写作方法20211729.com
一句话、一段话、一页纸：把复杂构想压缩成三个版本
The Network State in one sentence, one paragraph, and one page
导读
这是 Balaji 用「一句话、一段、一页」压缩推销「网络国家」构想的文本，HN 评论区几乎一边倒地拆解它。对写作者来说它是一份反面教材：看一个宏大宣言如何因为内部自相矛盾、重新发明旧制度、只思辨不实践、空泛断言缺证据而被逐条击穿。
可以直接用的写作要点
- 01
提出「全新」方案前，先交代它与既有制度（君主制、陪审团、俱乐部、非营利组织）的差别，否则读者只会认为你在重新发明轮子。
- 02
核心概念之间不能互相打架：一边要「明确的领袖」，一边要「去中心化加密」，读者一眼就能看出裂缝并据此否定全文。
- 03
与其写一本书论证「如果……会怎样」，不如先做出来再写；主张要落在可验证的实践上，纯思辨的文本在读者眼里近乎没有价值。
- 04
宏大断言必须配引用和具体证据，只堆口号式真理（「每代都有乌托邦」式概括）反而会暴露论证的空洞。
HN 评论精选
提案在重新发明君主制和陪审团，却没说明为什么更好。
这个提案犯了大多数技术提案的通病：重新发明已有的东西。「首先围绕一位明确的领袖建立社交网络」——这不就是君主制吗？「合法性来自人们是否自愿加入」——不喜欢就自己再造一个 Facebook？懂技术的人总该明白网络效应和先发优势。既然整个构想的核心是去中心化区块链，为什么还要中心化？为什么不干脆去中心化裁决（哦，我发明了陪审团）。真希望人们去发挥技术的长处，而不是把已有的东西重做得更糟；地方治理确实不好，但恐怕这些人根本不了解问题所在，所以从不花时间想怎么解决。
Traster查看英文原文
I think this suffers from the same problem that most technology based proposals suffer from: it reinvents something that already exist. >First, form a social network around a clear leader, a proven centralized mechanism for dispute resolution in a decentralized world. Yay! Monarchy! >Since anyone can found a network state, just like anyone can found a tech company or cryptocurrency, the legitimacy of this leader comes from whether people have opted in to follow them. And if you don't like it, go off an invent facebook yourself! If you're tech-literate surely you understand network effects and first mover advantage. Hell, why are you even centralizing at all, if the whole thing is centered around a decentralized blockchain? Why not decentralize adjudication (oh shit, I invented trial by a jury of your piers). I really wish people would focus on exploiting what technology does well, rather than focusing on re-implementing what we've already done, but worse. Local governance isn't great, but I suspect no one involved in this understands anything about the problems, hence why they spend no time thinking about how to solve them.
前提（中心化权威）与手段（去中心化技术）自相矛盾。
如果你需要一个「明确的领袖」、一个可信权威，那还要比特币式的加密货币和去中心化金融、「Web3」这些无政府技术干什么？直接用中心化数据库就行了，砍掉极其浪费的共识机制，总统甚至可以兼任数据库管理员。21 世纪技术民族主义的某些想法很棒，但硬塞技术时髦就站不住脚了。
pphysch查看英文原文
If you have a "clear leader", that is, a trusted authority, why do you need a Bitcoin-esque cryptocurrency and other anarchic blockchain/DeFi/"Web3" technology? Just use a central database, man. Cut out the tremendously wasteful PoX nonsense. The President could even be the DBA-in-Chief. Some of these ideas of 21st century techno-nationalism are great, but falls short when trying to shoehorn $techfad into it.
纯思辨不如实践，且未与既有路径区分开。
评论它、哪怕只是批评它，有价值吗？如果真把它当回事，你就不会写书而是直接去做（做完也许再写书）。可这篇文章读起来完全像初级程序员写自己理想中的框架或语言：纯属思辨，因此没有价值。当然，也许作者是想先抛出想法再动手，那我就要问：加密货币跟这到底有什么关系？它是一堆别扭的包袱，很糟糕地中介着一个群体在更大群体中的财务。另外，这和「先成立一个俱乐部、注册成非营利组织、抱团投票获得影响力，然后再推进下一步」有什么不同？历史上有大量乌托邦社区试过，你必须说明差异。人的第一选择永远是改造现有环境来满足需求，而不是从头复制一切——「基于某个单点假设重写一切」不是现实中的成功策略。
javajosh查看英文原文
Is there any value in commenting on this, even to critique it? If you take it seriously, then you don't write a book about it you DO it (then maybe write a book). But this reads exactly like what a junior programmer would write about their ideal framework or language - it's entirely speculative, and so, worthless. (Of course, maybe the authors want to nerd snipe before they try it out, in which case I'd ask: what does cryptocurrency have to do with anything? Crypto is uncomfortable baggage that mediates (poorly) between the finances of a group embedded in a larger group. Also, how does this differ from starting a club, incorporating as a non-profit, start voting as a bloc to gain leverage, and THEN do...whatever the next step is? Long history of utopian communities attempting it, and you need to differentiate. The first choice is always going to be modifying your host to suite your needs, rather than attempt to reproduce everything. Again, the parallel with jr. dev thinking - "rewrite everything based on speculative benefit of one thing" is not a successful strategy IRL.)
宏大主张缺证据，只是乌托邦写作传统的最新翻版。
每一代都有自己的乌托邦：从毕达哥拉斯的素食公社到 20 世纪 70 年代的美国嬉皮士。真正意义上的巨变几乎都是一夜之间由贫困人口的巨大一致支持所推动的革命，比如法国。这是 70 年代旧金山嬉皮公社意识形态的最新化身：抛出宏大而笼统的论断，却没什么引用，除了过度简化的老生常谈再无实质。我估计很快会看到「生物圈 X——现在加入更多加密货币！」，只是没有 Steve Bannon（没错，他 80 年代真的参与过生物圈 2 号）。
SavantIdiot查看英文原文
Every generation has their utopians, from the Pythagorean's vegan communes, to the US hippies in the 1970's. The only time massive change has happened relatively overnight has been revolution with enormous, unified support of an impoverished population, see: France. This is the latest incarnation of the the 1970's SF hippy commune ideology. In that it makes broad sweeping claims without many citations or substance beyond oversimplified truisms. I expect to see "BioSphereX-Now with more crypto!" any day now, except without Steve Bannon (yes, he was actually involved in BioSphere 2 in the 80's).
同类还值得读
被低估的工程技能：把文章写好
Undervalued Engineering Skills: Writing Well
这篇文章讨论为什么“写得好”是被低估的工程能力，以及它如何在团队协作、系统设计和职业发展中持续带来回报。
写作方法2019 · 1,109 分 · 371 评4 条写作要点编辑精选每行只写一句话：让写作和修改都更轻松的排版法
Writing one sentence per line
这是 Derek Sivers 的一篇短文，主张写作时每行只放一个句子，而不是按段落排版。
写作方法2022 · 849 分 · 297 评4 条写作要点编辑精选书面沟通是远程工作的超级能力
Written communication is remote work super power
这篇文章讨论远程工作中书面沟通的价值，评论区则补充了异步与同步沟通的适用场景，并提醒读者：备忘录、结论先行、结构化命令格式等书面沟通传统早已有之。
写作方法2020 · 831 分 · 190 评4 条写作要点编辑精选如何写出有用的文字
How to Write Usefully
保罗·格雷厄姆在这篇文章里给"有用的写作"下了一个可操作的公式：有用≈重要×新颖×正确×清楚，任一项为零，整篇文章就等于零；他还主张敢下确定判断、删掉让自己心虚的句子。
写作方法2020 · 821 分 · 285 评4 条写作要点编辑精选如何写得更清楚、想得更清楚，并学会复杂材料（PDF）
How to write more clearly, think more clearly, and learn complex material [pdf]
这份 PDF 讨论如何把文章写清楚、把思路理清楚，并用写作来学习复杂材料。
写作方法2022 · 779 分 · 153 评4 条写作要点编辑精选