每日科技文摘 2026-09-30Daily Tech Digest 2026-09-30
发布于 2026-09-30Published 2026-09-30
今日深读Today's Deep Read
AI 时代,为什么大多数人反而更有机会了?In the AI era, why does everyone else actually have more opportunity now?
Ethan Mollick 是沃顿商学院的教授,也是最早系统研究 AI 如何改变工作和学习的人之一。他的这篇文章提出了一个反直觉的观点:AI 能力的快速提升,对大多数人来说不是威胁,而是机会。Ethan Mollick is a Wharton professor and one of the earliest researchers to systematically study how AI changes work and learning. In this essay, he makes a counterintuitive argument: the rapid advancement of AI is not a threat to most people, but an opportunity.
他的核心概念叫 "能力悬置"(overhang)——当 AI 模型的能力远超普通人的使用水平时,两者之间就出现了一个巨大的落差。这个落差不是坏事,恰恰相反,它是普通人能抓住的机会窗口。His core concept is the "overhang" — when AI capabilities far exceed the average person's ability to use them, a huge gap emerges. This gap is not a problem; it is an opportunity window that anyone can take advantage of.
为什么会出现这种悬置?因为 AI 的学习曲线和人的学习曲线是完全不同的。AI 模型在一次训练中就能学会大量技能,但人类的学习是一个漫长、渐进的过程。我们花几年才能精通某个领域,而 AI 几乎瞬间就达到了很高的水平。这就造成了一个奇怪的现象:AI 在大多数任务上已经超越普通人,但绝大多数人还不会使用它。Why does this overhang exist? Because the learning curves of AI and humans are completely different. An AI model can acquire vast capabilities in a single training run, while human learning is slow and incremental. We spend years mastering a domain, while AI reaches high proficiency almost instantly. The result is a strange situation: AI surpasses most people on most tasks, yet most people still do not know how to use it well.
Mollick 提出了四个普通人可以在这个悬置时代保持竞争优势的领域。第一个是深度知识:AI 掌握大量信息,但它不理解你在自己领域的细微差别。一个资深工程师对系统架构的直觉、一个医生对患者病史的综合判断,这些深度知识是 AI 无法替代的。Mollick identifies four areas where ordinary people can maintain a competitive edge during this overhang. The first is deep knowledge: AI may know a lot, but it does not understand the nuances of your specific domain. A senior engineer's intuition about system architecture, a doctor's ability to synthesize a patient's full history — these forms of deep, contextual knowledge are irreplaceable by AI.
第二个是广博知识:跨领域的知识连接。AI 可以在单一领域表现出色,但真正的创新往往来自不同领域的交叉。懂编程的医生、懂心理学的产品经理、懂数学的音乐家——这种跨界的知识组合是 AI 难以复制的。The second is broad knowledge: cross-domain connections. AI can excel in a single field, but genuine innovation often arises from the intersection of different domains. A programmer who understands medicine, a product manager who knows psychology, a musician who grasps mathematics — these interdisciplinary combinations are hard for AI to replicate.
第三个是品味:知道什么是好的。AI 可以生成无数方案,但判断哪个方案最好的能力,来自于长期的实践和经验。品味不是天赋,而是通过大量阅读、大量实践积累起来的直觉。The third is taste: knowing what is good. AI can generate countless options, but the ability to judge which option is best comes from years of practice and experience. Taste is not innate talent; it is intuition built through extensive reading and doing.
第四个是能动性:决定做什么。AI 可以帮你完成很多事情,但它不知道你应该做什么。目标设定、价值判断、意义赋予——这些仍然是人类独有的能力。一个会用 AI 但没有明确目标的人,远不如一个目标清晰、善用 AI 工具的人。The fourth is agency: deciding what to do. AI can help you accomplish many things, but it does not know what you should do. Goal setting, value judgment, meaning-making — these remain uniquely human capabilities. A person who knows how to use AI but lacks clear goals will always lose to someone who has clear goals and uses AI as a tool.
他举了两个亲身实验:模型把 1977 年的纯文字游戏《Zork》改成了可玩的 3D 动作游戏,场景全由模型设计;另一个模型重建作家埃科在米兰公寓的藏书,逐帧读视频里的书脊,从两万七千个格子里认出约五千本书,并标出哪些只是猜测。这些活儿人类团队要做几周——但没有一个项目是自己发生的:题是他选的,他对 Zork 和埃科的了解让他几秒就看出 AI 哪里错了、下一步要什么。He cites two personal experiments: one model turned the 1977 text-only adventure Zork into a playable 3D action game, designing every scene itself; another reconstructed Umberto Eco's apartment library in Milan, reading video frames spine by spine to identify some 5,000 of 27,000 shelf slots, marking which were guesses. A human team would have taken weeks — but not one of these projects happened on its own. Mollick chose them, and his knowledge of Zork and of Eco let him see in seconds where the AI went wrong and what to ask for next.
所以他建议:别和 AI 比谁产出多,那是必输的比赛;带着四种人的独有优势和 AI 协作,做出两者单独都做不出的东西。他尤其强调“能动性”:所有人都说不清 AI 的边界在哪时,愿意自己去试——与其等别人宣布,不如自己试出来。他坚持做奇怪的 AI 实验:实验本身就是给前沿绘制地图。So his advice: do not compete with AI on producing output — a losing game. Bring your four human advantages to working with AI, and make things neither you nor it could make alone. He stresses agency: when nobody can say where AI's boundaries are, be the one who tests them — rather than waiting for an announcement, discover it by trying. That is why he keeps running weird AI experiments: the experimenting itself is the map-making.
这篇文章最有价值的地方,不在于它给出了四个答案,而在于它改变了问题的框架。面对 AI 时,大多数人问的是 "我该如何和 AI 竞争",而 Mollick 提醒我们问的是 "在 AI 能力超出大多数人的这段时间里,我该如何利用这个落差"。这种视角转换本身就很有价值。The most valuable aspect of this essay is not that it gives four answers, but that it reframes the question. When facing AI, most people ask "How do I compete with AI?" Mollick reminds us to ask instead "How can I exploit this gap while AI capabilities remain far ahead of most people's ability to use them?" This shift in perspective is itself the key insight.
AI 思考AI Thoughts
- 数学家们,这样来面对你们的“存在危机”Mathematicians, here's a way to think about your existential crisis
- 问题:AI 正在快速掌握数学证明和编程技巧,数学家的核心价值是什么?The problem: AI is rapidly mastering mathematical proofs and programming techniques. What is the core value of a mathematician?
- 关键想法:Kent Beck 提出 "特性与未来"(Features & Futures)框架。任何工作都有两个部分:可见的部分(特性、证明、代码)和不可见的部分(理解、抽象、判断)。AI 擅长前者,但后者同样重要。Key idea: Kent Beck proposes the "Features & Futures" framework. Every piece of work has two parts: the visible part (features, proofs, code) and the invisible part (understanding, abstraction, judgment). AI excels at the former, but the latter is equally important.
- 为什么成立:你写出一段代码或证明一个定理,这是 "特性"。但理解为什么这段代码是正确的、为什么这个证明优雅、什么时候应该用这个方案而不是另一个——这些是 "未来" 的工作。AI 可以在特性层面帮你,但无法替代你对问题的理解深度。Why it holds: Writing code or proving a theorem is the "feature." But understanding why the code is correct, why the proof is elegant, when to choose this approach over another — these are "future" work. AI can assist at the feature level, but cannot replace your depth of understanding.
- 能带走什么:不要和 AI 竞争谁写得更快、证明更短。要去培养那种 "知道为什么" 的深度理解,那才是你不可替代的核心能力。What to take away: Do not compete with AI on who writes faster or proves shorter. Cultivate the deep understanding of "why" — that is your irreplaceable core capability.
- 标题与内情故事:数学与科学中 AI 的理解与信任Headlines and inside stories: understanding and trust in AI for mathematics and science
- 问题:AI 可以给出预测结果,但我们能否信任它?The problem: AI can produce predictions, but can we trust them?
- 关键想法:Tapio Schneider 区分了两种知识:episteme(理解)和 techne(预测)。AI 擅长 techne —— 给出准确的预测。但在需要真正理解的地方,我们必须坚持 episteme。Key idea: Tapio Schneider distinguishes two kinds of knowledge: episteme (understanding) and techne (prediction). AI excels at techne — producing accurate predictions. But where genuine understanding is required, we must insist on episteme.
- 为什么成立:AI 可以预测天气、预测蛋白质结构、甚至辅助数学证明。但预测本身不等于理解。如果我们只信任预测而不追问理解,就会陷入一种危险:我们不知道 AI 为什么给出某个答案,也就无法判断它何时会出错。在科学和工程中,信任必须建立在可审计的因果链上。Why it holds: AI can predict weather, predict protein structures, even assist in mathematical proofs. But prediction is not the same as understanding. If we only trust predictions without asking why, we enter dangerous territory: we cannot know when AI will go wrong. In science and engineering, trust must be built on auditable causal chains.
- 能带走什么:AI 是强大的工具,但在使用时,我们必须区分 "能预测" 和 "能理解"。对于那些可以验证的环节,放心用 AI;对于那些需要信任的预测,一定要嵌入可审计的因果链。What to take away: AI is a powerful tool, but when using it, we must distinguish between "can predict" and "can understand." For verifiable tasks, use AI freely. For predictions that require trust, insist on embedding auditable causal chains.
- AI 规范从哪里来?Where do AI norms come from?
- 问题:企业该如何制定 AI 使用规范,而不是让规定显得生硬和抵触?The problem: How should companies establish AI usage norms without making policies feel rigid and resisted?
- 关键想法:Charity Majors 与心理学家 Cat Hicks 对话后提出,AI 规范不是自上而下制定的,而是自下而上涌现的。核心方法是:先倾听,让员工表达不满和担忧,然后形成共享的价值观。Key idea: After a conversation with psychologist Cat Hicks, Charity Majors argues that AI norms are not imposed top-down but emerge bottom-up. The core method: listen first, let employees express their concerns and fears, then develop shared values together.
- 为什么成立:强制性的 AI 使用政策往往会引发抵触。但如果先让员工说出他们真正关心的问题 —— 什么让他们感到威胁、什么让他们兴奋 —— 规范就会从共同的体验中自然生长出来,而不是由管理层强加。Why it holds: Mandatory AI policies tend to provoke resistance. But if you start by letting employees voice what truly concerns them — what feels threatening, what feels exciting — norms will grow naturally from shared experience rather than being imposed by management.
- 能带走什么:制定 AI 规范时,从对话开始,而不是从规则开始。倾听比宣布更重要。What to take away: When establishing AI norms, start with conversation, not rules. Listening is more important than announcing.
数学之美Beauty of Mathematics
每日 Lean 谜题 · 周三Daily Lean Puzzle · Wednesday ★★★☆☆☆☆
素数有无穷多个There Are Infinitely Many Primes
对任意自然数 $n$,都存在一个素数 $p$,使得 $p \ge n$。(换句话说:素数有无穷多个。)For every natural number $n$, there is a prime $p$ with $p \ge n$. In other words: there are infinitely many primes.
Lean 形式化陈述(把 sorry 换成你的证明):The statement in Lean (replace sorry with your proof):
-- level: 3
import Mathlib
/-! Euclid: prove there is no largest prime — for every $n$ there is a prime $p \ge n$. (Idea: look at a prime factor of $n! + 1$.) -/
theorem infinitely_many_primes (n : ℕ) : ∃ p, n ≤ p ∧ p.Prime := by
sorry
在浏览器中挑战 ▶Try it in your browser ▶ 下载题目 .leanDownload puzzle .lean
提示Hint
考虑 $n! + 1$ 的一个素因子。任何不超过 $n$ 的素数都整除 $n!$,所以没有一个能整除 $n! + 1$。Look at a prime factor of $n! + 1$. Every prime not exceeding $n$ divides $n!$, so none of them can divide $n! + 1$.
答案与讲解Solution & walkthrough
- 第一步(构造):取 $m = n! + 1$。因为 $n! + 1 \ge 2$,$m$ 必有素因子,设 $p$ 是它最小的素因子。对应 Lean 里的 let p := Nat.minFac (n.factorial + 1):先用 Nat.factorial_pos 知道 $n! + 1 > 1$(omega 收尾),再用 Nat.minFac_prime 得出 $p$ 是素数。Step 1 (construction): Let $m = n! + 1$. Since $n! + 1 \ge 2$, $m$ has a prime factor; take $p$ to be its smallest one. In Lean this is let p := Nat.minFac (n.factorial + 1): Nat.factorial_pos gives $n! + 1 > 1$ (closed by omega), and Nat.minFac_prime makes $p$ prime.
- 第二步(反设):目标是证明 $n \le p$。Lean 里 refine ⟨p, ?_, hp⟩ 把目标换成 $n \le p$,接着 by_contra 反设 $p < n$。数学上就是经典的反证:如果 $p$ 比 $n$ 小,那么 $p$ 落在 $2, 3, \dots, n$ 之中。Step 2 (assume the opposite): We must show $n \le p$. In Lean, refine ⟨p, ?_, hp⟩ reduces the goal to $n \le p$, then by_contra assumes instead that $p < n$. Mathematically this is the classic move: if $p$ is smaller than $n$, then $p$ is one of $2, 3, \dots, n$.
- 第三步(导出矛盾):$p$ 在 $2, \dots, n$ 里,就必然整除 $n!$(对应 Nat.dvd_factorial);可它又整除 $n! + 1$(Nat.minFac_dvd)。一个数同时整除两个相邻的整数,就必须整除它们的差 $1$(Nat.dvd_add_right),而素数不小于 $2$,不可能整除 $1$(hp.one_lt.ne')。Step 3 (the contradiction): Being among $2, \dots, n$, $p$ divides $n!$ (this is Nat.dvd_factorial). But $p$ also divides $n! + 1$ (Nat.minFac_dvd). A number dividing two consecutive integers must divide their difference, $1$ (Nat.dvd_add_right) — impossible for a prime, which is at least $2$ (hp.one_lt.ne').
- 第四步(结论):矛盾说明反设不成立,所以 $n \le p$,而 $p$ 是素数。既然对任意 $n$ 都存在不小于它的素数,素数当然有无穷多个——Lean 证完,数学上也证完。Step 4 (conclusion): The contradiction kills the assumption, so $n \le p$ and $p$ is prime. Since such a prime exists above every $n$, there are infinitely many primes — done in Lean, done on paper.
-- level: 3
import Mathlib
/-! Euclid: prove there is no largest prime — for every $n$ there is a prime $p \ge n$. (Idea: look at a prime factor of $n! + 1$.) -/
theorem infinitely_many_primes (n : ℕ) : ∃ p, n ≤ p ∧ p.Prime := by
let p := Nat.minFac (n.factorial + 1)
have hn : n.factorial + 1 ≠ 1 := by
have := Nat.factorial_pos n
omega
have hp : p.Prime := Nat.minFac_prime hn
refine ⟨p, ?_, hp⟩
by_contra h
push_neg at h
have h1 : p ∣ n.factorial := Nat.dvd_factorial hp.pos h.le
have h2 : p ∣ n.factorial + 1 := Nat.minFac_dvd _
have h3 : p ∣ 1 := (Nat.dvd_add_right h1).mp h2
exact hp.one_lt.ne' (Nat.dvd_one.mp h3)
人的思考A human's take
Euclid's theorem(维基百科条目) — Wikipedia contributors(含逻辑学家 Torkel Franzén 的分析)
这篇条目纠正了一个流行误解:欧几里得《几何原本》IX.20 的原证法其实是直接证明——任给一张有限长的素数表,都能造出一个不在表上的素数;把它改写成反证法时,“假设这就是全部素数”这句话在证明里根本没被用到(逻辑学家 Torkel Franzén 指出这一点)。这个辨析恰好解释了谜题的形式:Lean 定理写成“对每个 $n$ 造出一个 $p \ge n$”,正是欧几里得的原始结构;我们的答案虽然用反证收尾,构造核心仍是“取 $n!+1$ 的最小素因子”。作者还数了一下:这条定理至少有 200 个不同的证明——数学界对它的偏爱,本身就是它之美的一部分。The article corrects a popular misreading: Euclid's original proof (Elements IX.20) is in fact a direct proof — given any finite list of primes, it produces a prime not on the list; when people rewrite it as a proof by contradiction, the assumption 'these are all the primes' is never actually used (a point credited to the logician Torkel Franzén). The nuance explains the shape of today's puzzle: the Lean theorem says 'for every $n$, produce a prime $p \ge n$', which is exactly Euclid's original structure; our answer closes by contradiction, but its constructive core is still 'take the smallest prime factor of $n! + 1$'. The article also counts more than 200 distinct proofs of the theorem — the mathematical world's fondness for it is part of its beauty.
AI 随想An AI's reflection
以下由 AI 撰写。Written by AI.
欧几里得的这个证明有一种特殊的美感:它不去构造“下一个素数”,而是证明“存在最大素数”这个假设会自我反驳。Euclid's proof has a peculiar beauty: it does not hunt for the next prime; it shows that the hypothesis of a largest prime refutes itself.
精妙之处在于一个极小的算术动作——阶乘加一。$n!$ 被所有不超过 $n$ 的素数整除,所以 $n!+1$ 与它们每一个都互素;它的最小素因子只能从 $n$ 的外面来。整个证明只用整除的两条基本性质,却推出了一个关于无穷的结论——用有限的手段触及无穷,这是初等数论最迷人的瞬间之一。The elegance lies in one tiny arithmetic move — factorial plus one. $n!$ is divisible by every prime up to $n$, so $n! + 1$ is coprime to all of them, and its smallest prime factor can only come from beyond $n$. The whole argument uses two basic properties of divisibility yet reaches a conclusion about infinity — finitary means touching the infinite, one of the most enchanting moments in elementary number theory.
这个论证还教人区分两类知识:知道素数无穷多不难,难的是像这样把它压缩进十行以内的推理。有意思的是,Lean 里 Nat.minFac 承担了“最小素因子一定存在”这个在纸面上常被一句带过的前提——形式化强迫我们诚实:数学证明里每一个“显然”,都是编译器要求填补的洞。The proof also teaches us to tell two kinds of knowledge apart: knowing that primes are endless is easy; compressing that knowledge into a reasoning this short is the hard part. Notably, Nat.minFac in the Lean proof carries the premise 'the smallest prime factor exists' — the very step paper proofs wave through in a phrase. Formalization forces honesty: every 'obviously' in a mathematical proof becomes a hole the compiler demands you fill.
作为对照,高斯的素数定理说的是另一回事:素数不仅无穷,而且稀疏得很:在 $n$ 附近,大约每 $\ln n$ 个整数才碰得到一个素数。欧几里得只用整除的性质就拿到了无穷性,把“到底有多少”留给了两千年以后的人。今天你在浏览器里补上那个 sorry,等于把这条最短的路重走一遍——需要的不是计算,而是敢在一个“显然”面前停下来,问一句:凭什么。For contrast, the prime number theorem says something else entirely: not just that primes are endless, but that they thin out — near $n$, only about one integer in $\ln n$ is prime. Euclid got infinitude from pure properties of divisibility and left 'how many' to people two millennia later. Filling in that sorry in your browser is walking the shortest road again — what it requires is not computation, but the willingness to stop in front of an 'obviously' and ask: by what right?
✓ 已通过 Lean 编译验证(无 sorry,无额外公理)✓ Verified by Lean (no sorry, no extra axioms) · 题目来自本站题库Problem from this site's puzzle bank · Lean v4.35.0-rc3 · Mathlib 3f6737d
- 外星真理Alien Truth
- 问题:什么是所有智能生命都会认同的真理?数学和物理之外的世界有没有普适真理?The problem: What truths would every intelligent life form agree on? Beyond mathematics and physics, are there universal truths about the world?
- 关键想法:Paul Graham 提出 "外星真理" 的概念 —— 不是数学或物理那种所有文明必然发现的真理,而是那些智能生命在思考世界时会自然达成的共识。比如奥卡姆剃刀原则、受控实验的价值。Key idea: Paul Graham introduces the concept of "alien truths" — not the kind of truths (like those in math or physics) that every civilization must discover, but insights that intelligent beings naturally converge on when observing the world. For example, the principle of Occam's razor, the value of controlled experiments.
- 为什么成立:数学真理(如 2+2=4)对所有文明都一样,但 "简约的解释通常比复杂的解释更好" 这种洞察,需要智能生命在观察世界时自然形成。这类真理的数量远比我们想象的多,而哲学本质上就是在寻找这些外星真理。Why it holds: Mathematical truths (like 2+2=4) are the same for every civilization. But the insight that "simpler explanations are usually better than complex ones" is something intelligent beings must naturally form through observation. There are far more such truths than we imagine, and philosophy is essentially the search for these alien truths.
- 能带走什么:阅读这类文章时,不是在学知识,而是在训练一种思维方式 —— 识别那些 "任何智能生命都会认同" 的洞察。这种训练本身就是一种思维健身。What to take away: When reading such essays, you are not learning facts — you are training a way of thinking: identifying insights that "any intelligent being would recognize." This training itself is mental fitness.
- 黑洞像钟一样回响,之后的事更加奇异Black holes ring like bells. What happens next is stranger.
- 问题:两个黑洞合并后,剩下的东西为什么如此简洁?The problem: After two black holes merge, why is the remaining object so simple?
- 关键想法:黑洞合并后会产生 "ringdown" —— 就像被敲响的钟在振动后逐渐静止。最终留下的只是一个简单的克尔黑洞,只由质量和自转两个参数描述。这就是 "无毛定理"。Key idea: After merging, a black hole undergoes "ringdown" — like a struck bell that vibrates and gradually settles. What remains is a simple Kerr black hole, described by only two parameters: mass and spin. This is the "no-hair theorem."
- 为什么成立:合并中有一部分质量以引力波的形式辐射出去,但结果黑洞的视界面积只增不减——这种单行道式的不可逆,正是熵的行为。视界之外的人再也看不到里面发生过什么,于是黑洞“忘记”了它的历史:无论它由什么坍缩形成、合并过几次,从外面看都只剩质量和自转。Dyer 的深意是,这种“遗忘”不只是黑洞的特性——把视角拉远看整个世界,复杂的历史同样被抹平,只剩下熵增的方向。Why it holds: some mass is radiated away as gravitational waves during the merger, yet the area of the resulting horizon never shrinks — this one-way irreversibility is exactly how entropy behaves. Nothing outside the horizon can ever see what went on inside, so the black hole 'forgets' its history: no matter what collapsed to form it or how many times it merged, from the outside only mass and spin remain. Dyer's deeper point is that this forgetting is not special to black holes — zoom out on the world at large, and complex histories get smoothed away the same way, leaving only the arrow of entropy.
- 能带走什么:黑洞的 "遗忘" 提醒我们,宇宙倾向于从复杂走向简单。熵不仅描述了混乱的增加,也描述了信息的丢失。理解这一点,能让我们用全新的视角看待时间与变化。What to take away: A black hole's "forgetfulness" reminds us that the universe tends from complexity toward simplicity. Entropy describes not just increasing disorder but also information loss. Understanding this gives us a radically new perspective on time and change.
- 开放的问题,开放的数学Open problems, open mathematics
- 问题:AI 可以快速生成证明,数学界的开放共享文化会因此改变吗?The problem: AI can rapidly generate proofs. Will the open and sharing culture of mathematics change?
- 关键想法:Antonio Auffinger 观察到,AI 让 "快速发表" 变得容易,这可能导致数学家不再分享工作中的想法,转而追求局部优化。他呼吁数学界重新审视激励机制,保持开放的学术文化。Key idea: Antonio Auffinger observes that AI makes "fast publication" easy, which may discourage mathematicians from sharing ideas in progress and push them toward local optimization. He calls on the math community to reexamine incentive structures and preserve open academic culture.
- 为什么成立:数学的进步依赖于想法的早期共享——在证明完整之前分享思路,让同行指出漏洞或提供新方向。如果 AI 让发表变得廉价,而评价机制不变,数学家就会有动机藏起想法、抢先发表。生物学界在人类基因组计划中已经成功实践过类似的数据共享原则(百慕大原则)。Why it holds: Mathematical progress depends on sharing ideas early — circulating them before the proof is complete, so peers can point out gaps or suggest new directions. If AI makes publication cheap while evaluation systems stay the same, mathematicians will have an incentive to hide ideas and rush to publish. The biology community already practiced a similar data-sharing principle successfully during the Human Genome Project (the Bermuda Principles).
- 能带走什么:AI 不会取代数学的开放精神,但它会考验这种精神。社群需要主动设计激励机制,让开放成为理性选择,而不是道德牺牲。What to take away: AI will not replace the open spirit of mathematics, but it will test that spirit. The community needs to actively design incentive structures that make openness a rational choice, not a moral sacrifice.
- 给数学画像:一个 87 岁老分形的可视化实验Picturing Mathematics: an experiment in visualizing an 87-year-old fractal
- 问题:图像对数学有什么用?作者 James Propp 自称“低技术数学”信徒——更信任纸笔能亲手验证的东西,但也清楚自己想象力的边界:有些结构算得出却想不到,有些想得到却看不清。The problem: what are pictures good for in mathematics? James Propp calls himself a believer in 'low-tech math' — he trusts what he can check with pen and paper — yet he is equally aware of the limits of his imagination: some structures he can compute but not picture, and some he can conceive but not see.
- 关键想法:他在 2025 年的一次网络研讨会里做了个实验:用五分钟“推销”一个从没被令人满意地可视化过的数学对象(Ford 球),看有没有更懂图形的人愿意接手。答案是有人接了——Roice Nelson 和他一起把动画做了出来。文章讲的正是这种把“视觉直觉”当成数学工作的过程。Key idea: in a 2025 webinar he ran an experiment: spend five minutes pitching a mathematical object that nobody has ever visualized in a fully satisfying way — the Ford spheres — and see whether someone more skilled with graphics would take it on. Someone did (Roice Nelson), and the two of them built the animation together. The essay is about treating visual intuition as mathematical work.
- 为什么成立:Ford 球来自 Lester Ford 1938 年一篇叫《分数》的文章——比曼德尔布罗特提出“fractal”这个词还早 37 年。它的二维版本 Ford 圆是有理数的几何替身:分子分母越大,圆越小;圆彼此相切的方式编码着数论事实,还让“有理数处处都在、却又几乎不存在”这个关于可数稠密集的反直觉性质变得肉眼可见。把三维截面一帧帧看下去,眼睛会先于证明发现新结构——Propp 管它们叫“光环”和“日冕弧”。Why it holds: Ford spheres come from a 1938 article by Lester Ford called 'Fractions' — 37 years before Mandelbrot coined the word 'fractal'. Their two-dimensional cousin, the Ford circles, are geometric stand-ins for the rational numbers: the bigger the numerator and denominator, the tinier the circle, and the tangency pattern encodes number-theoretic facts. They also make a counterintuitive property visible: the rationals are everywhere and almost nowhere at once. Scrutinizing cross-sections frame by frame, the eye finds structure — 'halos' and 'solar arches' — ahead of any proof.
- 能带走什么:图像不能代替证明,但它能替你提出下一个问题。Propp 坦白:他还不知道怎么“盘问”这张图,让像素告诉他光环对应的精确数学——而这正是研究开始的姿势。对程序员来说,这和先画图、再对着图找 bug 或规律是同一种直觉工作。What to take away: pictures cannot replace proofs, but they can pose the next question for you. Propp admits he does not yet know how to 'interrogate' the image to say exactly what the halos correspond to — and that is precisely the posture in which research begins. For programmers, it rhymes with plotting data first and finding the bug or the pattern in the plot.
程序员随想Programmer Reflections
- 把想法变成文字Putting Ideas into Words
- 问题:为什么写作是检验想法是否清晰的最好方法?The problem: Why is writing the best test for whether your ideas are clear?
- 关键想法:Paul Graham 认为,写作不是思考的结束,而是思考的一部分。把想法写出来这个行为本身会改变你的想法,甚至产生新想法。Key idea: Paul Graham argues that writing is not the end of thinking, but part of thinking. The act of writing down ideas changes them and can even generate new ones.
- 为什么成立:思维是模糊的、网状的,而语言是线性的、有结构的。当你试图把想法写成句子时,你被迫做出选择:哪个概念先出现?因果关系是什么?这种 forcing function 会让隐藏的假设浮出水面。Graham 提出了 "陌生人测试":假装你的读者完全不知道你在想什么,如果你的解释让他困惑,说明你的想法还不够清晰。Why it holds: Thought is fuzzy and web-like, while language is linear and structured. When you try to write ideas as sentences, you are forced to make choices: which concept comes first? What is the causal relationship? This forcing function brings hidden assumptions to the surface. Graham proposes the "stranger test": pretend your reader knows absolutely nothing about what you are thinking. If your explanation confuses them, your idea is not yet clear enough.
- 能带走什么:不要等想法完全清晰了再写。写本身就是让想法变清晰的过程。如果你写不出来,不是因为你还没准备好,而是因为你的想法本身就还不够清楚。What to take away: Do not wait for ideas to become completely clear before writing. Writing itself is the process that makes ideas clear. If you cannot write it down, it is not because you are not ready — it is because your idea is not yet clear.
- 为什么 Lisp 的括号难读?What makes Lisp difficult to read?
- 问题:为什么 LISP 的括号对很多人来说难以阅读?The problem: Why are LISP parentheses difficult for many people to read?
- 关键想法:Paul Taylor 从认知科学角度解释:人脑在读取括号时,依赖格式塔感知原则中的 "近旁性"。C 风格的括号 { 和 } 在视觉上更靠近,更容易形成视觉分组;而 Lisp 的 ( 和 ) 之间有更多内容,认知负担更大。Key idea: Paul Taylor explains from a cognitive science perspective: when reading parentheses, the brain relies on the Gestalt principle of "proximity." C-style braces { and } are visually closer together, forming easier visual groups; while Lisp's ( and ) have more content between them, creating greater cognitive load.
- 为什么成立:这不是品味问题,而是人类视觉系统的物理限制。括号之间的距离影响了大脑对代码结构的快速解析。这个分析让 Lisp 社区的长期争论有了一个新的视角。Why it holds: This is not a matter of taste but of the physical limits of the human visual system. The distance between parentheses affects how quickly the brain can parse code structure. This analysis gives the long-standing Lisp community debate a new perspective.
- 能带走什么:编程语言的 "可读性" 不只是审美问题,它根植于人类的认知架构。理解这一点,可以帮助我们更好地设计语言、工具和代码风格。What to take away: The "readability" of a programming language is not just an aesthetic issue — it is rooted in human cognitive architecture. Understanding this can help us design better languages, tools, and coding styles.
- 系统的目的不是它所做的事The Purpose Isn't What It Does
- 问题:为什么我们常常误以为系统的目的是它所实际做的事?The problem: Why do we often mistake what a system actually does for what it is supposed to do?
- 关键想法:Tim Bray 批判了 POSIWID(The Purpose Of A System Is What It Does)这个观点。他指出系统通常有多个目的,意图的目的和实际的目的往往不同,新系统还会产生未预见的后果。Key idea: Tim Bray critiques the idea of POSIWID (The Purpose Of A System Is What It Does). He points out that systems typically have multiple purposes, the intended purpose and the actual purpose often differ, and new systems produce unintended consequences.
- 为什么成立:一个社交媒体的目的是 "连接人们",但它实际做的可能是 "最大化用户停留时间"。一个教育平台的目的是 "传授知识",但它实际做的可能是 "收集用户数据"。当我们只关注系统实际做了什么,而忽略设计意图时,就会失去对系统的理解和控制。Why it holds: A social media platform's purpose is to "connect people," but what it actually does may be "maximizing user engagement time." An education platform's purpose is to "teach knowledge," but what it actually does may be "collecting user data." When we only focus on what a system actually does and ignore its design intent, we lose understanding and control.
- 能带走什么:评估任何系统时,问两个问题:它被设计来做什么?它实际上在做什么?这两个答案之间的差距,往往是问题所在。What to take away: When evaluating any system, ask two questions: What was it designed to do? What does it actually do? The gap between these two answers is often where the real problem lies.
哲学与生活Philosophy and Life
- 父亲身份是诗意的Paternity is poetical
- 问题:成为父亲会损害创造力吗?这是一个被广泛接受但不加质疑的说法。The problem: Does becoming a father damage creativity? This is a widely accepted claim that goes unchallenged.
- 关键想法:Daniel Swift 通过文学史数据反驳了这个陈词滥调。柯勒律治成为伟大诗人的转折点恰好是成为父亲之后;80% 的诺贝尔文学奖获得者是父亲。成为父亲不是创造力的终结,而往往是某种诗意的觉醒。Key idea: Daniel Swift refutes this cliché using literary history data. Coleridge's turning point as a great poet came after becoming a father; 80% of Nobel Prize winners in Literature were fathers. Becoming a father is not the end of creativity, but often a poetic awakening.
- 为什么成立:这个说法之所以流行,是因为我们容易把 "有了家庭责任" 等同于 "没有时间写作"。但 Swift 展示了更复杂的真相:父亲身份改变了创作的主题和深度,它带来了一种新的感知方式 —— 对时间、传承、责任的感知。Why it holds: This claim is popular because we easily equate "having family responsibilities" with "no time to write." But Swift reveals a more complex truth: fatherhood changes the themes and depth of creation, bringing a new way of perceiving — a perception of time, legacy, and responsibility.
- 能带走什么:不要轻易接受关于人生重大转变的简单叙事。数据和文学史告诉我们,成为父亲这件事,远比 "拖累创造力" 要复杂和深刻。What to take away: Do not easily accept simple narratives about major life transitions. Data and literary history tell us that becoming a father is far more complex and profound than simply "hindering creativity."
- 土地伦理在一片草原上开花Where a land ethic blooms
- 问题:土地伦理是抽象的哲学概念,还是可以在日常决策中实践的?The problem: Is land ethics an abstract philosophical concept, or can it be practiced in everyday decisions?
- 关键想法:Craig Maier 讲述了一个父亲在坡地上做选择的故事:是种玉米(短期收益)还是保留草地(长期生态价值)?这个看似简单的农业决策,实际上是土地伦理的实践。Key idea: Craig Maier tells the story of a father making a choice on a slope: plant corn (short-term profit) or keep the prairie (long-term ecological value)? This seemingly simple farming decision is actually land ethics in practice.
- 为什么成立:土地伦理不是书斋里的哲学,而是活在家庭和社区中的实践智慧。一个父亲在坡地上的选择,影响的不只是今年的收成,而是土壤、水源、以及下一代能看到怎样的土地。这种思考跨越了代际,也超越了经济账本。Why it holds: Land ethics is not philosophy in an armchair, but practical wisdom lived within families and communities. A father's choice on a slope affects not just this year's harvest, but the soil, the water, and what the next generation will see. This thinking spans generations and transcends economic accounts.
- 能带走什么:伦理不只存在于宏大的哲学论述中,它也存在于我们每天做的微小选择里。下次面对类似决策时,问问自己:十年后,这块土地会是什么样子?What to take away: Ethics exists not only in grand philosophical treatises, but in the small choices we make every day. Next time you face a similar decision, ask yourself: What will this land look like in ten years?
工具角Tools Corner
- 用药房十字灯 Rickroll 你Rickrolling with a pharmacy cross
- 问题:法国药房门前的绿色十字灯,能播放视频吗?The problem: Can the green cross LED sign in front of French pharmacies play videos?
- 关键想法:Hugo Arnal 逆向工程了法国药房绿色十字 LED 灯的协议,然后用它播放了 Rick Astley 的 "Never Gonna Give You Up"。整个项目从硬件逆向到视频播放器,全部开源。Key idea: Hugo Arnal reverse-engineered the protocol of the French pharmacy green cross LED sign, then used it to play Rick Astley's "Never Gonna Give You Up." The entire project, from hardware reverse-engineering to video player, is fully open source.
- 为什么酷:这不是一个普通的 DIY 项目,而是把街头元素变成了艺术表达。24×24 像素的单色显示屏,限制极大,但作者找到了在限制中创作的方法。Why it's cool: This is not an ordinary DIY project — it turns a street element into artistic expression. A 24×24 pixel monochrome display with severe limitations, yet the author found a way to create within those constraints.
- 能带走什么:最好的创意项目往往来自对日常事物的重新想象。你不需要高端设备,只需要好奇心和一点逆向工程的勇气。What to take away: The best creative projects often come from reimagining everyday objects. You do not need high-end equipment, only curiosity and a bit of courage to reverse-engineer.