这也是Pickful团队几个月来每天重度使用的工具，根据工作场景，效率10倍百倍的提高。强烈推荐。

moon

· 6个月

Crypto

· 6个月

这是我见过最好的Claude Code+Obsidian的教学视频。

花一个小时看完，你将学会用 Obsidian 打造第二大脑——让 AI 记住你所有的想法，帮你挖掘规律、生成创意、落地执行。

字幕处理失败

中文 English

0:00 这是黑曜石，而黑曜石就是人们用来作为

0:05 他们的第二大脑的小工具

0:06 但真正酷的是他们将它和Claude代码配合使用，他们

0:10 得到了

0:10 疯狂的结果。这真的是个游戏规则改变者

0:15 我一直很慢接受黑曜石，因为对我来说看起来有点令人生畏

0:19 所以我请了我的朋友

0:21 Vin，他清楚地解释了黑曜石是什么，如何和Claude代码一起使用

0:28 如何设置这些命令，从Claude和所有

0:34 大型语言模型中获得最大收益，真是

0:35 一集令人难以置信的节目，真正改变游戏规则的一集，因为我觉得懂得

0:40 如何使用黑曜石和Claude代码的

0:43 人们，将能够过上更快乐、更健康、更富足的

0:48 生活

0:48 为什么？因为它可以随时为你提供惊人的想法

0:53 所以我知道坚持看到这集结尾的人

0:57 我认为对很多人来说

0:59 这将彻底改变他们使用AI的方式，并且

1:04 这将是一个超级有影响力的方式，因为你会在

1:07 合适的时间、合适的时刻获得更好的想法，这会让你更快乐、更健康、

1:11 更富有。享受这集节目吧

1:21 我请来了亲爱的朋友Vin，也就是互联网Vin出现在播客里。我

1:26 真的是求他来

1:28 我求他，求这位先生来

1:33 教我们一件非常具体的事情。Vin，到这集播客结束时，

1:39 大家会学到什么？

1:40 我希望你了解如何把Claude代码和黑曜石

1:46 作为思维伙伴一起使用

1:47 我希望你了解如何停止

1:51 反复向智能代理解释事情，只需传入特定

1:55 文件，我还希望你明白

1:57 如何利用黑曜石和Claude代码去察觉

1:59 一些关于你思考方式的东西，而没有这些工具你自己

2:03 是发现不了的

2:04 好，从你口中说出，愿上天听见。我们开始吧

2:08 好的，首先什么是Claude代码？

2:14 Claude代码是一个可以使用

2:17 命令行界面的智能代理，基本上是你可以用来

2:23 控制你的电脑的工具，

2:25 你可以通过自然语言来操作它，对吧？

2:27 比如我可以说创建一个文件，或者在我的桌面创建一个文件，

2:32 内容是“Hello Greg”，用纯文本形式

2:36 它就会去执行这个操作

2:40 这真酷，这之前是无法

2:44 做到的

2:46 以前我得去桌面，打开文本编辑器，然后创建那个文件

2:51 对吗？

2:51 现在这个文件就在我的桌面上了

2:53 我可以说打开这个文件，瞧，Hello Greg。太疯狂了吧？

3:00 现在

3:03 有趣的是，如果你有这样一个可以控制

3:06 并操控电脑的智能代理

3:08 这意味着你描述给它的任何事情，它都能开始做了，所以

3:14 当你向它描述一个项目时

3:16 或者你和它进行长时间对话时，

3:19 它能做的事情越来越复杂，信息越多，

3:24 它能做的事情就越复杂

3:25 但问题是，如果你得，比如说你写了很长的项目描述，或者你和代理

3:28 进行了长达一小时的对话，谈论具体的

3:32 项目，我不想每次都新建会话去解释这些东西

3:35 我不想反复解释这些

3:40 很多人在网页上使用Claude或ChatGPT，它们

3:40 有记忆功能，

3:43 但你无法控制，你不知道那记忆里有什么，

3:48 你不知道它知道什么，不知道它不知道什么，所以

3:49 需要有某种方法，

3:53 方便快捷地向智能代理传递信息，

3:57 信息越好，传递越快，

3:59 它能为你做的事情就越多，

4:04 你也能越快地把任务交给它代理

4:06 好，所以

4:09 现在

4:11 即使我，比如说，我写了一个

4:14 大的项目描述，

4:15 创建一个描述项目的文件，

4:19 这个项目是这样的，

4:21 一个代办清单应用，

4:25 设计非常简约，

4:31 能读取

4:35 我的所有日程，

4:38 消息，

4:41 我的Slack，

4:43 并将它们转化成它认为我应该做的任务清单

4:45 这只是个想法，对吧？

4:49 这个文件可以放在我的桌面，

4:53 当我用Claude代码时，我可以引用那个文件，随时传入，

4:59 这很重要，

5:03 因为这是上下文，对吧？对，游戏的关键就是

5:07 给智能代理提供好的上下文

5:08 没错

5:11 我也不想反复这么做，当我花几天时间处理这个项目，

5:16 我不会记得我们之前谈过什么，所以我需要某种

5:20 文件可以传进去。抱歉，Greg，刚才说什么？

5:25 对，这就是很多人在用Claude代码时面临的问题，

5:27 他们用着用着，

5:31 会觉得还行，但不算改变游戏规则

5:32 That I can like pass in. Oh, sorry Greg. What's that?

5:35 Yeah, and that's that's sort of the problem that a lot of people are facing

5:39 with cloud code is like they're using it and then

5:42 They're saying well, it's it's okay. It's not like game-changing

5:45 问题是他们没有在正确的时间提供正确的

5:49 上下文。是的，完全正确，所以这里是

5:53 这是一个项目

5:56 描述，它写了，显然我可以传入这是一个

5:59 这是我刚创建的一个通用示例

6:02 但你可以让这些变得非常复杂。你可以将它们逐步构建成强大的

6:05 文件

6:07 所以我们知道

6:09 Cloud code 可以创建文件，它可以重复执行，也可以

6:11 读取文件，对吧？所以现在我可以说，比如我创建了一个新会话。这里是

6:16 一个新会话，现在我可以说我

6:19 想要处理

6:21 这个项目，然后我这里，这就是

6:26 待办 - 应用，搞定。现在我不需要再解释文件了，我

6:30 不需要重新解释项目，对吧？

6:32 所以它会读取这个文件，然后开始像

6:36 这帮我省了很多时间，很棒的项目，在深入之前有几个

6:39 问题来规划第一个会话

6:40 接下来继续。那么，什么是 Obsidian？

6:43 Obsidian 是一种工具，类似于一个界面，它位于

6:50 一组

6:53 Markdown 文件之上，对吧？这里它正在读取一个 Markdown

6:58 文件。我用 Obsidian 做故事开发，对吧？

7:01 我有每日笔记

7:04 这是我的每日笔记，这也是一个 Markdown 文件

7:09 我应该进行自己的基本分析，思考事物如何保持纯粹

7:14 当它们成长并变得更主流时，对吧？

7:16 这只是我有的一个文件，我有

7:20 一个关于 Greg Eisenberg 的文件，我还没放进去，很奇怪。

7:25 是的

7:25 我会做文件，记录我从别人那里学到的东西，

7:30 诸如此类

7:32 所以我为所有事情

7:34 有不同的文件，对吧？有趣的是，使得“vault”（保险库）这样的东西不同于普通文件夹的是

7:38 Obsidian 与整个系统交互，

7:41 它被称为“vault”，让它区别于文件夹的是？

7:44 Obsidian 不仅仅是与一个

7:49 文件夹里的文件

7:51 交互

7:55 它还能让你建立

8:00 文件之间的联系。比如我可以说今天我在一个播客节目中

8:03 与 Greg Eisenberg 合作，这个文件就链接到了那个 Greg Eisenberg 的文件

8:08 非常有趣，非常有趣，所以很多人，像很多人

8:13 非常喜欢使用 Obsidian 和类似工具，

8:17 就是因为这个

8:19 构建互相关联的能力

8:21 这是单纯的电脑文件夹无法做到的

8:26 显示这种相互关系

8:28 所以当你开始

8:32 持续建立这些相互关系时会非常有趣

8:35 对了，这里有个小的可视化，在这里

8:41 每一个圆圈代表一个文件，

8:45 并显示它们如何与我写过内容的其他文件相连

8:50 比如这里是个人代理

8:53 基础设施

8:55 对，我可以看一下

8:58 我觉得我还应该加点评论，说说做这个演示有多难

9:01 这里面有很多个人信息，因为这是我的

9:02 个人资料

9:06 所以我甚至不知道屏幕上会显示什么，对吧？

9:07 但这就是做这种演示的一部分，虽然奇怪

9:10 但很有趣

9:14 你可以看到个人代理基础设施链接到像 Genetic AI

9:15 这里有个链接到 telegram

9:19 这里也有个链接到 Toby

9:21 和 Shopify 创始人有关的链接。还有链接到 Presence Log Quad Bot

9:23 你知道，然后这里我有一个播客叫“其他东西”，然后

9:28 你会看到我显然在

9:32 思考很多东西，对吧？

9:35 所以我

9:40 还可以，比如说我去 Greg Eisenberg，打开本地图谱

9:44 这里显示我写过关于 Greg Eisenberg 的所有内容

9:48 关于时间限制的笔记，如何使用 Obsidian，这很

9:52 有趣

9:53 如果我说“我也爱你”，那我就说“我也爱你”

9:56 如果我在听节目并发现不同的模式，

10:00 我可以把它关联回 Greg。非常有趣，但

10:04 关键是

10:06 人们喜欢 Obsidian 是因为这些相互关系

10:10 你可以打开一个文件，然后你

10:13 打开这个文件，然后想，哦，有提到 Greg Eisenberg，

10:15 我可以点开，直接跳到相关文件

10:17 这很有趣，没错。它的工作方式更像是

10:22 你的大脑运作方式，你的大脑一直在连接各种模式

10:25 问题来了。是的，我知道这很有趣，

10:31 但这怎么能让我产出更好的结果？

10:36 正是如此

10:37 接下来，Obsidian 推出了一个新工具叫 Obsidian CLI，

10:39 它让你可以使用 Cloud Code，

10:43 它能读取你 Obsidian vault 中所有文件，vault 就是

10:48 一个文本文件夹

10:53 但通过 Obsidian CLI，Cloud Code 不仅能读取和访问这些文件，

10:55 还可以获取这些文件之间的相互关系信息

11:00 你可以看到 Cloud Code 可以知道，这个文件连接了那个，

11:02 还有那个，和那个

11:08 这对于 Cloud Code 理解你，以及它能

11:09 做到什么来说，非常有趣

11:14 非常关键

11:16 对于你来说意义重大

11:20 以及 Cloud Code 能做的事情

11:22 了解

11:24 你正在处理的所有事物之间的关系，可以开始

11:28 显现出你思考内容的模式

11:31 你并不是

11:32 亲眼看到一些你可能已写了一年的想法

11:35 在这个知识库中

11:36 它可能是一个潜在的想法，能立刻提醒你说，嘿

11:39 你知道自己一直在写关于这个相同的模式和创业公司吗？

11:43 在这个特定项目中？

11:45 你在每一条笔记中工作

11:47 你在不同领域做的笔记，第一次看到它们

11:51 可能会产生巨大的灵感闪现效果

11:53 它能带来巨大进步，你在学习和

11:56 理解以及你对世界的看法

11:59 还有你所做的工作

12:01 所以我现在写了，我想演示这到底如何运作

12:06 我如何传递

12:07 信息给一个没有 obsidian 和 cloud code 根本无法做到的代理

12:11 代码

12:11 这是我用的一些命令，我不希望你害怕

12:16 这些东西

12:18 我知道看起来很复杂

12:20 但是

12:21 这是我

12:23 我已经准备好的一些命令，这是我创建的终端

12:27 在 obsidian 中运行

12:29 你不必使用它，你也可以在自己的终端

12:33 会话里用任何工具操作

12:35 但我放在 obsidian 里，是想把它们都整合展示

12:38 并且展示你如何集成和定制

12:41 这个环境

12:42 所以

12:44 这是个酷东西。slash context / context 加载关于我生活的全部背景

12:50 工作和当前状态读取

12:52 上下文文件、每日笔记，并跟进反向链接构建完整画面。所以

12:56 我就直接展示给你看

12:57 比如说我新开了个会话

12:59 在一个

13:02 我桌面上的 cloud 里，现在我准备开始做点什么

13:06 但在开始前，我先输入 context demo

13:09 它会读取大量关于我当前状态的文件，完成

13:13 我已经预加载了所有这些上下文

13:19 你看它开始读取所有这些文件

13:21 正在读取读我文件，正在读取关于 new 的背景，这是我正在做的一个媒体公司

13:25 它还在读取其他内容

13:27 正在读取我的个人其他项目，这是我的节目

13:30 正在读取个人工作流上下文，所以我无需担心

13:34 它不知道我想了解的关键信息

13:37 我只需执行一个命令，它就完成了全部信息的收集

13:40 然后我可以用 slash today，它是早晨复盘，会拉取日历任务

13:45 我过去一周的每日笔记消息，生成当天的优先计划

13:49 这为什么重要？

13:50 当然，你可以设置一个代理，给它访问你的日历和

13:55 你的任务还有消息等权限

13:58 但那缺少了所有关于你在思考什么和为什么的信息

14:01 如果我每天写笔记关于某项

14:05 特定技术或项目或兴趣，那我的

14:08 日历是否会主动反映这些？它是否匹配我

14:12 实际上正在写的主题？

14:14 如果代理有这些上下文，它能更有效地给你

14:19 关于你应做什么或不应做什么的信息，或者更有效地为你

14:24 决定日历中应有什么或不应有什么

14:27 还有一个命令 slash closed day

14:29 日结处理

14:32 提取行动项，显现知识库关联，检查置信度标记

14:36 需要不断更新

14:37 我有很多假设，我会给它们一个置信度

14:42 等级

14:42 这是我正在做的一个想法，我对此很有信心。还有另一个想法

14:46 我正在做，我不太确定

14:48 这些都是

14:50 日常操作内容，但

14:52 这是我用 obsidian 最多的用途，作为思考工具。我

14:58 非常非常喜欢将大型语言模型作为思考伙伴一起工作

15:03 这是我使用大型语言模型的最喜欢方式

15:05 我知道有人喜欢用代理和大型语言模型来构建东西

15:09 但我更喜欢用它们陪我一起思考，创造当我觉得

15:14 你知道

15:15 我确实有一种新颖的看法。所以

15:17 我们来看这里

15:20 所以 ghost

15:22 这是个命令，让它用我会的方式回答问题，构建

15:27 一个来自知识库的声音档案

15:29 用那种声音写作，然后评估可信度

15:31 我可以直接问“我怎么看 AI”，我给你展示一下

15:36 挑战主题，如果用知识库自身历史测试当前信念

15:41 寻找矛盾、相反证据和思路转变

15:44 这为什么重要？

15:46 如果我想确保自己作为一个人持续成长

15:50 以及在技能上

15:52 我想确保自己的观点不至于过度偏见或局限

15:57 这就可以挑战我

15:59 浮现

16:01 显露知识库暗示但未明说的想法

16:04 从零散前提、未命名的模式、未表达的方向得出结论

16:08 这非常非常有用，因为很多时候

16:12 你知道，我可能会陷入一个阶段，只是在多种方式中提出想法

16:18 多年

16:18 仅仅有人对我说一句简单的话：

16:23 嘿

16:23 这只是给这个想法命名。嘿，你知道你一直在绕着

16:27 这个模式转吗？

16:27 重大突破

16:30 Slash drift 它将我声明的注意力意图与

16:36 实际行为在30到60天内进行比较

16:38 显现我在回避的内容

16:41 播客上的人们，听众们可能会喜欢这个深度的30天

16:45 保险库扫描，结合跨领域模式卡组

16:48 检测和图谱分析，生成跨所有领域的想法。

16:52 这给了我

16:53 不仅是我应该处理的事情的想法，它还给我工具

16:57 和类似的东西的想法

16:58 但它也给了我想法，比如我应该看的电影

17:01 应该再买的产品，所有这些都受我

17:05 在保险库里写的东西影响

17:06 Trace 追踪想法在保险库中随时间的发展轨迹。我们来看

17:13 一些东西

17:14 trace 演示

17:17 我已经做过这个，操作方式是我创建了一个

17:21 标签页，然后我可以直接输入 Claude

17:25 Trace，我不得不为所有这些创建了演示版本

17:28 命令，因为我的保险库里有太多个人信息，但我

17:33 甚至不知道，无法控制

17:36 屏幕上会显示什么，我只是个糊涂的问题，比如

17:40 我们看到的那些命令

17:42 是

17:44 你创建的命令，还是 obsidian 自动生成的？那些是

17:48 我创建的命令，

17:52 你可以通过让 Claude 代码创建特定

17:56 命令来非常轻松地创建它们

17:56 嗯，我们可以包括

17:59 我会在节目说明里加上

18:02 一个链接，大家可以点击获取 VIN 的一些

18:08 技能，如果可以的话，没问题。你可以看到这里

18:15 我所做的是只输入 slash trace demo，demo 是因为我公开展示

18:18 我是如何使用 obsidian 的，

18:21 这个操作会追踪我的，

18:24 我如何发展、导航这个想法穿越我的保险库

18:29 我这里有个例子，很有趣

18:33 我做了它，然后让它运行，它做的就是，正如所说的

18:39 全部追踪

18:40 你与 obsidian 的关系是如何演变的，让我先构建一个

18:43 词汇地图并在保险库中搜索

18:45 所以

18:47 它开始浏览保险库，读取所有

18:52 这些不同的文件，可以看到所有通过 obsidian CLI 连接的文件

18:54 这真厉害

18:58 这是我自己绝对做不到的，去

19:02 阅读所有这些文件，

19:04 了解它们如何相互关联

19:06 这作为人类我是完全做不到的，然后它会输出到

19:10 这里

19:10 我现在拥有构建这个追踪所需的一切。这是完整的演变

19:13 追踪我如何使用 obsidian

19:16 首次出现2025年1月11日，时间跨度13个月。这是针对这个

19:22 保险库里的所有文件特定的

19:25 它显示

19:27 保险库之前2024年12月，我如何在多伦多西区做笔记的文章

19:32 日期为12月1日

19:33 2024年，描述了一个完整的系统，其中 obsidian 没有任何角色

19:38 这个系统是通过 Mac Whisper 的音频转存，LLM 对话循环，Canopio 用于

19:42 空间映射，纸质笔记本用于认知摩擦

19:45 未来平台用于跟踪，Arena 用于片段，流程是捕捉、

19:48 处理、结晶化

19:50 这是基线，obsidian 不在其中，发现和设定

19:54 怀疑阶段2025年1月至5月

19:56 保险库中的第一条每日笔记，既兴奋又不确定

20:01 我可能也会把转录放这里，作为存储方式

20:06 目前的理论是，以定向链接的方式做笔记

20:09 并不太有用，但我不确定

20:11 然后

20:14 它显示选定的工具

20:16 最初我用的是反向链接。它显示

20:20 我如何使用 obsidian 的笔记，一个关键的反向链接认知

20:24 最初，我用反向链接指向通用主题，如播客、健身或

20:28 电影制作

20:29 我意识到这不是使用 obsidian 的最佳方式

20:32 最重要的是为我的每个模式

20:36 理论、项目和视角创建笔记，把它们从脑海中写出来，

20:39 然后链接到那些笔记

20:41 所以它就是从

20:44 我写过的内容中抽取，形成了这一

20:50 概念的历史

20:51 我可以用任何事情做这件事，比如初创公司，

20:54 特定项目、食物、我的人际关系，就像是

20:59 一个爱好

20:59 任何东西，没错。然后第四阶段2026年1月，一个爆发式建设月

21:05 一切仍需

21:09 我积极提示和管理每个部分

21:11 下一步的解锁是弄清楚如何让代理任务自动运行

21:15 自动化

21:15 摩擦不再是 obsidian 本身，而是保险库和代理执行之间的界限

21:19 所以你看，我真的在逼自己，

21:20 这很酷

21:24 这对我理解自己

21:28 对这个工具的使用如何进化非常有用

21:32 我觉得太荒谬了，我只是

21:38 做笔记，

21:40 关于我生活中各种不同事情，甚至作为父亲，

21:44 我都能反思我学到的不同东西。我就是觉得这

21:49 疯狂的是，和

21:50 一台电脑能掌握这么多关于我的信息，并揭示这些模式

21:55 我自己做不到这么快，这真是个很棒的工具

22:01 对我来说

22:01 就像我现在可以直接在这里写，边想事情边写

22:05 它还能给我灵感，对吧？

22:08 关于

22:09 我的生活和我正在做的项目

22:11 所以我可以说，我的Obsidian关系的演变很有趣

22:19 随着时间的推移

22:21 这让我思考很多

22:25 我和电脑的关系是如何发展变化的

22:32 从小时候到现在

22:35 现在

22:37 很有趣

22:39 这些事情是如何随着时间积累发生的，而我们并没有真正意识到

22:46 是吗？

22:46 这就是一个想法的例子，对吧？

22:51 所以我认为掌握Obsidian的一部分

22:56 是反思，将反思融入日常生活，因为很多

23:03 人

23:04 我们从一个会议赶到另一个会议，非常忙碌，我们是父母

23:07 嗯

23:10 我们长大后当然会在笔记本上写东西

23:14 但我觉得随着年龄增长，我们写作和反思反而越来越少

23:19 是的

23:19 你知道，如何？

23:22 你是如何将反思融入生活的？

23:27 嗯，我想

23:30 对我来说，反思有两个有趣的理由

23:34 写笔记也很有趣

23:37 其中一个是

23:39 可以回头看这些笔记，比如我现在

23:43 显然，我可以用智能助手

23:44 但我去看这些笔记时

23:47 意识到我这个人一直在变化，我的技能

23:51 一直在变，项目也在不断发展

23:53 反思是生活中极其美妙的一部分，可以看到

23:58 事情如何随着时间变化

23:59 你如何变化，世界如何变化

24:03 但还有一个功能性原因

24:07 我喜欢频繁做笔记的原因是这是我产生想法的方式

24:10 当我坐在电脑前

24:12 写东西的时候，想法就来了，比如这件事，我

24:15 刚刚写下了

24:15 只是一个快速笔记，实时写的，随手写的

24:18 对

24:20 通过写出来，我感觉更能内化它，而且我喜欢有好的

24:25 想法

24:26 我喜欢进步，因为我喜欢好点子，也喜欢进步

24:30 写作就是我的方式，所以我认为

24:33 如果你想养成写作习惯，我觉得首先你要

24:36 把它和这个想法联系起来：这是你进步的方式，这是你

24:40 产生想法的方式，这是你

24:44 形成原创思维的方式

24:46 另一个我要说的是

24:48 写作

24:51 现在是你委派任务给智能助手的重要方式

24:52 这是一个全新的层面。所以如果你能养成

24:56 写作习惯

25:01 你就能给智能助手提供更多上下文，进而

25:02 极大提升你能委派的任务量和

25:07 能做的事情

25:11 我希望这是个好回答

25:13 这和

25:15 这和OpenClaw有什么关系呢？

25:17 如果你想到OpenClaw，

25:21 本质上

25:23 它就是你的延伸，可以去做事情，

25:30 你知道的，

25:32 独立完成

25:35 或者在你的指导下完成。所以你怎样将Obsidian的命令、

25:41 OpenClaw和反思

25:44 协调结合呢？

25:47 我觉得比如这里有个我使用的命令示例，

25:52 就是一个日程安排命令

25:54 所以

25:56 我要求它做的是安排

25:59 我说今天2月20日下午2点能安排和Greg Eisenberg的会议吗？

26:03 对

26:05 它会检查我的日程等

26:08 并且会查看我的每日笔记

26:11 它会查找我关注的事，然后给我一些

26:14 建议。比如它说

26:16 你的日程已经满了，今早你有Greg播客录制

26:20 接着是团队午餐和和Peter、Vince的会议

26:23 你2月17日的笔记显示，Greg的那集一直是重点，有专门的Greg Eisenberg笔记

26:26 没有2点的会议，所以建议是不安排

26:28 实际上你可能根本不需要额外开会

26:31 这才是正确答案

26:35 这和OpenClaw有什么关系？

26:37 嗯

26:39 OpenClaw就是一个自治智能代理，如果你设置好，

26:44 它可以不需要你一直提示，自动处理事情

26:48 它可以自主做决定，为你搭建东西

26:49 独立完成

26:51 所以现在OpenClaw可以

26:53 像我刚才做的那样执行命令，也能自主执行

26:57 它可以读取我的知识库，找到关联，做出决策

27:00 然后采取行动

27:05 代表我，更深刻地理解我自己，而不是像

27:10 管理一个代理或与另一个人

27:12 类的人谈论某件事。我只是专注于管理这个知识库。这是

27:16 新的来源

27:17 我不断努力让这个知识库拥有所有

27:21 所需信息，这样我就可以委派给代理，

27:25 代理可以直接从这个知识库来源获取信息并做出决策，如果它

27:29 做出错误的决策，

27:30 我会更改知识库的内容，而不一定是直接与代理

27:34 合作。

27:35 这大概是我对这个话题的猜测。

27:42 我觉得非常有趣。是的，我想

27:44 让我有点担心的是，如果Obsidian真的是你的

27:49 第二大脑，

27:50 给OpenAI访问你的第二大脑是

27:54 可怕的，是的，很可怕，我想说这是这项技术

28:00 最根本的

28:06 奇怪的元素之一，我故意给

28:12 Obsidian，呃，不是，给OpenAI或者任何代理访问大量信息，

28:19 我就是故意这么做，

28:21 因为我和这项技术的关系是：我想了解这些东西是什么，我想

28:26 了解它们揭示了什么，

28:28 你知道我们和计算机的关系是如何变化的，

28:31 但是，

28:33 这很怪，得真正考虑你

28:37 分享给这些

28:38 代理多少信息，以及这是正确的决定还是错误的决定，

28:43 我认为

28:44 隐私作为一个概念将如何发展和

28:48 变化，以及我们为之奋斗或不为之奋斗，都会非常有趣，

28:52 还有未来的

28:53 我们的社会和世界，即使对这些命令中的每一个，我都必须创建

28:58 它们的一个新版本，一个演示版本，

29:01 这样我在播客录制时不会泄露太多个人信息，

29:05 即使如此，这也是个权衡，

29:08 你知道，我可以输入演示版本，但谁知道屏幕上会显示什么，

29:11 你知道的。

29:12 你还想展示哪些其他命令？

29:16 所以，

29:20 有

29:21 Connect（连接）

29:22 它允许我连接两个领域，利用

29:26 知识库的链接图谱，

29:27 我可以说我

29:30 在这里做了一个，

29:32 然后我让它

29:35 连接

29:38 电影制作和世界构建，它会读取所有这些

29:41 不同的文件，

29:43 然后开始说好，让我们把这两个概念连接起来，

29:47 比如笔记和电影制作，

29:50 笔记和电影制作邻近区。所以我写了35部电影观看清单，这是我和Toby的第一次

29:55 会议，

29:57 笔记在世界构建邻近区，

30:01 世界构建的文章，

30:03 作为新媒体公司。这些是我正在思考的不同内容，所以

30:07 桥接一个是采访门户和构建的世界，

30:10 如果我在电影制作中注意到具体某件事并提出问题，

30:14 它会打开通向某人的内心世界的门户，那个世界往往

30:18 是丰富的概念、信念和愿景宇宙。

30:20 关于世界构建的文章，我

30:22 希望我的博客能向你展示我重视什么、相信什么、担心什么，像

30:27 一座古埃及的坟墓，

30:28 我希望我的博客成为一个地方，你即使在我离开后也能来这里挖掘和研究，

30:32 这些是我写的内容，我能看到这些想法如何

30:35 相互连接，

30:36 桥接二「常驻纪录片」=持续的世界构建，

30:40 常驻纪录片是一种创意策略，公司通过纪录片持续叙述

30:45 他们角色的追求、冲突和愿景，

30:49 这些是我写的内容，它向我展示这些内容

30:53 的连接方式，我

30:54 认为这可以变得

30:57 非常有趣，取决于你愿意

31:01 连接哪些内容，

31:01 根据你在知识库中写的内容，你也可以玩得很疯狂，

31:05 比如我想，我可以连接烤肉卷和初创公司，看看

31:06 它们之间产生了什么样的联系，

31:10 这真是很有趣，因为所有这些发生得非常快，而且我

31:12 不需要向大语言模型解释任何事情，

31:15 我只要输入内容，就像

31:18 斜杠 connect filmmaking world building，

31:20 你用的很多例子都是个人反思。是的。你

31:22 怎么想的？

31:26 你知道，

31:32 比如会议记录，也许你有像Granola或者Gemini这种

31:32 做笔记，

31:36 你知道的，做笔记或者把笔记放进Obsidian，

31:41 顺便说一句，当我说笔记时，可以是你没参加过的会议，

31:42 他们可能只是，

31:45 你知道，Tommy和Vince开过会，我想把它放进这里，

31:51 是的，

31:55 你怎么看这事？

31:56 这是个好问题。我认为你可以随意使用这些知识库，

32:01 无论是

32:02 以什么方式，

32:05 你想放入任何文本都可以，比如你想把Granola会议

32:09 的文字记录放进来，

32:10 你都可以放进去，只要你保证，

32:14 知晓内容的安全即可，

32:15 你必须确保你知道

32:18 也许你是在这么做，对吧？就是开会然后

32:21 然后你就想着，好吧

32:22 这些是，也就是项目一，然后每次开会

32:26 你做会议记录，然后放到这里，对吧？所以你

32:30 就是和Greg Eisenberg开会

32:32 然后

32:34 再加上，然后再然后

32:36 你知道的

32:37 这就创建好了，现在你可以直接丢你的特伦特

32:40 你可以直接把会议记录丢进这里

32:42 然后它就在保险库里，然后你可以传给智能助手或者

32:45 智能助手会发现它，对吧？

32:46 尤其是如果你开始标记，比如说我要标记回，比如说

32:50 我的播客什么的，现在

32:52 它就被连接了，所以智能助手有了更多上下文，然后

32:57 知道这个特伦特的记录和这个文件有关，很棒

33:00 我觉得这取决于你

33:02 我觉得你放多少信息进这里，取决于你

33:05 想怎么用保险库，想怎么

33:08 把任务交给智能助手，也许你甚至想创建不同的保险库

33:11 用于不同的用途，对我来说

33:14 我用大型语言模型和智能助手来

33:18 提升我自己对某些主题的理解程度

33:21 所以我用它做很多反思之类的事

33:23 所以我不希望智能助手自己写进文件里

33:27 比如我可以轻松让它这么做，我甚至可以说，比如说这里

33:31 我让它写了一些我今天会谈到的命令的说明

33:35 今天

33:35 但是我不想让它做文件，因为我想

33:39 控制我Obsidian里所有文件

33:41 保险库，因为我总想引用我自己对事情的看法

33:45 而不是它对事情的看法，如果它开始自己在

33:49 这个保险库建文件

33:50 那我就不清楚它在找的模式，是它写的东西的模式，还是

33:54 我写的东西的模式

33:57 所以我给自己定了一条规则，就是严格区分

34:02 这两种东西

34:03 我只想在这里旁边写东西，然后我会拿那个内容

34:08 然后写我自己的想法

34:11 应该包括什么内容

34:13 对。是的，是的，继续。我刚想说，

34:17 我能理解用它来做自我反思的力量

34:24 我也能理解

34:27 人工智能的强大之处

34:30 你知道

34:32 去网上找基于趋势的信息，然后

34:36 以你想要的方式提炼它，

34:39 并且把它放进你的世界，这也很有趣。是的

34:44 完全正确，我觉得还有，比如说，如果你让Obsidian，

34:51 如果你让

34:51 Cloud Code去浏览你的Obsidian文件，生成想法

34:55 这些

34:58 你知道，你应该做的工具的想法，那么你就可以说

35:01 好，太棒了

35:02 如果我有个想做的工具，帮我生成一个

35:05 描述，然后我就去做工具

35:08 没错

35:10 是的，我想展示这个

35:13 这样就少了反思的成分

35:16 比如说，我做了个叫idea browser的网站。

35:23 每天

35:24 都会给出一个经过验证的创业想法，理论上有人可以

35:29 拿这些信息放进Obsidian保险库里，然后基于

35:35 这些

35:36 实质上，帮他们做出真正的东西，没错，完全正确

35:41 我想给你看看，因为我觉得它真的能让它

35:45 从反思的领域走出来，进入创造的领域，

35:49 但唯一的问题是，这需要一点时间，

35:51 没关系。对，还有就是用这些

35:58 命令我注意到的另一个模式是，它们需要一点时间，

36:01 因为它们要读很多文件，我觉得这是用Obsidian和

36:05 用Cloud Code访问我这个Obsidian保险库的一个很大区别，

36:07 我注意到的是

36:12 我所有的

36:12 请求都慢了很多，就是因为它读的文件多了很多。

36:17 你看看这个

36:18 这是一个想法演示，我会运行一个全面的想法生成，

36:23 让我先并行收集保险库结构和上下文，然后

36:27 如果你看，会发现它做了很多有趣的事情，

36:30 对

36:31 它在找Obsidian孤儿文件，

36:33 也就是说孤儿是指那些

36:39 孤立无援、没有连接文件的文件

36:40 很有趣，它能知道这些

36:42 Obsidian死角、Obsidian已解决、Obsidian标签计数，它是在

36:47 尝试弄清这些文件之间的联系

36:50 然后它说，好，读取每日笔记

36:55 然后找到了一份叫new context的文件，是我在做的新媒体项目

36:59 然后

37:00 又读取了其他内容，就是播客那边的上下文。我还想说，

37:04 我管理

37:07 我写，我为比较详细的项目创建上下文文件，我会

37:13 给你看

37:13 我本来不知道要不要展示这个，因为它很私密，不过

37:18 对于其他的东西，

37:18 你看看这个其他工作的上下文，最近发生了什么变化？

37:23 前期投入的简介，是通过去旧金山、纽约录制嘉宾

37:27 嘉宾

37:27 这都是非常私人的内容

37:29 但其他东西，节目的核心信念和研究

37:34 坚实的基础，最好的对话就像发现般坚实

37:38 你知道，有个团队正在努力。所以接下来发生的是

37:44 非常个人化，但这是

37:46 上下文刚刚被拉入，所以现在它知道谁在做我的播客

37:53 最近的假设是什么？我正在探索哪些假设？

37:56 它刚拿到了那个信息，这只是它做的事情之一

38:01 个人工作流程上下文

38:02 超级个人的文件，但它显示了

38:06 你懂的，比如

38:09 我的日程安排是怎样的，知晓我生活中必须做的事情

38:13 个人生活

38:13 所以它拉取了我喜欢怎样工作，不喜欢怎样工作的信息

38:17 个人代理基础设施

38:20 还有一件事。看看拉出来会怎样

38:24 这是一个项目，我想踏出一步，增强我的个人

38:28 基础设施工作流程委派

38:30 无论你怎么用代理来描述它，理解它的含义

38:32 越来越多地委派给代理

38:34 实施方法

38:36 这就是我在写的内容

38:39 关于我个人如何使用代理的文件，且那是

38:44 文件的一个例子

38:46 它正在读取，这只是其中之一。你也看到了其他内容

38:49 个人工作流程

38:50 它把所有这些都考虑进我让它做的任务里，也就是

38:54 帮我生成点子

38:55 从你的日记、日程和volstructure里收集数据。这需要

38:59 一会儿，因为它又从多个来源拉取数据，这是这套系统的

39:03 特点：它要处理大量信息，很多信息

39:08 大量信息

39:09 所以它会更久，你知道，它已经运行了五分钟了

39:13 是的，这就是我所注意到的，但

39:15 对我来说，这正是我想要的，我想要LLM给我

39:22 非常非常有上下文的回应

39:24 针对我写的内容和思考，我想这就是

39:28 我和代理最好的协作方式，我只专注于

39:32 持续

39:34 记录我当前项目的进展、我的理解和我发现的

39:39 有趣之处，我想保持这些信息

39:41 尽可能最新和深入

39:45 可能

39:45 所以无论何时我和代理交流，它都能最好地展现

39:49 我在那个时刻的状态，当我请求代理帮助时

39:53 是的，这就是目标

39:55 这是我们都应该问自己的问题，就是

40:00 代理是否拥有关于我项目、偏好、梦想、希望和目标的最新信息

40:04 因为它的好坏，取决于它信息的最新程度，对吧

40:10 没错

40:15 百分之百，代理所掌握的信息质量，完全

40:16 决定了它能为你做什么

40:21 对，如果它对你了解不多，帮不了你太多

40:23 对

40:27 但如果它了解很多，它就能为你做很多事，我觉得

40:27 甚至有时是你自己都未意识到的事

40:32 有点奇怪这么说，但就是这样

40:36 这很合理，因为

40:40 归根结底

40:42 它是在串联点滴，是的，city和Obsidian

40:44 和Claude代码都在串联点滴

40:50 现在

40:53 作为一个企业主或个人生活中的个体，要串联这些点其实很难

40:54 为什么很多人在某种程度上会去？

40:59 教练、治疗师那里

41:01 是的

41:03 你去看治疗师，

41:06 一般

41:10 你说话最多，对吗？想想看

41:12 你在大量反思，而治疗师或教练

41:16 是在引导你

41:19 从某种角度说，这就是这个系统在做的事，我不是说

41:20 别去看治疗师，只是说

41:24 我的观点是它帮助你揭示

41:27 那些点，以及如何串联它们

41:33 是的，绝对如此，对我来说，这真的很令人兴奋

41:37 是的，这是个疯狂的时代

41:41 计算机的发展

41:46 我们来看这个，这任务完成了。这是一个点子生成报告

41:49 vault关系探索。这相当

41:53 详尽，对吧

41:56 来一份点子报告

41:58 我觉得这会很好地展示我们如何从

42:04 反思迈向可执行的行动

42:07 所以

42:09 结构亮点。这依然是Obsidian的东西，比如孤立文件

42:14 值得注意

42:15 这里有些防务技术内容，这是加拿大

42:19 正在增长的一个主题

42:20 大量的知识投入，独自思考，还有孤儿代理软件

42:24 孤儿意味着这些文件

42:26 我没有真正链接它们，只写过一次或类似的

42:29 未解决的链接揭示隐性兴趣

42:34 隐藏的关联，还是反思内容，很好。Obsidian

42:39 Cloud code的组合系统对我很有效

42:42 它带来了真正的突破和思考产出

42:45 当天域结构被强制实施时，这基本上是我开始

42:49 把我的日程分开，每天有具体重点。这很棒

42:54 这非常真实，Greg Eisenberg那集是个推动力

42:58 它把几个月对Obsidian和代理的思考压缩成清晰

43:02 的论点和示范，非常真实

43:04 你知道，上节目并做这个，迫使我

43:07 综合我所知道和呈现的一切

43:11 但这里我们开始进入可操作的内容，构建工具

43:16 基于每日笔记的斜杠毕业斜杠命令每日笔记创意提取器

43:23 充满了创意

43:24 抱歉，每日笔记充满了创意标签和有趣的想法，但从未

43:27 得到发展，库里有九个创意标签

43:30 但有数百个未被发现的洞见，构建一个命令，扫描最近的每日

43:35 笔记，识别带标签或未带标签的创意，并提示你决定

43:39 创建独立笔记、添加到现有文件或忽略

43:43 这将每日笔记流变成结构化的创意管道

43:47 针对新手的Obsidian库，我只需要管理和设置一个中央ob

43:51 sidian新手库

43:52 这意味着，就像

43:55 我创建这个所有的库，里面有我所有的创意和模式及

43:59 诸如此类的东西

44:00 我为什么不为我的团队创建一个呢？

44:02 作为团队，我们可以去提问，我们都可以

44:07 贡献进去

44:09 开始使用的工具来了

44:12 这是什么？针对所有外部文档的类型？

44:15 有趣的时间分块应用，强制按日期分块，一个

44:21 按领域强制每日分块的应用，意思是我每天专注一件事

44:25 生活中的某个方面。它说为什么不创建一个强制执行的时间分块应用

44:30 很有趣

44:32 待实施的系统

44:35 一句话，代办Obsidian代理处理任务。这就是演示

44:39 第三版Greg Eisen的

44:41 Eisenberg预设。你已经在想象了，下一步是实际

44:45 构建它，从小做起，对吧，安排一次本周的电话讨论某个主题

44:49 在每日笔记中，然后让Otis或clogbot或open claw接手

44:53 处理它

44:54 所以它可能意味着你可以直接从笔记中委派任务，我是

44:58 这么理解的，非常有趣

45:00 是的，我认同，内嵌委派，也许这就是一种新的

45:06 用户体验模式

45:07 我甚至不知道，你可以把它内建到这些不同工具里

45:09 待调查的主题

45:14 Christopher Alexander的模式语言应用于数字空间，很有趣

45:18 Black Mountain College作为体育场馆模型，体育场是一种物理空间

45:23 我们在多伦多有

45:24 无作者媒体这一概念

45:26 深圳硬件生态如何运作，写作和发布的内容

45:32 有用的东西

45:33 语境架构论文，计算机作为场所，软件书籍将成为

45:38 时尚

45:39 多伦多理论究竟是什么？

45:42 编辑思维小刊

45:45 待进行的对话，不是你，是现实中的人，Aaron体育场

45:49 工作坊主持，关于成为技术编程的核心

45:52 这里是一个空间，我们没有，多伦多要毁了，还有另一个人的项目

45:57 关于制作项目的旗舰系列

45:59 Steph 和 Obsidian CEO，关于库作为一个场所

46:01 所以这是

46:05 这真疯狂。它建议了我应该见的人，前五名

46:10 高影响力，马上做，构建毕业命令或执行手动每周创意审查

46:16 ？

46:16 这太疯狂了，伙计。这真的很疯狂

46:21 就像

46:23 而且它是纯文本，不是，不含

46:28 图片

46:29 虽然不易阅读，但我有点喜欢，因为它

46:34 朴素

46:35 你明白我的意思吗？是的，我喜欢那种美学，因为我是个书呆子

46:38 但你可以直接说，你可以直接说，显然你

46:41 可以说，能不能把这个

46:43 变成一个漂亮可读的HTML文件，放在我的桌面上。这个很难读

46:51 它会做的，嗯。对吧，所以，如果你不喜欢

46:54 那样的话

46:55 随你怎么做。你明白我的意思吧？我喜欢这样

46:58 这种美学

47:01 但这就是如何从反思中走出来的办法，当然，知道吗，当然

47:06 当然了

47:06 我们也可以说，如果我们不想做这个

47:10 也可以说它推荐那个斜杠毕业命令，所以我可以

47:15 直接说构建斜杠毕业命令

47:20 对吧，这很有趣，这是你很多命令开始构建的方式吗？

47:24 ？

47:25 它就开始建议，然后你就说，去构建吧

47:29 嗯，我最初是自己动手做的，比如我在想

47:34 自己设计命令

47:35 但后来，我开始问代理，说等等

47:39 你觉得哪些命令有趣，此外可能有用的还有

47:43 另一件我想做的事是，我喜欢

47:46 在使用LLM时提升抽象层次

47:51 我的意思是我可以说，做一个命令

47:55 告诉我每天该关注什么，这就是我想的命令

48:02 但当我，另一种做法是，你可以退一步，然后我说

48:03 基于我的Obsidian库和你对我的了解

48:07 形成对我理解水平的认知，

48:12 考虑到我是个怎样的人，以及我正在做的项目，基于此

48:16 建议我应该使用哪些命令，能让我从初级

48:21 提升到更高水平

48:23 对，就像让它为我推荐命令，而不是我来推荐命令

48:28 然后我从中选择

48:29 看看这个，这就是代理基于它读了我的库后的想法

48:35 基于我做的笔记，对吧？所以看看这是啥，日记创意提取器

48:37 So look at this. This is this is it was this was the agent's idea based on what

48:41 it read in my vault

48:42 Based on the notes. I'm taking right so let's see what this is daily note idea

48:46 extractor

48:46 想法、见解和原创思维在每日笔记中积累细节

48:51 但很少升级成独立笔记，无法通过反向链接复利增长

48:55 通过反向链接

48:56 此命令扫描近期每日笔记，筛选最佳候选项并辅助

49:00 决定推广为想法或其他内容，对吧？

49:03 当然

49:05 事情就是这么运作的

49:06 扫描所有近期每日笔记，交叉参考现有

49:10 库，呈现候选项

49:12 选定想法升级为独立笔记时，若创建新笔记则在库根目录

49:18 创建该笔记

49:19 将笔记写成迷你论文或工作文档，捕捉核心观点

49:24 或起源于该每日笔记的问题与背景

49:27 并作为反向链接连接到其他库中的笔记

49:31 这些都是它捕捉核心观点或问题的方法

49:33 你可能看到这觉得，这不过是代理生成的文本

49:37 确实如此

49:39 但对我来说感受不同，因为我知道

49:43 我写了很多关于这些事情的内容，我了解，甚至

49:48 迷你论文这回事

49:49 这些词对我来说有特定含义

49:52 这真的太疯狂了，非常有上下文感

49:55 我知道它在说什么，因为我花了很多时间在这个工具上

49:59 也花了很多时间写作

50:00 所以，是的，我创建了它，这个命令也会创建它

50:03 对，就是这样

50:07 这真是不可思议，因为我只是随便做笔记

50:10 同时有一个并行代理在查看我的笔记

50:13 给我提供如何改进工作流程提升生活的想法

50:17 不仅可以给建议，它还能直接构建

50:20 东西

50:20 任务完成

50:22 我们就在这里有它，使用“/graduate”命令，我可以直接点击

50:25 它就会运行

50:27 真疯狂

50:29 如果我是 OpenAI 或 Anthropic，我一定会买下 Obsidian

50:34 因为它是缺失的环节

50:40 是的，真疯狂，这就是缺失的环节，事实上有人

50:43 顺便说一句，你说服我了

50:46 我已经下载了 Obsidian，我认为它是免费的，对吧？是的

50:51 开源的

50:52 我下载了，但还没

50:57 创建我的笔记库，因为我想

50:59 你知道

51:01 我知道它会很棒

51:04 我知道我会深入使用它

51:06 这实际上超出了我的预期，这真是

51:09 事实

51:12 就是

51:13 毫无道理

51:15 毫无道理

51:16 这事实——如果你认真使用大型语言模型来整理想法，并充分利用它们

51:23 如果你认真构建所谓的个人操作系统

51:28 If you're serious about building, you know, what people are calling a personal

51:32 OS

51:32 但你没有用像这样的去中心化笔记工具

51:38 用 Markdown 作为基础的

51:42 那么你就没有正确使用大型语言模型，至少没用到极限

51:49 没错，你没有充分利用它

51:53 所以我觉得困难的是

51:56 它确实需要

51:58 花费大量时间

52:00 并想要设置好它，需要

52:04 like

52:07 是的，这确实需要时间，而且界面太

52:12 令人畏惧，因为它就是一块空白画布

52:15 不像是“嘿，你应该在这写你的偏好”或者

52:19 你只能自己

52:20 自己想办法

52:24 是的，必须自己想出点子

52:26 但是

52:29 这还是非常棒的，因为即便和其他人合作

52:33 我们也得想办法给他们解释

52:33 我觉得很酷

52:37 that

52:38 现在我们可以跟这些代理合作，虽然还是得解释给它们听

52:42 但只需解释一次，因为一旦写进了文件

52:46 文件里

52:46 我们就可以随时引用这个文件，解释项目或偏好

52:50 它一直存在，也可以传入使用

52:53 文件就像是

52:55 完美的记忆，是吧？

52:58 人类有记忆，我们会回忆

53:01 但大量研究表明，我们记得的东西其实

53:06 和真实情况

53:07 完全不一样，例如

53:09 我们去密西索加理发时

53:13 我本以为那是我理过的最好的一次发型

53:18 记忆里是个好发型，但谁知道呢

53:22 那可能是我理过最糟的发型

53:25 现在

53:29 Obsidian 或你最终用的任何工具

53:32 如果我把它写下来

53:35 记忆，也就是 Markdown 文件是完美的

53:40 这样当我链接或调用它时，得到的就是完美的

53:46 嗯

53:47 数据点，关于这些文件的另一点是，你

53:51 希望？

53:52 它们基本上没有偏见，跟写下反思的人偏见一样

53:56 在那个时间点是这样，是的

54:01 这真疯狂，伙计，

54:05 是啊，真的很疯狂，而且这里面有各种不同的方面，

54:10 它，

54:10 还有它的隐私性，以及这意味着什么，

54:14 还有它的力量，事实上现在你可以用自然语言和这些

54:17 计算机工作，直接委托它们，

54:20 嗯，这里还有像我这样的人，正在使用这些

54:24 工具，试图搞清楚如何以这种方式将任务委托给代理，

54:27 还有像我这样更极端的不同类型的人们，

54:31 不断推动它们，

54:32 我只是觉得活在这个时代太疯狂了，因为我觉得我们可能，

54:37 潜在地，

54:38 正在见证人类与计算机关系的根本转变，

54:44 我真的很高兴能在这个时期活着，

54:47 我很好奇这一切会如何展开？

54:50 嗯，酷的地方在于？

54:52 99.99%的人不会花时间，

54:59 真正设置这样的东西并将其变成日常生活的一部分，

55:05 而且

55:05 这个，

55:07 可以说是领先的一种状态，

55:12 在更加高效、

55:14 快乐、更健康、

55:16 更赚钱的职业生活方面，就是使用这种结合大型语言模型的工具，我

55:22 同意，

55:22 我不是说今天就去下载Obsidian，我跟他们没有任何关系或什么的，

55:27 但我是说，

55:29 选一个，我们似乎都应该做的事情，我是在给自己这个建议，

55:34 就是说，

55:36 我再也没有理由不去写笔记和进行反思了，

55:42 对，用Markdown文件。是的，在一个大型语言模型使用Markdown文件的世界中，Markdown文件，

55:48 才是氧气，

55:50 人们认为tokens（令牌）是氧气，是的，但其实不是，

55:55 Markdown文件才是记忆。想想人类是什么，

56:00 是的，你知道，

56:03 人类是什么，

56:06 是人的能量呢，还是它的记忆？你知道我们曾被称作，

56:11 你懂的，

56:12 这就是个哲学问题，也许两者兼而有之，

56:17 但，

56:19 我觉得，

56:22 Markdown文件确实很有趣，

56:27 它们被低估了，

56:30 为了拥有真正的，

56:34 计算机体验，在如今这个时代，确实，

56:38 这里发生了一些事情，

56:39 某种根本性的转变，确实很棒，

56:42 是啊，我的工作有点糟糕，你知道，我是在实时学习，对吧？

56:46 而且我甚至没有合适的词汇来解释这一切，

56:49 是的，

56:50 现在它们太高级了，伙计，我不知道，我正在尝试，

56:54 实时搞清楚，

56:55 这就是为什么我觉得，比如说，我展示一些东西，对我来说，

56:58 我会做一些事情，或者我会看到一些东西，我的朋友们会笑我，

57:02 因为我会坐在电脑前，

57:04 痴迷地思考，我想那是因为我真的很喜欢电脑，

57:09 我无法相信，

57:10 这是真的，

57:13 我可以只是像小时候一样在电脑上做笔记，

57:17 然后，

57:18 突然间，这个代理可以扫描这些笔记并构建东西，

57:23 正因如此，

57:24 它能连接我永远看不到的模式，太疯狂了，伙计，

57:28 太疯狂了，

57:30 归根结底，你说得对，它不过是一堆相互关联的，

57:34 Markdown文件而已，

57:35 是的，

57:36 酷，伙计，感谢你，我不知道你能不能看到我现在的思维，

57:40 我彻底惊呆了，

57:41 谢天谢地，

57:44 我想对你做对的那件事，我每次都这么说，伙计，

57:48 但我会一直说下去，永远都说给你听，

57:51 我真的非常非常非常感谢你所做的一切，

57:54 我觉得你的模式识别和模式匹配能力，

57:58 真的是被严重低估了，我觉得你做的很多事情，

58:02 如果不专心注意，是很难发现的，

58:05 我只想说，感谢你所做的一切，

58:08 你总是在节目中带来新的声音，我看到了，真的很感谢，

58:12 真的很感谢，

58:13 能认识你真是一种荣幸，是的，感谢这个机会，感谢你，伙计，

58:16 我感激你，Vin，你是传奇。我会附上链接，

58:18 告诉大家去哪里关注，

58:21 被严重低估的网络名人Vin，

58:24 在X（推特）、他的YouTube节目，

58:26 播客的节目说明和简介里，你都可以去看看，

58:30 各位，请多玩玩这些工具，

58:36 告诉我你的想法，也请告诉Vin你的想法，拜托，

58:41 然后，

58:48 我会恳求你再次回到节目，

58:49 希望你一定会回来，伙计，谢谢你，谢谢，

58:51 And I hope you come back on again for sure man. Thank you. Thank you

0:00 This is obsidian and obsidian is this little tool that people are using as

0:05 their second brain

0:06 But what's really cool about it is they're pairing it with Claude code and they

0:10 're getting

0:10 Crazy results out of it. It's literally a game changer now

0:15 I've been slow to adopt obsidian because to me it's been a little daunting to

0:19 look at so I had my friend

0:21 Vin and he clearly explains what obsidian is how to use it with Claude code

0:28 How to set up these commands that really drive the most out of Claude and all

0:34 the LLM's and it's an incredible

0:35 Episode like a really game-changing episode because I think that people who

0:40 understand how to use obsidian and how to use Claude code

0:43 Together, they're going to be able to live happier healthier and wealthier

0:48 lives

0:48 Why because it gives incredible ideas to you on tap

0:53 So I know that the people that stick around to the end of this episode

0:57 I think that for a lot of them

0:59 It's going to absolutely change how they use AI and it's gonna be a super

1:04 impactful way because you're gonna get better ideas at the

1:07 Right time the right moment and it's gonna make you happier healthier and

1:11 wealthier. Enjoy the episode

1:21 I've got my dear friend vin also known as internet vin on the podcast. I

1:26 literally begged him to come on I

1:28 begged him I begged this man to come on and

1:33 To teach us a very specific thing Vin by the end of this podcast episode. What

1:39 are people gonna learn?

1:40 I want you to have an understanding of how you can use Claude code and obsidian

1:46 as a thinking partner

1:47 I want you to have an understanding of how you can stop

1:51 Having to explain things to agents over and over again and just pass specific

1:55 files in and I want you to understand

1:57 How you can use obsidian and Claude code to notice?

1:59 Things about the way you think that you would not have noticed on your own

2:03 without these tools

2:04 All right from your lips to God's ears. Let's get into it

2:08 Okay, so first is like what is Claude code?

2:14 So Claude code is this like agent that you can use

2:17 in a command line interface, so it's just basically this tool you can use that

2:23 you can control your computer and

2:25 You can use it through natural language, right?

2:27 So I can say make a file or make a file on my desktop

2:32 That says hello Greg and plain text

2:36 Right and it's gonna go and do this

2:40 That's really cool. That's that's something that's that's new. That wasn't

2:44 possible before before this

2:46 I had to go to the desktop open some text editor and then like create that file

2:51 , right?

2:51 And now this file is on my desktop

2:53 So I can say open the file. There we go. Hello Greg. That's crazy, right?

3:00 Now

3:03 What's interesting about this is if you have this agent that can like control

3:06 and do things on your computer

3:08 That means that whatever you can describe to it like it can start to do and so

3:14 when you if you describe a project to it

3:16 Or you get into these long conversations with an agent

3:19 It can do more and more complex things the more information it has the more

3:24 complex things

3:25 It can do but the problem is that if you have to let's say, you know

3:28 Like I write some super long description about a particular project or I have

3:32 like an hour conversation with this agent about a particular

3:35 Project, it's like I don't want to have to create a new session to explain that

3:40 all

3:40 I don't want to have to explain that over and over and over again a

3:43 Lot of people are using like Claude or Chachi PT on the web and it has things

3:48 like memory

3:49 But you you can't like control you don't know what's in that memory, right?

3:53 You don't know what it knows and what it doesn't know and so there needs to be

3:57 some way of like, you know

3:59 Passing information into these agents that is easier and faster and the better

4:04 information you can give it and the faster

4:06 The information you can give it the more stuff if it can do for you and the

4:09 better the faster you can delegate to it

4:11 Okay, so

4:14 now

4:15 Even if I let's say let's say I had like, you know, let's say I wrote like a

4:19 big project description here, right?

4:21 Create a file that describes, you know a project

4:25 About

4:31 To-do list app that

4:35 is very minimally designed and

4:38 reads from

4:41 all of my calendar

4:43 my messages and

4:45 My slat and slack and

4:49 Interprets it into a task list that of tasks

4:53 That it thinks that I should do. I don't know some some idea, right?

4:59 So now this this is a file that could be on my desktop and

5:03 what I can do when I use cloud code is I can reference that file and pass it in

5:07 whenever I want and

5:08 why that's important is because

5:11 It's the context, right? Yes whole game is

5:16 Feeding the beast good context. Yes

5:20 Exactly, and I don't want to have to do this over and over again and and when I

5:25 work on this over days

5:27 I'm not going to remember like what we talked about right so I want some kind

5:31 of file

5:32 That I can like pass in. Oh, sorry Greg. What's that?

5:35 Yeah, and that's that's sort of the problem that a lot of people are facing

5:39 with cloud code is like they're using it and then

5:42 They're saying well, it's it's okay. It's not like game-changing

5:45 And the issue is they don't have they're not they're not feeding the right

5:49 context at the right time. Yes exactly and so here's

5:53 So here's like a project

5:56 Description that it wrote and obviously I can pass this in and this is like a

5:59 this is like a general one that I just created

6:02 But you can make these like very complex. You can build them into like robust

6:05 files right over time

6:07 so we know that

6:09 Cloud code can create files and it can repeat and it can

6:11 Read files, right? So now I can say let's say I created a new session. So here

6:16 's a new session and now I can say like I

6:19 Want to work on

6:21 This project and I go here and it's gonna be

6:26 To do - app here it is boom now. I didn't need to explain the file again. I

6:30 need I didn't need to explain the project again, right?

6:32 So it's gonna read this file and it's gonna start like

6:36 You know that saved me a lot of time great project before diving in a few

6:39 questions to scope the first session

6:40 So that'll continue. So now what is obsidian?

6:43 right obsidian is this tool that it's kind of like an interface that sits on

6:50 top of a

6:53 Collection of markdown files, right? So here like this is reading a markdown

6:58 file. How I use obsidian story development, right?

7:01 I have a daily notes

7:04 This is my daily notes. This is a this is also a markdown file

7:09 I should do my own fundamental analysis into think into how things stay pure

7:14 when they grow and become more mainstream, right?

7:16 This is just a file that I have I have

7:20 You know like a file on Greg Eisenberg that I haven't put it. That's weird.

7:25 Yeah

7:25 And so I make files like notes of things that I'm learning from people and

7:30 stuff like that, too

7:32 So I have different

7:34 Files for everything right and the interesting thing that makes a vault which

7:38 is obsidian what obsidian interacts with this whole thing

7:41 It's called a vault what makes it different than a folder is that?

7:44 Obsidian is not only interacting with just like a you know

7:49 a folder of files

7:51 But what it does is it also allows you to make inter

7:55 To connect relationships between files so I can say today. I am on a

8:00 podcast with

8:03 Greg Eisenberg now this file is linked to that Greg Eisenberg file

8:08 Super interesting super interesting and so when people like like like people

8:13 There's a lot of people who really really like of using obsidian and tools like

8:17 obsidian because of this

8:19 ability to form interrelationships

8:21 This is unique to just having a folder a folder on your computer cannot show

8:26 these interrelationships and

8:28 so it gets really interesting when you start to

8:32 Keep making these like interrelationships over time

8:35 right and so what happens here's a little visualization and so in that here

8:41 these are each one of these circles is a file and

8:45 In and it's showing how it's like connected to all of these other files where I

8:50 've written things about so here's like personal agent

8:53 infrastructure

8:55 Right and so I could look you know

8:58 And I guess I should also add just kind of a comment on this and what was

9:01 difficult doing this demo

9:02 There is like so much personal information in here because this is like my

9:06 personal thing

9:07 So I don't even know like what's gonna show up on the screen here, right?

9:10 But that's part of doing demos like this, which is which are kind of weird and

9:14 interesting

9:15 But you can see personal agent infrastructure link links to like a Gentic AI.

9:19 There's like a link here to telegram

9:21 There's a link here to like Toby

9:23 With the founder of Shopify. There's a link to like presence log quad bot

9:28 You know and then here's like I have a podcast to call the other stuff and like

9:32 you'd see I'm obviously doing a lot of like

9:35 Thinking about that a lot right and so I

9:40 Can also let's say if I go to Greg Eisenberg and I go to local graph

9:44 So here's like all the times I've written about Greg Eisenberg

9:48 Right notes on time constraints how I use obsidian, which is just kind of

9:52 interesting

9:53 So if I'm less too then I love you, too

9:56 If I'm listening to a show and I'm picking up different patterns

10:00 I can I can reference that back to to Greg. So that's really interesting, but

10:04 here's the thing

10:06 The reason why people love obsidian is because of these interrelationships the

10:10 idea that you could open a file and then you know

10:13 I just open this file and then I'm like, oh interesting. I mentioned Greg Eisen

10:15 berg. I can click that and it goes to that file

10:17 That's interesting, right? It shows it works more to it. It works more like the

10:22 way your brain works your brain connects these patterns all the time

10:25 Here's the question. Yeah, yeah, so I see why it's interesting

10:31 But how does this get me better output?

10:36 Exactly. Yeah

10:37 So the next thing is obsidian

10:39 released this new tool called obsidian cli and

10:43 what that allows you to do is it allows you to use cloud code and

10:48 It can go and it can read all of the files in your obsidian vault, which is a

10:53 folder of text files

10:55 But with the obsidian cli it can give cloud code not only those files that it

11:00 can read and access

11:02 But it can also give cloud code information about the interrelationships of

11:08 those files

11:09 So you can see so cloud code can see that Oh, this file is connected to this

11:14 file and this file and this file

11:16 And that gets very interesting in terms of what cloud code can understand about

11:20 you and what cloud code can

11:22 understand about

11:24 All of the relationships between the things that you're working on it can start

11:28 to surface patterns about what you're thinking about that

11:31 You are not

11:32 Seeing for yourself some idea that you might have been writing about for a year

11:35 in this vault

11:36 It could be a latent idea and it can just immediately say like hey

11:39 Did you know that you've been writing about this same pattern and startups or

11:43 in this particular project?

11:45 You're working on in every single note

11:47 You're making across these different domains and and seeing that for the first

11:51 time can be like a huge light bulb effect

11:53 It can cause like huge progressions and you're learning and you're

11:56 understanding and your point of view on the world

11:59 But also in what you're working on

12:01 So I've written now I wanted to demonstrate how that actually works in terms of

12:06 how I can pass

12:07 Information into an agent that would be impossible without obsidian and cloud

12:11 code

12:11 So here's some commands that I have that I use and I don't want you to be

12:16 afraid of like all this stuff

12:18 I know this can look intense

12:20 but

12:21 here here's what I'm

12:23 Here's what I've got some commands and this is just terminal that I've created

12:27 and I'm running it in obsidian

12:29 You don't need to use this you can just also do this in your own terminal

12:33 session on in whatever tool you want

12:35 But I put it in obsidian because I want to see it all together

12:38 And I wanted to show you the ways in which you can like integrate and customize

12:41 this environment

12:42 so

12:44 Here's a cool thing. So context slash context load full context about my life

12:50 work and current state reads

12:52 context files daily notes and follows backlinks to build a complete picture. So

12:56 I'll just show you that right here

12:57 So like let's say I open a new session

12:59 in in a

13:02 In cloud just on my desktop and now it's like I'm about to work on something

13:06 But before I work on it, I can just type context demo now

13:09 It's gonna read a whole bunch of files about where I'm currently at done

13:13 Like I've already preloaded I've already preloaded in all this context now so

13:19 you can see it's gonna start reading all these files

13:21 It's reading a read me. It's reading context about new which is a media company

13:25 that I'm working on. It's reading about other stuff

13:27 It's reading my personal the other stuff is my show

13:30 It's reading a personal workflow context and so I don't have to worry about it

13:34 not knowing the key information that I wanted to know

13:37 I just did that one command and now it's gonna get all that information done

13:40 So I can use slash today, which is a morning review pulls calendar tasks

13:45 I messages in the past week of daily notes and into a prioritized plan for the

13:49 day. Why does this matter?

13:50 Well, okay sure you can set up an agent and give it access to your calendar and

13:55 your tasks and I messages and things like that

13:58 But it's miss it's that's that doesn't have all of the information about what

14:01 you're thinking about and why if I'm writing daily notes about some

14:05 Particular technology or project or thing that I'm interested in does my

14:08 calendar reflect actively like does it match the subjects?

14:12 I'm actually writing about

14:14 If an agent has that context you can it can it can more effectively give you

14:19 Information about what you should do or not to or it can more effectively make

14:24 decisions on what should be in your calendar or not in your calendar

14:27 Here's another one slash closed day

14:29 and if day processing

14:32 Extracts action items surfaces vault connections checks confidence markers

14:36 needs to be updating

14:37 So I have a bunch of hypotheses that I think about and I give them a confidence

14:42 rating

14:42 This is an idea. I'm working on I feel very solid about it. Here's another idea

14:46 . I'm working on I'm not sure about it

14:48 So these are like

14:50 daily operations things, but

14:52 This is what I use obsidian for the most, which is thinking tools. I

14:58 Really really really like working with LLM's as a thinking partner

15:03 It's my favorite way of using LLM's

15:05 I know people like to use agents and LLM's to build things

15:09 But I really like using them to think alongside me and build when I feel like,

15:14 you know

15:15 I really have a novel way of viewing things. So

15:17 Let's see here

15:20 So ghost

15:22 Here's he's a command to have it answers the question the way I would it builds

15:27 a voice profile from the vault

15:29 Rights in that voice then evaluates the fidelity

15:31 So I can just say what do I think of AI and I'm going to show you this

15:36 Challenge topic if pressure tests current beliefs using the vault's own history

15:41 finds contradictions counter evidence and shifts in thinking

15:44 Why does that matter?

15:46 Well, if I want to make sure that I'm continually developing as a human being

15:50 and as a as a as in my skills

15:52 I want to make sure that you know the POV I have is an overly biased or limited

15:57 . So this can challenge me

15:59 emerge

16:01 surface ideas the vault implies but never states

16:04 Conclusions from scattered premises unnamed patterns unarticulated directions.

16:08 This is super super useful because a lot of times

16:12 You know, I can be stuck just surfacing ideas in a in a in a lot of different

16:18 ways

16:18 Like for years and just having someone say a simple thing to me that just says

16:23 hey

16:23 This is just naming the idea. Hey, did you know that you keep circling around

16:27 this pattern?

16:27 huge breakthroughs

16:30 Slash drift it compares it compares my stated attention intentions against

16:36 actual behavior over 30 to 60 days

16:38 Surfaces what I am avoiding

16:41 Ideas people on this podcast the listeners will probably like this one deep 30-

16:45 day vault scan with cross-domain pattern deck

16:48 Detection and graph analysis to generate generate ideas across all domains.

16:52 This gives me

16:53 Not just ideas on like things I should work on like it gives me ideas for tools

16:57 and things like this

16:58 But it also gives me ideas. I'm like films. I should watch

17:01 Products I should buy again all influenced by like like things. I'm writing

17:05 about in my vault

17:06 Trace tracks how an idea has evolved over time across the vault. So let's see

17:13 some of this stuff

17:14 the trace demo

17:17 So I did this one already and the way this would work is I just like create a

17:21 tab here and I could just be like Claude

17:25 Trace and I had to create demo versions of all of these

17:28 Commands because of how much personal information is in my vault, but still I

17:33 don't even know like I can't even control

17:36 What is going to show up on the screen and I I have just a dumb question like

17:40 all those commands that we saw

17:42 is that

17:44 Command say you created or is that what obsidian created that those are

17:48 commands that I created and

17:52 You can create them very easily by just asking Claude code to create a specific

17:56 command

17:56 Mm-hmm. We can include

17:59 I'll include in the show notes in the description

18:02 Just like a link where people can click to get access to some of VIN's

18:08 Skills if that's if that's yeah for sure. Yeah, so you could see here

18:15 What I did was I just typed slash trace demo demo is just because I'm showing

18:18 this publicly how I use obsidian and

18:21 so what this does is it tracks my I

18:24 The way I've developed neat the way I've navigated this idea through my vault

18:29 So I have an example here. It's really interesting

18:33 So here I did it and I and I and I let it run and what it's doing is as it says

18:39 like all trace

18:40 How your relationship with obsidian has evolved let me start by building a

18:43 vocabulary map and searching across a vault

18:45 so it

18:47 When you have a like so now it starts going through the vault reading all of

18:52 these different files and it can see all of the files

18:54 That are connected using the obsidian CLI. This is a lot man

18:58 This is this is something that I would never be able to do on my own and to

19:02 read all of these files

19:04 Just know what how they're interconnected with each other

19:06 This is just not possible for me to do as a human being and then it spits to

19:10 this out

19:10 I now have everything I need to construct this trace. Here's the full evolution

19:13 trace how I use obsidian

19:16 First appeared January 11, 2025 timespan 13 months. That's in relation to this

19:22 vault specifically meaning all the files in here

19:25 It says

19:27 Pre-volts December 2024 the essay how I take notes in the west end of Toronto

19:32 dated December 1st

19:33 2024 describes a complete system where obsidian plays no role at all

19:38 The system is audio dumps via Mac whisper LLM dialogue loops canopio for

19:42 spatial mapping physical notebooks for cognitive friction

19:45 And future land for tracking arena for fragments the pipeline is capture

19:48 process crystallization

19:50 This is the baseline obsidian is not in the picture discovery and set

19:54 skepticism January to May 2025

19:56 The first daily note in the vault raw excitement mixed with uncertainty

20:01 I could probably drop transcriptions in here too as a way of storing them

20:06 The current theory is that in terms of note-taking by directional linking is

20:09 not that useful, but I don't know

20:11 then

20:14 It says the chosen tool

20:16 Originally I was back ring link. It says here

20:20 How I use obsidian note a pivotal realization about back linking

20:24 Originally, I was back linking to general terms podcasts or physical fitness or

20:28 filmmaking

20:29 I'm realizing it's not the most useful way to use obsidian

20:32 The most important thing to do is to create notes for each of my patterns

20:36 Theories projects of perspectives and get them documented out of my head and

20:39 then link to those notes

20:41 so it's just it's just like pulling from

20:44 Things I've written and it's forming this historical at this history of this

20:50 concept

20:51 And I could do this with anything. I can do this with like startups

20:54 I can do this with a particular project with foods my relationships like a

20:59 hobby

20:59 Anything right and then phase four January 2026 a month of explosive building

21:05 Everything still required and then like everything still requires me to

21:09 actively prompt and manage each section

21:11 Session the next unlock is figuring out how to get agents tasks to run

21:15 automatically

21:15 The friction is no longer with obsidian itself, but with the boundary between

21:19 the vault and agent execution

21:20 so you can see I'm really pushing myself, right and

21:24 It's cool. This is a this is a very

21:28 useful thing for me to understand how my

21:32 use of this tool is evolving and it's just I think it's just absurd that I can

21:38 just be making notes and then

21:40 about all of these different things through my life like even as a parent I

21:44 Can I can like reflect on the different things? I'm learning. I just think this

21:49 is insane that and

21:50 A computer can have this much information about me and surface these patterns

21:55 I would not be able to do this on my own and this fast and what a great tool it

22:01 is for me

22:01 Like like I can just like now write in here and I'm you know as I'm thinking

22:05 about things and it gives me ideas, right?

22:08 about

22:09 My life and the projects that I'm working on

22:11 So I can say, you know, it's interesting the way that my relationship with obs

22:19 idian has evolved over time

22:21 It makes me think a lot about

22:25 The way in which my relationship with computers has evolved

22:32 over time since I was a kid

22:35 to now

22:37 it's interesting

22:39 How these things just happen and compound over time and we don't really realize

22:46 it?

22:46 So it's just like a no, right? It's an idea. So that's an example of something

22:51 so I think a part of getting good at obsidian sounds like

22:56 Reflect, you know inserting reflection into your everyday life because a lot of

23:03 people

23:04 You know, we're moving from meeting to meeting we're busy we're parents

23:07 You know

23:10 We grow up and we of course write things down in notebooks and stuff like that

23:14 But I feel like as we get older we actually write and reflect less and less.

23:19 Yes

23:19 You know, how do you?

23:22 How have you been able to insert reflection into into your life?

23:27 Yeah, I think

23:30 For me, it's really about there's there's two reasons that I think reflection

23:34 is interesting and making notes a lot are interesting

23:37 one is

23:39 That it's great to be able to look back on them like for me now

23:43 Obviously, I can use an agent

23:44 but for me to go back and see these notes and

23:47 Realizing that oh like I'm a person that's continually changing my skill is

23:51 continually changing projects are continually evolving

23:53 It's it's just a mate is an amazing part of life to be able to reflect on how

23:58 things are changing over time

23:59 And how you are changing over time and how the world is changing over time

24:03 But the other thing is that there's like a functional reason to a reason

24:07 I like to make notes a lot is because that's how I generate ideas when I like

24:10 get when I sit down on my computer

24:12 And I write things down. That's where ideas come for example. This thing. I

24:15 just wrote here

24:15 It's just a quick note. I'm just doing it in real time. I'm just making it up

24:18 right now, but

24:20 By writing it out I feel like I internalize it more and I like having good

24:25 ideas

24:26 I like progressing so because I like having good ideas and I like progressing

24:30 Writing is how I do that and so I think

24:33 To you know if you want to cultivate

24:36 Like a writing as a habit. I think first you have to

24:40 Connect it with the idea that this is how you progress and this is how you

24:44 generate ideas and this is how you have

24:46 form original ways of thinking

24:48 the other thing I'll say is

24:51 that

24:52 Writing right now is a big way of how you delegate things to agents

24:56 That's like a whole new that's a whole new aspect of it. So if you can develop

25:01 a writing habit

25:02 You have a lot more context that you can pass over to an agent which then

25:07 Dramatically, I think increases the amount of like things you can delegate and

25:11 the amount of things you can build

25:13 I hope that was a good answer

25:15 How does

25:17 How does this relate to open claw because if you think of open claw

25:21 It's essentially

25:23 You know at the best case an extension of you. Yeah, that could go and do

25:30 things, you know

25:32 independently

25:35 Slash based on your guidance. So how can you use commands obsidian and

25:41 Open claw and reflection

25:44 harmoniously

25:47 Yeah, so I think if you look at like here's an example of one command that I do

25:52 It's just like a schedule command

25:54 and so

25:56 What I asked this thing to do as I said schedule

25:59 I said can I take a meeting with Greg Eisenberg today at Feb 20 at 2 p.m

26:03 right and

26:05 What this does is of course you can look at my calendar and stuff like that

26:08 But it's also going to look through my daily notes

26:11 It's gonna look through what I care about and then it's gonna give me some

26:14 perspective. So it says

26:16 Your day is stacked. You're already recording on Greg's podcast this morning

26:20 followed by a team lunch outing and meeting with Peter and Vince

26:23 Your Feb 17 notes show the Greg episode has been top of mind the bulk as a

26:26 dedicated Greg Eisenberg note

26:28 No, not a - so the recommendation is no not at 2 p.m

26:31 But you might not need a separate meeting at all. Yeah, that's that's actually

26:35 the correct answer. How does that relate to open claw?

26:37 well

26:39 Open claw is just like autonomous agent that can go and do things if you

26:44 Set it up to do that like it can do things without you having to prompt it all

26:48 the time

26:49 It can just go and make decisions and build things for you

26:51 on your own

26:53 So now what open claw can do is in the same way

26:57 I just did this command open clock and do this on its own as well and

27:00 It can go and read my vault find connections and then make decisions

27:05 On behalf of me with like a deeper understanding of me and now instead of like

27:10 managing an agent or talking to another

27:12 Human about working on something. I just focus on managing this vault. This is

27:16 like the new source

27:17 I just continually try and make it so that this vault has all of the

27:21 information needed so that I can delegate to an agent and

27:25 The agent can just pull from this vault source and make decisions and if it's

27:29 not making the right decisions

27:30 I'm changing something on the vault. I'm not necessarily working with the agent

27:34 specifically

27:35 Kind of kind of that's my speculation on that subject

27:42 I think it's very interesting. Yeah, I think

27:44 One thing that worries me a little bit about it is if obsidian is really your

27:49 second brain

27:50 giving open claw access to your second brain is

27:54 Scary yeah scary and I would say that is the

28:00 Fundamental

28:06 The weird element of this technology I would say and I have purposely given

28:12 Obsidian I mean sorry a cloud code or any agent access to a lot of information.

28:19 I've purposely done that because my

28:21 Relationship with this is I want to understand what these things are and I want

28:26 to understand what they're revealing about

28:28 You know how our relationship with computers is changing

28:31 but

28:33 It's weird. It's like you have to really think about how much information you

28:37 're sharing with these

28:38 agents and whether that's the right decision or not the right decision and

28:43 I think

28:44 It's going to be very interesting to see how privacy as a concept evolves and

28:48 changes and what we fight for or don't fight for and like

28:52 the in the future of

28:53 Our society in our world even with every one of these commands. I had to create

28:58 a new version of them a demo version

29:01 So that I wouldn't reveal too much personal information while I'm on screen on

29:05 this podcast and even then it's like a it's a toss-up

29:08 You know I could type the demo version, but who knows what's going to be shown

29:11 on screen, you know

29:12 What other commands do you want to show?

29:16 so

29:20 There's

29:21 Connect

29:22 Which is allow it allows me to take two domains and connect them using the

29:26 volts link graph

29:27 so I can just say I

29:30 Did one here?

29:32 And I just asked it to

29:35 Connect

29:38 filmmaking in world building and so it goes through and it reads all these

29:41 different files and

29:43 Then it can start to say okay. Let's let's connect these two concepts

29:47 so notes and filmmaking

29:50 Notes and filmmaking's neighborhood. So I was like 35 film watch list my first

29:55 meeting with Toby

29:57 notes in the world building neighborhood and

30:01 The world building essay

30:03 New as a media company. So these are different things. I'm thinking about so

30:07 bridge one the interview portal and the constructed world

30:10 In filmmaking if I noticed something specific and ask the question about it

30:14 It would open a portal into a person's internal world, which is often a vast

30:18 universe of concepts and beliefs and visions

30:20 So world building essay I

30:22 Want my blog to show you what I value what I believe what I worry about like a

30:27 tomb from ancient Egypt

30:28 I want my blog to be a place that you dig up and examine long after I'm gone

30:32 These are things that I've written and I can start to see how these ideas

30:35 connect together

30:36 Bridge to always on documentary equals continuous world building

30:40 Always on documentary is a creative strategy where companies continuously narr

30:45 ativize their characters pursuits conflicts and visions through documentary

30:49 So these are like things that I'm writing about and it's showing me the ways in

30:53 which these are connected I

30:54 Think this can get

30:57 Very interesting depending on the kinds of things that you're willing to

31:01 connect together

31:01 You could get probably pretty crazy with it as well depending on what you're

31:05 writing about in your vault

31:06 I could connect like shawarma and startups if I wanted to for example and see

31:10 the kind of connections that are coming between these things

31:12 again

31:15 Really interesting because all of this is happening super quickly and I don't

31:18 need to explain any of this to an LLM

31:20 I can just type something it's like

31:22 slash connect filmmaking world building

31:26 The the a lot of the examples you're using is personal reflection. Yes. Do you

31:32 think about?

31:32 You know

31:36 For example, no taking in meetings like maybe you have like granola or Gemini

31:41 notes taking

31:42 you know

31:45 taking notes or and in sort of putting it into obsidian or

31:51 And by the way when I say notes then those could be meetings that you're not

31:55 even in they could just be like

31:56 You know tommy met with Vince and they had this meeting and I want to put it in

32:01 here. Yes

32:02 How do you think about that?

32:05 So that's a really good question. So I think you can use these vaults however

32:09 you want to use them in terms of like

32:10 You could put any text you want in here if you want to put your granola meeting

32:14 transcripts in here

32:15 You can put them in here and you have to just make sure you know

32:18 Maybe you're doing something like this, right? So you're just like meetings and

32:21 then you're like, okay

32:22 These are this is you know project one and then every time you do a meeting

32:26 You take your granola notes, and you just put them in here, right? So you're

32:30 like meeting Greg Eisenberg

32:32 then

32:34 plus then and then

32:36 you know

32:37 That follows created now and you can just drop your Trent

32:40 You can just drop your meeting notes in here

32:42 And now that's in the vault and then you can pass that into the agent or the

32:45 agent will discover it, right?

32:46 Especially if you start tagging like oh, I'm gonna tag this back to you know,

32:50 like my podcast or something there now

32:52 It's connected so now the the the agent it has like more context and now it

32:57 knows that Trent this transcript is related to this other file great

33:00 I think that's up to you

33:02 I think the way that you the amount of information that you put in here is up

33:05 to how you want to use the vault and how you want to

33:08 Delegate to things to agents and maybe you even want to create different vaults

33:11 for different purposes for me

33:14 for me, I use LLMs and agents as a way to

33:18 Increase my own level of understanding of subjects

33:21 So I use it for a lot of reflection and things like that

33:23 So I don't want an agent to write into the files

33:27 Like I could easily get it to do that like I could just say like even here

33:31 I've asked it to write a description of some commands that I can talk about

33:35 today

33:35 But I don't I don't I don't want it to make a file to do this because I want to

33:39 control all the files in my obsidian

33:41 Vault because I always wanted to pull from what I think about things

33:45 Right not what it thinks about things and if it starts making its own files in

33:49 this vault

33:50 Then I don't know like is like when it's finding these patterns is it finding

33:54 patterns about things it's written or is it finding patterns about things?

33:57 I've written so I create a rule for myself, which is like a strict separation

34:02 between these things

34:03 I only wanted to write things on the side here and then I will take that and

34:08 and and and write what I think

34:11 Should be included

34:13 Right. Yeah. Yeah, go ahead. No, I was just gonna say like

34:17 You know, I could see the power of just using it for your own reflections

34:24 I can also see the power of

34:27 You know AI is really good at

34:30 you know

34:32 Going out on the internet finding information based on trends and stuff like

34:36 that distilling it in a way that you want and

34:39 Having that being put into your world is also interesting. Yes

34:44 totally, I think that and also like let's say if you asked the obsidian if you

34:51 asked the

34:51 Cloud code to go through your obsidian file and generate ideas

34:55 which

34:58 You know ideas for tools that you should build well, then you can just like say

35:01 okay cool

35:02 If that's if I have an idea for a tool that I should build just generate a

35:05 description of that and then just build the tool

35:08 So exactly

35:10 Yeah, I want to show this one

35:13 So that it's like less on reflection

35:16 Like for example, like, you know, I'm a I built this thing called idea browser.

35:23 com and every single day

35:24 We give this validated startup idea like someone theoretically can go and

35:29 You know grab that information put it in an obsidian vault and then based on

35:35 that

35:36 Basically, you know help them build the actual thing, right? Totally. Yes

35:41 So I want to show you this because I think it'll it'll really make it

35:45 They'll take it out of the realm of reflection and into the realm of building,

35:49 but the only issue is this takes a bit so

35:51 That's okay. Yeah, so the other thing is that with all of these like with these

35:58 commands another pattern that I'm noticing is that they take a bit

36:01 Because it's reading so many files and I would say that's a big difference

36:05 between using obsidian and

36:07 Like are using cloud code with access to this obsidian vault that I'm noticing

36:12 is all of my

36:12 Requests are taking way longer and it's just because it's reading so much more.

36:17 So like look at this one

36:18 So this is ideas demo. So I'll run a comprehensive ideas generation

36:23 Let me start by gathering vault structure and context in parallel and then if

36:27 you look it's really interesting to see what it's doing

36:30 right, so

36:31 It's like obsidian orphans

36:33 Right, so it's it's it's it's like I guess orphans are like files that are on

36:39 their own not connected to things

36:40 Right. So that's interesting that it knows that

36:42 Obsidian dead ends obsidian resolved obsidian tag counts. So it's just trying

36:47 to it's trying to figure out like some connection between all these things

36:50 And then it says okay daily read so it's reading my daily notes

36:55 Then it found this file called new context, which is new is this media company

36:59 I'm working on then

37:00 It's like read file the other stuff context that that's podcast. The other

37:04 thing. I would say to you guys is I do manage

37:07 I write I create context files for projects that are pretty extensive. I'll

37:13 show you I

37:13 Was I didn't know if I was gonna show this because it's very personal but like

37:18 for the other stuff

37:18 Look at this other stuff working context. What shifted recently?

37:23 Front loading profile by traveling, San Francisco, New York City to record

37:27 guests

37:27 This is super personal stuff

37:29 But what is the other stuff the format core beliefs of the show researches the

37:34 foundation solid the best conversations feel like discovery solid

37:38 You know, here's a team that's working on it. And so what happens is this again

37:44 very personal, but this is

37:46 Context that it just pulled in so now it knows who's working on my podcast

37:53 What are the what are the recent hypothesis hypotheses? I'm exploring

37:56 And it just got that information. That's just like one of the things that did

38:01 personal workflow context

38:02 super personal file, but it shows like

38:06 you know, like

38:09 What my daily schedules is like, you know things that I have to do like in my

38:13 personal life

38:13 So it's like pulling that how I like to work how I don't like to work

38:17 personal agent infrastructure

38:20 There's another thing. Let's see what happens if I pull this up

38:24 This is a project in which I want to take a step towards increasing my personal

38:28 infrastructure workflow delegation

38:30 However, you want to describe it with agents understanding what it means

38:32 delegate to agents more and more

38:34 implementation approach

38:36 So this is like what I'm writing about

38:39 The file about how I'm thinking about using agents personally and again, that's

38:44 an example of one of the files

38:46 It's reading. That's just one of them. So you saw like the other stuff the

38:49 personal workflow

38:50 And it's factoring that all in to this task. I've asked it to do which is

38:54 generate ideas for me

38:55 Gathering data from your daily notes calendar and volstructure. This takes a

38:59 moment since it's pulling from multiple sources again one of the

39:03 Things with this is that it's just going through a lot of information man a lot

39:08 of information

39:09 So it takes longer, you know, it's already been going for five minutes

39:13 Right, and so that's that's something I'm noticing, but

39:15 For me, that's what I want. I want that I want I want a response from LLM's

39:22 that is very very contextual

39:24 To the things that I'm writing about and I think a lot and I think that's how

39:28 me and an agent can work best together where I just focus on

39:32 continually

39:34 noting that my could the where I'm currently at in terms of the projects I'm

39:39 working on and what my understanding is and what I find

39:41 Interesting. I want to maintain that and make it as current and as deep as

39:45 possible

39:45 So whenever I'm talking to an agent, it has the best representation at all

39:49 times of who I am in that moment when I ask the agent for something

39:53 Yeah, that's the goal, right

39:55 That's the question that we all should be asking of ourselves, which is yes

40:00 Does the agent have the most up-to-date information on the projects on my

40:04 preferences and my dreams and my hopes and my goals

40:10 Because you're it's only as good as the up-to-date version of that. Correct.

40:15 Yes

40:16 100% the quality the the quality of information that the agent has entirely

40:21 determines what it can do for you

40:23 Right if it doesn't know a lot about you, it's not going to be able to do a lot

40:27 for you

40:27 But if it knows a lot then it can it can it can do things for you that I

40:32 Think like even some of your in some it's kind of weird to say but I mean like

40:36 that you don't even know about yourself in ways

40:40 I mean it makes sense right because

40:42 Ultimately what this is doing

40:44 Like to distill it to its core. It's connecting the dots. Yes city and obsidian

40:50 and Claude code here are connecting the dots

40:53 now

40:54 It's actually quite difficult as a as a business owner or just as a personal in

40:59 our personal lives to connect the dots

41:01 Like why do people in a lot of ways go to?

41:03 You know coaches therapists. Yes

41:06 You know if you go to a therapist and

41:10 You have someone

41:12 Who is you're doing most of the talking, right? Think about it

41:16 You're doing a lot of the reflection and the therapist and coach is sort of

41:19 guiding you

41:20 That that's what this is doing in a lot of ways and I'm not saying by the way

41:24 that don't go to your therapist, you know just

41:27 But I'm but my point is it helps it helps you helps you uncover

41:33 What are their dots and how you can connect them?

41:37 Yeah, absolutely and for me. Yeah, it's just it's just really exciting

41:41 And yeah, it's just it's just a crazy time it with computers

41:46 So let's look at this this thing finished. So this is an idea generation report

41:49 vault relationship exploration. This is this is pretty

41:53 extensive right like

41:56 to get an idea report

41:58 Like I think this is this is really gonna show how we can move you move from

42:04 reflection to something actionable

42:07 so

42:09 Structural highlights. So again, this is just obsidian stuff, right orphans

42:14 worth noting

42:15 There's some defense technology stuff here. Just a theme. That's a growing in

42:19 Canada

42:20 Massive intellectual investment sitting in isolation also orphaned agentic

42:24 software. So orphan just means these are files

42:26 I haven't really linked to random notes. I just wrote once or something

42:29 Unresolved links that reveal latent interests

42:34 Hidden relationships again all reflection stuff fine. What's working obsidian

42:39 cloud code is a combined system is working for me

42:42 This is producing genuine breakthroughs and thinking and output

42:45 Day per domain structure when enforced. This is basically I started

42:49 Splitting my schedule where each day has a specific focus. This is cool

42:54 And this is very true the Greg Eisenberg episode as a forcing function

42:58 It's compressing months of thinking about obsidian and agents into a clear

43:02 thesis with demos very true

43:04 You know coming on the show and doing this forced me to

43:07 Synthesize everything I knew and presented

43:11 But here's where we're going to the actionable stuff tools to build

43:16 The slash graduate slash command daily note idea extractor based on daily notes

43:23 are full of idea

43:24 Sorry daily notes are full of idea tags and interesting thinking that never

43:27 gets developed the vault has nine idea tags

43:30 But hundreds of undiscovered insights build a command that scans recent daily

43:35 notes identifies ideas tagged or not and prompts you to decide

43:39 Create a standalone note add to an existing file or dismiss

43:43 This turns a daily note stream into a structured idea pipeline

43:47 Obsidian vault for new it says I just have to manage and set up a central obs

43:51 idian vault for new

43:52 what that means is in the same way that

43:55 I'm creating this of all vaults and it has all my ideas and my patterns and

43:59 everything like that

44:00 Why would I not create one for my team?

44:02 Where they like as a team we can go and ask this well questions and we can all

44:07 contribute to it

44:09 Here we go tools to start using

44:12 What is this types for all external documents?

44:15 Interesting a time blocking act that enforces date a time blocking app that en

44:21 forces day per domain meaning since I'm trying to focus on one thing each

44:25 Day one aspect of my life. It's saying why not create a time blocking app that

44:30 forces you to do that interesting

44:32 systems to implement

44:35 One sentence an obsidian agent handles arrests. This is literally the demo

44:39 three version of the Greg Eisen per

44:41 Eisenberg prep. You're already imagining it. The next step is to actually

44:45 building it start small right schedule a call with person about topic

44:49 This week in a daily note and have Otis or clogbot or open claw pick it up and

44:53 handle it

44:54 So it's saying maybe you can delegate right from the note itself is how I'm

44:58 interpreting that super interesting

45:00 Yeah, I just in line in line delegation like maybe that's even like a you knew

45:06 ux pattern

45:07 I don't even know right that you could build into like these different tools

45:09 Subjects to investigate

45:14 Christopher Alexander's pattern language applied to digital spaces interesting

45:18 Black Mountain College as a model for the stadium stadium is a physical space.

45:23 We have in Toronto

45:24 Authorless media as a concept

45:26 How Shenzhen's hardware ecosystem actually works things to write and publish

45:32 that be useful

45:33 Context architecture essay the computer as a place software book will become

45:38 fashion

45:39 What Toronto theory actually is?

45:42 editorial thinking zine

45:45 Conversations to have this isn't you these are real people Aaron stadium

45:49 workshop hosts about becoming an anchor of technical programming

45:52 This is a space. We haven't Toronto to ruin a can another person program about

45:57 making program the flagship series

45:59 Steph and go obsidian CEO about the vault as a place

46:01 And so this is like

46:05 Yeah, this is crazy. It's suggesting people I should meet right top five

46:10 High impact do now build the graduate command or do a manual weekly idea review

46:16 ?

46:16 This is crazy, dude. This is actually crazy

46:21 like

46:23 And the fact that it's in plain text and it just not be it's not there's no

46:28 images

46:29 It doesn't make it easy to read but I kind of like it because it's it's like

46:34 dressed down

46:35 You know what I mean? Yeah, I mean I like that aesthetic because I'm a nerd

46:38 But you could just say you know, you could you could just say obviously you

46:41 could just say can you turn this?

46:43 Into a beautiful readable HTML file that is on my desktop. This is hard to read

46:51 And it's gonna do that. Mm-hmm, right? So, I mean if you don't that says if you

46:54 don't like that

46:55 Just do it. However, you want to do it. You know what I mean? I like it like

46:58 this in this kind of like this aesthetic

47:01 But yeah, that's how you can move out of reflection and of course, you know, of

47:06 course

47:06 We can also say like here, you know, if we don't if we don't want to do this

47:10 We could also say like it recommended that slash graduate command. So I could

47:15 just say build the slash graduate command

47:20 Right, which is interesting and is that how you started building a lot of your

47:24 commands?

47:25 Like it it sort of it started suggesting and you're like just go build it

47:29 Well, I started off actually like building them myself like I'm like oh trying

47:34 to think about commands myself

47:35 But then yeah, I said I started asking the agent like well wait a second

47:39 What commands do you think would be interesting and and just and this might be

47:43 useful another thing that I'd like to do is I like to

47:46 Move to like higher levels of abstraction when I'm using an LL LLM

47:51 What I mean by that is I could say like oh make a command that

47:55 Tells me what I should focus on each day and that's that's like a command that

48:02 I thought of

48:03 But when I when I another thing you can do is you can step back and I can say

48:07 Based on my obsidian vault and what you know about me

48:12 Form an understanding of what you think my level of understand like

48:16 Where you think my skill level is in terms of a person and the projects that I

48:21 'm working on and based on that

48:23 Suggest the kinds of commands I should use that would take me from the level of

48:28 mat to a higher level

48:29 Right get like you know get it to suggest the commands for me instead of me

48:35 suggesting the commands and I could pick between them

48:37 So look at this. This is this is it was this was the agent's idea based on what

48:41 it read in my vault

48:42 Based on the notes. I'm taking right so let's see what this is daily note idea

48:46 extractor

48:46 Ideas insights and original thinking accumulate in detail in daily notes

48:51 But rarely graduate into standalone notes working where they can compound

48:55 through backlinks

48:56 This command scans recent daily notes surfaces the best candidates and helps

49:00 decide what to promote into like an idea or something, right?

49:03 Sure

49:05 So this is how it's going to work

49:06 It scans all the recent daily notes. It cross references with the existing

49:10 vault. It presents candidates

49:12 It graduates selected ideas if if creating a new standalone note create the

49:18 note in the vault route

49:19 Write the note as a mini essay or working document that captures the core claim

49:24 or question context from the daily note where it originated

49:27 Connections to other vaults notes as backlinks now

49:31 All this stuff like it captures the core claim or question

49:33 You might be looking at this and think like okay. This is just the text that

49:37 the agent generated and it is

49:39 But also it hits differently for me because I know

49:43 Like I'm writing a lot about these things. I know I know like even like the

49:48 mini essay thing

49:49 These are words that mean specific things to me

49:52 Which is just so it's just so crazy. It's very contextual

49:55 I know what it's talking about because I spend a lot of time in in this tool

49:59 and I spend a lot of time writing

50:00 So yeah, I created and it's going to create that command

50:03 Right and it's like

50:07 That's that's nuts because I'm just going about making notes

50:10 And I have this parallel agent that is looking at my notes

50:13 And giving me ideas on how I can prove my workflow improve my life

50:17 And then and then not only can it just suggest it. It could just build the

50:20 thing

50:20 And it's done

50:22 And then we have it right here slash graduate. I could just hit it

50:25 And it's going to run

50:27 That's crazy

50:29 If if I'm open AI or anthropic, I'm buying obsidian

50:34 Right because it's the missing link

50:40 Yeah, it's nuts. It's the missing link what the fact that there are people

50:43 Like you've sold me on this by the way

50:46 I've downloaded obsidian. I think it's a free tool, right? Yes, it is. It's

50:51 open sourced

50:52 I've downloaded it, but I have not

50:57 created my vault because I wanted

50:59 You've been

51:01 I knew that this was going to be great

51:04 I knew that I would go through this

51:06 This actually exceeded my expectations like this is

51:09 the fact

51:12 like

51:13 It makes no sense

51:15 It makes no sense

51:16 the the fact like that if you are if you are using if you are serious about

51:23 Using LLMs to take your ideas and put in and get the most out of them

51:28 If you're serious about building, you know, what people are calling a personal

51:32 OS

51:32 And you are not using a centralized note-taking tool like this

51:38 Yeah, that uses markdown as the foundation

51:42 Then you are not using LLMs properly. Yeah, or at least not at the limit

51:49 Yeah, yeah, exactly. You're not getting the most out of it. Yeah

51:53 You're not getting the most out of it. So I think what's what's difficult about

51:56 this is that it requires

51:58 it does require a lot of time

52:00 and to actually set it up properly it requires

52:04 like

52:07 Yeah, I mean it takes a lot of time and and and the UI is in and is is so

52:12 daunting in the sense that it's a blank canvas

52:15 And it's not like hey, you should like write your preferences over here or you

52:19 know, you kind of just have to

52:20 you have to

52:24 Come up with these ideas yourselves. Yes

52:26 but

52:29 That's still so amazing, right because I mean even when we work with other

52:33 humans

52:33 We have to find a way to explain things to them and I just think it's so cool

52:37 that

52:38 Now we can work with these agents and we still have to explain things to them

52:42 But we only need to explain them once because once we get it down on into a

52:46 file

52:46 We can always reference that file that explanation of a project or a preference

52:50 or anything and it's always there and you can pass it in

52:53 Yeah, a file is like

52:55 Essentially perfect a perfect memory. Yes, right?

52:58 Human beings have memories like we recall things

53:01 Yes, but there's tons of studies that show that what we remember in fact is

53:06 completely

53:07 different than a reality for example

53:09 When we went and got our that haircut in Mississauga

53:13 I could have thought that I had the best haircut

53:18 You know, that's what my memory remembers. There was a great haircut, but who

53:22 knows it could have been the worst haircut I had ever gone

53:25 Now

53:29 Obsidian or whatever tool you end up using

53:32 Like you know if I had written

53:35 Yes, like the memory the file the markdown file is perfect

53:40 So that when I link that or I recall it it is going to give me a perfect

53:46 uh

53:47 Data data point and the other thing about you know, these files is that you

53:51 hope?

53:52 Well, they're not biased basically. They're as biased as the human being is

53:56 In terms of writing the reflections. Yes at that moment in time. Yes

54:01 It's crazy man and

54:05 Yeah, it's just crazy and there's all there's all of these different aspects to

54:10 it

54:10 There's the privacy of it and what that means

54:14 There's the power of it the fact that now you can just work with these

54:17 computers in natural language and just delegate to them

54:20 Um, there's a fact that there's people like me that are that are using these

54:24 tools and trying to figure out how to delegate stuff to agents in this way

54:27 There's people that are like me that are even more hardcore in different ways

54:31 and and pushing them

54:32 And I just think it's such a crazy time to be alive because I think we are

54:37 potentially

54:38 Watching a fundamental shift in the human relationship to computers

54:44 And it's just I'm just really happy to be alive while this is happening and I'm

54:47 curious like how is this all going to unravel?

54:50 Well, what's what's cool about this is?

54:52 99.99 of people are not going to spend the time

54:59 to actually set up something like this and make it a part of their daily lives

55:05 and

55:05 the

55:07 Alpha so to speak is in terms of leading a more productive

55:12 happier healthier

55:14 better

55:16 More money-making career is in in in using something like this with an llm. I

55:22 think so

55:22 I'm not saying download obsidian today and I have no affiliation or whatever

55:27 with them, but i'm saying like

55:29 Pick a it sounds like what we should all be doing and i'm talking i'm giving

55:34 myself this this advice is like

55:36 There's no excuse anymore for me not to be writing down and reflecting

55:42 Yes into markdown files. Yes in a world that where llms use markdown files is

55:48 the oxygen. Yes

55:50 Like people think tokens are the oxygen. Yes, but they're not

55:55 Yes, the markdown files are the memories like think about what a human being is

56:00 Yes, you know

56:03 Is a human being

56:06 The the energy of a human being or is it the memories of it, you know what we

56:11 were called, you know

56:12 I mean, that's like a philosophical question and maybe it's a bit a bit a bit

56:17 of both, but it's

56:19 you know, I think that

56:22 there's something really really fascinating about md files as

56:27 an underrated about them in order to

56:30 to have a true

56:34 Computer experience and in today's day and age. Yeah, there's definitely

56:38 something going on here

56:39 Some fundamental shift. Yeah, it's awesome

56:42 Yeah, and like I have bad my work, you know, i'm learning in real time, right?

56:46 Like and I I don't have the right vocabulary to even explain this

56:49 Yes

56:50 And now they're too high man. I don't do why i'm trying to i'm trying to figure

56:54 it out in real time

56:55 That's that's why I think like I know I show something and and for me

56:58 I'll do something or I'll see something and my friends are like they kind of

57:02 laugh because i'll just be sitting at my computer

57:04 Just tripping out and I think it's because I really like computers and and I

57:09 cannot believe

57:10 That this is possible. I cannot believe

57:13 That I can just be making notes on my computer like I have been since I was a

57:17 kid

57:18 And then all of a sudden this agent can scan through it and build things

57:23 because of it

57:24 And and like connect patterns that I could never see it's nuts man

57:28 it's nuts and

57:30 And at the root of it, you're right. It's just a collection of interrelated

57:34 markdown files

57:35 Yeah

57:36 Cool man. I appreciate you. I don't know if you can see my mind, but my mind is

57:40 blown right now

57:41 Thank god. Yeah

57:44 I wanted to do right by you. I also just like I say this every time man

57:48 But i'm just gonna keep always saying it to you all the time

57:51 I really really really really appreciate everything that you do

57:54 I think your pattern recognition and your pattern matching is like

57:58 Like really underrated. I think there's a lot of things that you do that I don

58:02 't think it's like difficult to see if you're not really paying attention

58:05 I just want to say like thank you for for everything that you do

58:08 You're always putting on like new voices on your show. I see it. I really

58:12 appreciate it

58:13 And it's just been it's just an honor to know you and yeah, just thanks for the

58:16 opportunity. Thanks for everything man

58:18 I appreciate you vin your legend. I'll include links

58:21 for where to follow

58:24 criminally under followed internet vin

58:26 on on x on his youtube show

58:30 Podcast in the show notes and description. You can go and check him out there

58:36 Uh, people please play with some of these tools

58:41 Uh, and let me know what you think let vin know what you think please and uh

58:48 then

58:49 I will beg you to come back on the show

58:51 And I hope you come back on again for sure man. Thank you. Thank you

查看原文 →

TaiYuan TaiYuan · 6个月

好东西~

moon moon · 6个月

超级好用

Download Pickful App

Better experience on mobile

iOS Android APK

iOS

Android

APK