读书笔记 · 约 22,079 字 · 预计推送 3 部分

演讲精读：Hamming《You and Your Research》——如何做出一流研究，兼论对夜间研究系统选题方式的启发 未开始

目录

题记与说明

一、七维提取

1. 核心问题：这份工作在解决什么问题

Kaiser 转录稿开头一句话就是全篇的问题陈述：

"This talk centered on Hamming's observations and research on the question 'Why do so few scientists make significant contributions and so many are forgotten in the long run?'"
（为什么只有极少数科学家做出重要贡献，而大多数最终被遗忘？）

更准确地说是三层递进的问题：

2. 主要主张（核心论点清单）

#主张原文关键句
1运气被高估：一流成果会"重复地"出自同一批人"Consider Einstein. Note how many different things he did that were good. Was it all luck? Wasn't it a little too repetitive?" / 引 Pasteur："Luck favors the prepared mind." / "The particular thing you do is luck, but that you do something is not."
2天赋不是瓶颈"Most of you in this room probably have more than enough brains to do first-class work. But great work is something else than mere brains."（Pfann 区熔提纯、Clogston 电缆两个反例：起手"数学不好、不善言辞"的人靠一个成功获得勇气后连奖项拿遍）
3勇气是第一性格变量"Courage is one of the things that Shannon had supremely." Shannon 对随机码问出不可问题："What would the average random code do?" —— "Who but a man of infinite courage could have dared to think those thoughts?"
4知识与产出是复利"Knowledge and productivity are like compound interest... Given two people with exactly the same ability, the one person who manages day in and day out to get in one more hour of thinking will be tremendously more productive over a lifetime."（书章版补了量化感："I do not know what compound interest rate to assign, but it must be well over 6%—one extra hour per day over a lifetime will much more than double the total output."）
5但蛮力无效，要"聪明地用力""drive, misapplied, doesn't get you anywhere... Just hard work is not enough - it must be applied sensibly."（他观察 Bell Labs 里比他更努力却没产出的朋友）
6容忍模糊性是伟大科学家的标志"Great scientists tolerate ambiguity very well. They believe the theory enough to go ahead; they doubt it enough to notice the errors and faults so they can step forward and create the new replacement theory. If you believe too much you'll never notice the flaws; if you doubt too much you won't get started."（Darwin 自传：强迫自己记下每一条反例，否则它们会从记忆中消失）
7创造力来自潜意识，所以要"饿着潜意识"逼它干活"creativity comes out of your subconscious... If you are deeply immersed and committed to a topic, day after day after day, your subconscious has nothing to do but work on your problem... Keep your subconscious starved so it has to work on your problem, so you can sleep peacefully and get the answer in the morning, free."
8⭐ 不做重要问题就做不出重要工作"If you do not work on an important problem, it's unlikely you'll do important work. It's perfectly obvious." / "The average scientist... spends almost all his time working on problems which he believes will not be important and he also doesn't believe that they will lead to important problems."
9⭐ 重要性的定义 = 重要后果 × 可攻击性"It's not the consequence that makes a problem important, it is that you have a reasonable attack. That is what makes a problem important."（反例：时间旅行、瞬移、反重力——"guaranteed a Nobel Prize"，但没人有攻击手段，所以不是"重要问题"）
10⭐ 伟大科学家脑内常备 10–20 个重要问题，新线索一出现就全押"Most great scientists know many important problems. They have something between 10 and 20 important problems for which they are looking for an attack. And when they see a new idea come up, one hears them say 'Well that bears on this problem.' They drop all the other things and get after it."
11⭐ 每周留 10% 的"Great Thoughts Time" 专门想大问题"at some urging from John Tukey and others, I finally adopted what I called 'Great Thoughts Time.' When I went to lunch Friday noon, I would only discuss great thoughts after that." / "I committed 10% of my time trying to understand the bigger problems in the field." 由此预判："nine out of ten experiments were done in the lab and one in ten on the computer... it would be reversed"——后来被证明正确。
12橡子与橡树：在可能出事的地方保持活跃"You can't always know exactly where to be, but you can keep active in places where something might happen... you can stand on a mountain top where lightning strikes; you don't have to hide in the valley where you're safe."
13名声是毒药：别只攻大问题，要继续种小橡子"After information theory, what do you do for an encore? The great scientists often make this error. They fail to continue to plant the little acorns from which the mighty oak trees grow." / "The Institute for Advanced Study in Princeton... has ruined more good scientists than any institution has created."
14⭐ 开着门工作：短期少产出，长期做对题"if you have the door to your office closed, you get more work done today and tomorrow... But 10 years later somehow you don't know quite what problems are worth working on; all the hard work you do is sort of tangential in importance. He who works with the door open gets all kinds of interruptions, but he also occasionally gets clues as to what the world is and what might be important."（他明说因果未证："I cannot prove the cause and effect sequence"，但相关性稳定）
15⭐ 不做孤立问题，只做一类问题的代表"I made the resolution that I would never again solve an isolated problem except as characteristic of a class." / "The business of abstraction frequently makes things simple." / "You should do your job in such a fashion that others can build on top of it... The essence of science is cumulative."
16⭐ 做完还不够，必须会"卖""it is not sufficient to do a job, you have to sell it... If they don't stop and read it, you won't get credit." 三件套：写好文章、正式报告、非正式场合发言。书章版量化："at least 50% of the time must go for the presentation."
17教育你的老板：设定自己的截止日期Schelkunoff 事件结尾："Ever after that Schelkunoff said, 'You set your deadlines; you can change them.'"
18别为小事和系统打仗，但要会用系统"If you will learn to work with the system, you can go as far as the system will support you." / "Which do you want to be? The person who changes the system or the person who does first-class science?" / "The appearance of conforming gets you a long way."
19别给自己找借口"Don't try an alibi. Don't try and kid yourself. You can tell other people all the alibis you want... But to yourself try to be honest."
20值不值？价值在挣扎本身"the truth is, the value is in the struggle more than it is in the result... The success and fame are sort of dividends."

3. 方法：他怎么论证的

这不是论文，是一次"经验研究式"的演讲，方法论相当自觉：

4. 数据与证据（含数字）

证据出处强度评估
伟大科学家脑内常备 10–20 个重要问题转录稿（"They have something between 10 and 20 important problems"）个人访谈归纳，无样本量
每周 10% 时间做 Great Thoughts转录稿（"I committed 10% of my time"）自我实验报告
汇报呈现应占 ≥50% 时间书章版 Q&A（"at least 50% of the time must go for the presentation. It's a big, big number."）自我估计
知识复利率"远超 6%"；每天多思考 1 小时 → 终生产出翻倍以上书章版（"it must be well over 6%—one extra hour per day over a lifetime will much more than double the total output"）类比量化，无实测
实验地点从实验室搬到计算机：9:1 → 1:9 预判转录稿个人预言自称验证，未给数据
知识每 17 年翻倍；340 年后"百万个学科"不可持续转录稿 Q&A粗估外推
每约 7 年换一次子领域；他自己"numerical analysis → hardware → software"转录稿 Q&A经验法则
Dave McCall 对照组：一桌化学家中唯一响应提问者，数月后升部门主管、后入 NAE转录稿个案，n=1 对照
Berkeley 裂变数据"已采集未归约"而错失发现转录稿（他自注"can't vouch for the truth of it"）转述故事，他自己标注不可考
Pfann（区熔提纯）"collected all the prizes in the field"；实验室旧址成为国家纪念地书章版补充个案
Shannon 年龄≈Tukey 的 surprise → Bode"复利"劝学转录稿轶事（注意：演讲版只说"John was a genius and I clearly was not"，书章版才有 6% 数字）
转录稿现场：Bellcore 约 200 人听众，演讲约 50 分钟 + Q&AKaiser 转录稿引言（"an overflow audience of some 200 Bellcore staff members"）事实性

证据总评：这是 n=几十的、以成功者为中心的个案归纳集，含大量自我报告与转述轶事（Hamming 自己也承认："I was sitting in an airport talking to a friend... I can't vouch for the truth of it"；以及承认样本偏倚 "It's a biased sample, but I still think it is worth the struggle"）。它的力量在于内部一致性：每条主张都有机制解释（潜意识、复利、开放门的信息价值），且互相咬合成一个系统，而不是一堆鸡汤格言。但它不是统计学意义上的证据，主张 8/9/10/11（选题四件套）是其中最核心也最难验证的部分。

5. 结论：一张"个人研究操作系统"的清单

把 20 条主张压缩成一个可执行的闭环，Hamming 的结论其实是一套个人研究操作系统：

  1. 自我设定：先允许自己想做出一流工作（"Yes, I would like to do first-class work"——社会的谦虚规范是第一道障碍）；
  2. 题库管理：长期维护 10–20 个"重要且可攻击"的问题，用每周 10% 的固定时间审视它、扩大它（Great Thoughts Time）；
  3. 火力匹配：新想法/新工具/新论文出现时，对照题库——"bears on this problem" 则 drop everything；
  4. 变形术：孤立问题泛化为一类问题；缺陷换个视角变资产（"It ain't what you do, it's the way that you do it"）；做别人能在其上盖楼的累积性工作；
  5. 信息面：开门工作、挑"能激发回响的人"（"sound absorbers" 是诅咒）、读书为找问题而非找答案；
  6. 销售面：写作/正式演讲/即席发言三技能，呈现占时 ≥50%；
  7. 自我管理：饿着潜意识、公开立 flag、教育老板、不为小事与系统开战、每 7 年换邻域、不给自己找借口；
  8. 价值观：价值在挣扎本身，名声是红利——"Go forth, then, and do great work!"

6. 局限（作者承认的 + 推断的）

7. 对我的启发：结合夜间研究系统的选题方式

先陈述观察到的系统现状（均来自本仓库只读材料）：

对照 Hamming，得到七条启发（按可操作性排序）：

启发 1：backlog 缺一个"10–20 重要问题库"，应从"课题队列"升级为"问题库×课题队列"两层结构。
Hamming 的核心机制是方向相反的两条流水线：伟大科学家是 problem → idea（常备 10–20 个问题，新想法出现时问"this bears on my problem"），而夜间系统目前只有 idea → note（先排队课题，再做笔记）。补法很轻：建一个 backlog/problems.md（或 reading.md 头部注释区），维护 ≤20 个"重要且可攻击"的问题；每条新 backlog 课题入队时强制回答"这个问题服务于 problems.md 里的哪一条"——答不上来的课题不是不能做，但应该自知是"acorn（闲子）"而不是"重要问题"。matrix.csv 的"对我的启发"列应定期回写 problems.md，形成"读过的东西改变问题清单"的闭环——否则复利只记账不滚存。

启发 2：:: 要求字段其实就是 Hamming 的"reasonable attack"，但"重要后果"维度完全缺失，选题应过两道检验。
现状抽样显示 :: 字段普遍写的是攻击手段（"以 Windows 单机可落地为准""给配置建议"），这非常好地实现了 Hamming 的 attack 检验。但没有任何机制问后果："这条做完会改变我的哪个决策、产出哪个工件？半年后我还会在乎吗？" 建议的选题口诀：后果（会改变什么决策）× 攻击（:: 字段）双真才入队；只有攻击没有后果的课题是"精致的手工活"（Hamming 嘲讽的 "another decimal place"）。

启发 3：给清单留 10% 的"Great Thoughts 条目"——夜间系统最缺的就是 Friday afternoon。
Hamming 的 10% Great Thoughts Time 是一个固定仪式，而夜间系统的全部轮次都在执行既定课题，没有任何一轮在问"我领域里最重要的问题是什么、我的 field 正走向哪里"。可落地的最小改法：在 reading.md/research-topics.md 里加入一类元课题，例如"回顾 problems.md：哪些问题这半年没人碰？夜读清单的邻域分布是否失衡？"（每 10 条课题配 1 条元课题），让系统自己定期抬头看路。Hamming 靠这个仪式在 1980 年代预判了计算/实验 9:1 反转——夜间系统同款问题应该是"agent 时代，'读论文→做笔记'这个动作本身会被什么取代？"

启发 4：邻域过窄，违反 acorn 原则——每 5 条留 1 条"邻域外的小橡子"。
抽样 53 条课题几乎全部落在 LLM/agent/检索/工具链一个邻域。Hamming 的警告是双向的：名人只攻大问题会废（"fail to continue to plant the little acorns"），而在一个领域耗尽原创性也会废（"every seven years make a significant shift... you tend to use up your ideas"）。夜间系统的对冲：固定比例（≈20%）条目是"有可能长成橡树的橡子"——不在当前舒适邻域、但满足"后果×攻击"的小课题。跨邻域阅读正是"lightning 高地"：AI 工程的下一个重要问题大概率不是从 AI 工程文献里读出来的。

启发 5：夜间系统是"关门模式"，需要补一条 open-door 的输入通道。
Hamming 的开门/关门观察精确命中清单驱动系统的盲区：关门的系统"今天更高效，十年后不知道什么问题值得做"。当前只有 kb 流水线有"发现"型开放式条目（外部信号可插队），reading/research 两条流水线完全靠预定队列。轻量改法：research 清单允许一种 - [ ] 发现 :: ... 条目类型，且排队不按严格 FIFO（README 已实现失败降权，可复用同一机制给"发现"条目加权），让偶发的开放门线索不至于排到三个月后。

启发 6：笔记是弹药不是成品——"卖"的环节缺失，精读产量应有一部分回流为可发布的合成品。
Hamming："not sufficient to do a job, you have to sell it"，且呈现应占 ≥50% 时间。夜间系统的产出（七维笔记 + matrix.csv）停在"个人弹药库"层；kb 流水线反而有 digest.md 的"卖"层，knowledge 流水线没有对应物。他的"Hamming's Method"故事展示了 selling 的正确姿势：把一次作业重构成"第一次用数字机在模拟机主场击败模拟机"的宣言——同样的结果，换了问题表述就从琐碎变成重要。对系统的含义：backlog 里应存在一类"合成课题"（把已有 3–5 条笔记组装成一篇可发布的对比文章/决策指南），并让成品链接回源笔记——"The essence of science is cumulative"的私有版：让六个月后的自己能在笔记上盖楼，而不是重新精读一遍。

启发 7：直接吸收 Hamming 的三个"个人 API"，成本近乎为零。

一个诚实的反向提醒：Hamming 的隐含代价函数（牺牲家庭时间、常年应激、"nice guys end last"）不应照单全收进夜间系统。夜间系统的存在理由恰恰是把深工作放进无人值守时段，保护日间生活——这是对 Hamming 式偏执的工程化缓和，而不是向它投降。他的方法可以全收，他的价值观（工作压倒一切）不必全收。

二、重要原文引述汇编（中英对照，均为 1986 Kaiser 转录稿，另注明者除外）

  1. 方法论起点
"I saw I was a stooge... And to put the thing bluntly, I was envious. I wanted to know why they were so different from me. ... I became very interested in the difference between those who do and those who might have done."
  1. 运气论（全篇最重要的一段辩证）
"There is indeed an element of luck, and no, there isn't. ... The particular thing you do is luck, but that you do something is not." ——引 Pasteur："Luck favors the prepared mind." 引 Newton："If others would think as hard as I did, then they would get similar results."
  1. 重要问题的定义（对个人选题系统最有操作价值的一句）
"It's not the consequence that makes a problem important, it is that you have a reasonable attack. That is what makes a problem important."
  1. 题库机制
"Most great scientists know many important problems. They have something between 10 and 20 important problems for which they are looking for an attack. And when they see a new idea come up, one hears them say 'Well that bears on this problem.' They drop all the other things and get after it."
  1. 伟大思想时间
"I finally adopted what I called 'Great Thoughts Time.' ... I committed 10% of my time trying to understand the bigger problems in the field, i.e. what was and what was not important. I found in the early days I had believed 'this' and yet had spent all week marching in 'that' direction. It was kind of foolish. If I really believe the action is over there, why do I march in this direction? I either had to change my goal or change what I did. It's that easy."
  1. 开门/关门
"if you have the door to your office closed, you get more work done today and tomorrow, and you are more productive than most. But 10 years later somehow you don't know quite what problems are worth working on; all the hard work you do is sort of tangential in importance."
  1. 不做孤立问题（转录稿）
"I made the resolution that I would never again solve an isolated problem except as characteristic of a class."

书章版补：按 drunkard's walk，无愿景者的路径长度 ∝ √N，有愿景者 ∝ N——"the difference between having a vision and not having a vision is almost everything."

  1. 销售
"it is not sufficient to do a job, you have to sell it. 'Selling' to a scientist is an awkward thing to do... But the fact is everyone is busy with their own work. You must present it so well that they will set aside what they are doing."
  1. 失败四联因（演讲总结）
"they don't work on important problems, they don't become emotionally involved, they don't try and change what is difficult to some other situation which is easily done but is still important, and they keep giving themselves alibis why they don't."
  1. 价值观
"the truth is, the value is in the struggle more than it is in the result. The struggle to make something of yourself seems to be worthwhile in itself. The success and fame are sort of dividends."

结尾："Go forth, then, and do great work!"

  1. 阅读方式（Q&A，对知识库系统最相关）
"If you read all the time what other people have done you will think the way they thought. ... You need to keep up more to find out what the problems are than to read to find the solutions. ... it is not the amount, it is the way you read that counts."
  1. 七年之变（Q&A）
"Somewhere around every seven years make a significant, if not complete, shift in your field... before you use up all the old ones."
  1. Wiener 之 50%（Gwern《On Really Trying》所引，机制层面补强 Hamming）
"Once a scientist attacks a problem which he knows to have an answer, his entire attitude is changed. He is already some 50% of his way toward that answer." —— Norbert Wiener, *The Human Use of Human Beings*, p.124
  1. Narayanan 的分阶段修正
"Note that the last point is at odds with advice given to more experienced researchers. Richard Hamming, in a famous talk titled 'You and your research', advised researchers to pick the most important problem that they have a shot at solving."（原文随后给出新手替代建议：选能完成的小问题学流程）

三、版本考据备忘

*精读完成于 2026-11-05，夜间研究系统 nightly-knowledge 流水线。本笔记仅新增本文件，未改动 matrix.csv、backlog、nightly-log.md 及任何其他条目笔记。*

← 资料库今日播报 →