agent PRs