Skip to content
BranchesTags

Latest commit

 

History

272 Commits

Folders and files

NameName
Last commit message
Last commit date
.github
.github
 
 
.tools
.tools
 
 
ai-news
ai-news
 
 
archive
archive
 
 
articles
articles
 
 
.gitignore
.gitignore
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 
SECURITY.md
SECURITY.md
 
 

Repository files navigation

🌱 Tony Articles

盛大白的每日深度思考 · Tony's Daily Deep-Dives

Last commit Stars Follow @shengdabai

每日更新 / Daily 中英双语 / Bilingual 知识花园 / Garden 文章 / Articles

🇨🇳 中文文章（由近到远） ｜ 🇬🇧 English Articles (Newest First) ｜ 📚 全部双语文章 ｜ 🌐 知识花园 Garden

📚 全部文章 · All Articles

中文版与英文版均按发布日期由近到远排列。2021–2024 年旧文已补齐英文重写版。 Both language editions are ordered newest to oldest. The 2021–2024 archive now includes English re-creations.

📰 AI 圈每日热点 · AI Daily News

每天中午 12:00 自动整理过去 24 小时 AI 圈最值得关心的 6-8 条精选,面向 AI 工具、Agent、独立开发和自我进化系统,中英双语。 Curated daily at 12:00 — the 6-8 most signal-rich items from the past 24 hours of AI, with a focus on AI tooling, agents, independent builders, and self-evolving systems.

🇨🇳 中文阅读

这里是什么

我是盛大白(Tony),一名自由职业中文培训师。这个仓库是我每天的一篇深度思考。

不是碎片化的灵感速记,不是 AI 一键生成的口水文,而是一套完整写作工作流的产物—— 我把每天的观察、阅读、对话、突然冒出来的念头随手记进笔记系统, 再由一套固定的创作流程,把这些零散的火花打磨成一篇有出处、有逻辑、能引发思考的文章。

如果你想看更生鲜的、未经打磨的笔记原貌,欢迎逛我的知识花园 / Notes Garden 🌐—— 那里是这些文章的素材源头,也是我每天真正在用的第二大脑。

为什么做这件事

我相信几件事:

  • 跨时空连接 —— 真正有力的洞见,常常来自不同领域、不同时代、不同学科的意外撞击。
  • 工程化破界 —— 用系统化的方法把「想清楚一件事」变成可以每天重复的流程,而不是等灵感。
  • 自进化系统 —— 不是写完一篇就完事,而是让人和这套写作系统都在持续迭代、复利成长。
  • 用 AI 让人变强,而不是变懒 —— AI 是放大镜,不是替身。如果一个工具让你停止思考,那你用错了。
  • 长期主义、践行导向 —— 受李笑来影响,写作是体力活,只要你愿意,随时都可以。

如果上面这些理念让你有共鸣,这个仓库就是为你准备的。

每天的文章是怎么「长」出来的

一篇文章,从我的笔记到发布到这里,要经过 13 个阶段的工程化流水线。简化成 6 大步骤:

① 采集 · Gather

流程从我的笔记系统里召回当天积累的素材开始——可能是清晨听到的一句话,可能是地铁上读到的一段文字,可能是和朋友的一次争论。不依赖记忆,只读真实记录下的痕迹。

② 选题 · Choose

一天可能记了几十条笔记。选题这一步只挑那个最有张力、最贴近我价值理念的点——通常是一个能引发「咦?这事儿原来可以这么看」的反常识连接。其他都先放下。

③ 求证 · Verify

给文章里每一个关键事实——年份、数字、引用、人名、事件——找到真实出处。不靠记忆硬写。这是最耗时但最关键的一步,文章的可信度全在这里。

④ 自我反驳 · Steelman the opposition

主动找反例、找逻辑漏洞、找幸存者偏差。一个论点要是经不起自己反驳,就更经不起读者反驳。把它逼到墙角,要么推翻、要么重建得更强。

⑤ 成文与打磨 · Draft & polish

写初稿,调节奏,配类比,让它读起来像「人话」——不是 AI 那种四平八稳的辞令,而是有呼吸、有起伏、有温度的文字。

⑥ 双语再创作 · Bilingual re-creation

中英两版各自地道。不是机器翻译——而是用各自语言的内在节奏重新写一遍,让英文读者读到的是英文该有的样子,中文读者读到的是中文该有的样子。

整套流程每天中午 12:00 自动跑完一轮,产物推送到这里。

你会在这里读到什么

  • 关于学习与自我进化:怎么让自己持续变强,怎么避开学习的陷阱。
  • 关于 AI 与人的关系:在 AI 时代什么是不可被复制的、什么是真正属于你的护城河。
  • 关于跨时空连接:历史里的人和今天的我们,有什么惊人的共通处。
  • 关于践行与长期主义:那些「听过 100 遍但还没做到」的事,如何真正做到。
  • 关于作为中文教师的观察:语言、文化、教育、技术,在我这个独特位置上交织的视角。

怎么跟我互动

  • ⭐ Star 一下这个仓库,新文章会更稳定地推到你的 GitHub 主页
  • 👀 点 Watch → Custom → Releases(我会定期发月度精选 Release)
  • 💬 在 Issues 里告诉我你想看什么主题
  • 🌐 逛我的知识花园,看原始笔记
  • 📚 按时间翻阅全部中文文章，共 579 篇

同系列仓库 · 一起逛逛

这几个仓库出自同一套「在公开处用工程化方法做事」的实践,可以串着看:

  • 🌿 tony-knowledge-garden —— 公开的知识花园:AI、中文教学、学习方法的原始笔记与连接。
  • 🔎 Small-yet-smart-programs —— 自托管扫描器,每天发掘「小而美、可复制、能赚钱」的产品机会。
  • 🛠️ Tony-Claude-Code-Skills —— 我自己在用的 Claude Code skills、agents 与 Codex 工作流合集。

🇬🇧 Read in English

What This Is

I am Tony (盛大白), a freelance Chinese-language teacher. This repo is my daily deep-dive — one piece every day.

Not a quick note. Not AI-generated filler. Each piece is the output of a complete writing workflow: throughout the day I capture observations, readings, conversations, and stray thoughts into my note system; then a fixed creative pipeline turns those scattered sparks into a sourced, reasoned, thought-provoking article.

If you want to see the rawer, unpolished version of these ideas, visit my Notes Garden 🌐 — that is where the source material lives, and where I actually think every day.

Why I Do This

A few beliefs:

  • Cross-domain connections — the most powerful insights usually come from unexpected collisions between different fields, eras, and disciplines.
  • Engineering past personal limits — turn deep thinking into a repeatable system, instead of waiting for inspiration.
  • Self-evolving systems — finishing one piece is not the point; both the writer and the system should keep improving with compound interest.
  • Use AI to amplify yourself, not replace yourself — AI is a magnifying glass, not a stand-in. If a tool makes you stop thinking, you are using it wrong.
  • Long-termism, executed daily — writing is a manual craft. If you are willing, you can always start, anytime.

If any of this resonates, this repo is for you.

How Each Article Comes To Be

Every article goes through a 13-stage engineered pipeline, condensed into 6 steps:

① Gather

The process starts by recalling the day's raw material from my note system — a sentence overheard at dawn, a paragraph on the subway, an argument with a friend. Never from memory; always from a real captured trace.

② Choose

A day might produce dozens of notes. This step picks the single one with the most tension and closest fit to what I care about — usually an unexpected connection that surprises me. Everything else gets dropped.

③ Verify

For every key fact in the piece — years, numbers, quotes, names, events — find a real source. Never write from memory alone. This is the slowest but most critical step; the credibility of the whole article lives here.

④ Steelman the opposition

Actively hunt for counter-examples, logic holes, survivorship bias. If an argument cannot survive my own attack, it definitely cannot survive the reader. Push it into a corner, then either kill it or rebuild it stronger.

⑤ Draft & polish

Write the draft, tune the rhythm, find the analogies. Make it sound like a human wrote it — not AI flat, balanced phrasing, but prose with breath, rise, and warmth.

⑥ Bilingual re-creation

The Chinese and English versions are each native in their own language. Not machine translation — I re-write each in its own internal rhythm, so English readers get English-shaped prose and Chinese readers get Chinese-shaped prose.

The whole pipeline runs automatically once a day at noon (China time), and pushes here.

What You Will Find Here

  • On learning and self-evolution: how to keep getting stronger, how to dodge the common traps.
  • On AI and human work: in the age of AI, what cannot be replicated, what is actually yours.
  • On cross-time connections: striking parallels between historical figures and us today.
  • On execution and long-termism: ideas you have heard a hundred times but still have not acted on — how to actually do them.
  • On the view from a Chinese-language teacher: language, culture, education, and technology, intersecting at this peculiar vantage point.

How to Engage

Sibling repos · worth a look

All from the same practice of doing things in public, with an engineering mindset:

  • 🌿 tony-knowledge-garden — my public knowledge garden: raw notes and connections on AI, Chinese teaching, and learning.
  • 🔎 Small-yet-smart-programs — a self-hosted scanner that surfaces small, replicable, profitable product ideas every day.
  • 🛠️ Tony-Claude-Code-Skills — the Claude Code skills, agents, and Codex workflows I actually use.

📚 年份导航 · Browse by Year

全部文章已统一放进 articles/zh/ 与 articles/en/，每个年份页同样按由近到远排列。

All pieces now live under articles/zh/ and articles/en/. Every year page is also ordered newest first.

👤 关于我 · About Me

盛大白 Tony Sheng

  • 自由职业中文培训师 · Freelance Chinese-language teacher
  • 教过 6000+ 学员 · Taught 6,000+ learners
  • 终身学习者 · Lifelong learner
  • 相信「用 AI 让人变强,而不是变懒」 · Believer that AI should make you stronger, not lazier

找到我 · Find me:

「写作是体力活,只要你愿意,随时都可以。」

Writing is a manual craft. If you are willing, you can always start, anytime.

⬆ 回到顶部 · Back to top

About

🌱 工程化跑出来的深度思考实验场 · 每日中英双语原创 | Daily bilingual deep-dives on growth, learning & AI, written through a 13-step pipeline

notes.zturnsgo.com/

Topics

aiai-and-learningbilingualchinese-englishdaily-writingdeep-thinkingessaysknowledge-gardenlearningpersonal-growthwritingwriting-pipeline

Resources

Readme
MIT license

Security policy

Security policy
Activity

Stars

2 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

Packages

Used by

Contributors

Languages